diff --git "a/run-2024-07-20T09:26:09+00:00.log" "b/run-2024-07-20T09:26:09+00:00.log"
--- "a/run-2024-07-20T09:26:09+00:00.log"
+++ "b/run-2024-07-20T09:26:09+00:00.log"
@@ -3529,4 +3529,1161 @@ Non-default generation parameters: {'max_length': 200, 'early_stopping': True, '
 
 94%|█████████▍| 348320/371472 [5:33:24<2:02:45,  3.14it/s]
 94%|█████████▍| 348321/371472 [5:33:25<2:01:23,  3.18it/s]
 94%|█████████▍| 348322/371472 [5:33:25<1:53:21,  3.40it/s]
 94%|█████████▍| 348323/371472 [5:33:25<1:54:44,  3.36it/s]
 94%|█████████▍| 348324/371472 [5:33:25<1:51:47,  3.45it/s]
 94%|█████████▍| 348325/371472 [5:33:26<1:50:00,  3.51it/s]
 94%|█████████▍| 348326/371472 [5:33:26<1:49:29,  3.52it/s]
 94%|█████████▍| 348327/371472 [5:33:26<1:58:55,  3.24it/s]
 94%|█████████▍| 348328/371472 [5:33:27<2:00:12,  3.21it/s]
 94%|█████████▍| 348329/371472 [5:33:27<1:57:54,  3.27it/s]
 94%|█████████▍| 348330/371472 [5:33:27<1:52:35,  3.43it/s]
 94%|█████████▍| 348331/371472 [5:33:28<2:00:36,  3.20it/s]
 94%|█████████▍| 348332/371472 [5:33:28<1:57:33,  3.28it/s]
 94%|█████████▍| 348333/371472 [5:33:28<1:54:47,  3.36it/s]
 94%|█████████▍| 348334/371472 [5:33:28<1:51:11,  3.47it/s]
 94%|█████████▍| 348335/371472 [5:33:29<1:50:11,  3.50it/s]
 94%|█████████▍| 348336/371472 [5:33:29<1:47:10,  3.60it/s]
 94%|█████████▍| 348337/371472 [5:33:29<1:48:30,  3.55it/s]
 94%|█████████▍| 348338/371472 [5:33:30<1:45:56,  3.64it/s]
 94%|█████████▍| 348339/371472 [5:33:30<1:50:06,  3.50it/s]
 94%|█████████▍| 348340/371472 [5:33:30<1:57:33,  3.28it/s]
                                                           
{'loss': 2.5126, 'learning_rate': 1.5607425283888895e-07, 'epoch': 15.0}
 
 94%|█████████▍| 348340/371472 [5:33:30<1:57:33,  3.28it/s]
 94%|█████████▍| 348341/371472 [5:33:30<1:55:26,  3.34it/s]
 94%|█████████▍| 348342/371472 [5:33:31<2:04:46,  3.09it/s]
 94%|█████████▍| 348343/371472 [5:33:31<2:04:33,  3.09it/s]
 94%|█████████▍| 348344/371472 [5:33:31<2:05:14,  3.08it/s]
 94%|█████████▍| 348345/371472 [5:33:32<2:03:05,  3.13it/s]
 94%|█████████▍| 348346/371472 [5:33:32<2:01:19,  3.18it/s]
 94%|█████████▍| 348347/371472 [5:33:32<2:03:02,  3.13it/s]
 94%|█████████▍| 348348/371472 [5:33:33<1:59:27,  3.23it/s]
 94%|█████████▍| 348349/371472 [5:33:33<1:57:55,  3.27it/s]
 94%|█████████▍| 348350/371472 [5:33:33<1:53:34,  3.39it/s]
 94%|█████████▍| 348351/371472 [5:33:34<1:50:57,  3.47it/s]
 94%|█████████▍| 348352/371472 [5:33:34<1:56:07,  3.32it/s]
 94%|█████████▍| 348353/371472 [5:33:34<1:51:54,  3.44it/s]
 94%|█████████▍| 348354/371472 [5:33:34<1:48:50,  3.54it/s]
 94%|█████████▍| 348355/371472 [5:33:35<1:48:52,  3.54it/s]
 94%|█████████▍| 348356/371472 [5:33:35<2:04:43,  3.09it/s]
 94%|█████████▍| 348357/371472 [5:33:35<2:01:52,  3.16it/s]
 94%|█████████▍| 348358/371472 [5:33:36<1:59:10,  3.23it/s]
 94%|█████████▍| 348359/371472 [5:33:36<1:56:01,  3.32it/s]
 94%|█████████▍| 348360/371472 [5:33:36<1:59:23,  3.23it/s]
                                                           
{'loss': 2.5462, 'learning_rate': 1.5602577086341013e-07, 'epoch': 15.0}
 
 94%|█████████▍| 348360/371472 [5:33:36<1:59:23,  3.23it/s]
 94%|█████████▍| 348361/371472 [5:33:37<1:56:52,  3.30it/s]
 94%|█████████▍| 348362/371472 [5:33:37<1:51:45,  3.45it/s]
 94%|█████████▍| 348363/371472 [5:33:37<1:59:26,  3.22it/s]
 94%|█████████▍| 348364/371472 [5:33:38<1:58:19,  3.25it/s]
 94%|█████████▍| 348365/371472 [5:33:38<2:03:09,  3.13it/s]
 94%|█████████▍| 348366/371472 [5:33:38<1:59:03,  3.23it/s]
 94%|█████████▍| 348367/371472 [5:33:38<1:54:28,  3.36it/s]
 94%|█████████▍| 348368/371472 [5:33:39<1:54:11,  3.37it/s]
 94%|█████████▍| 348369/371472 [5:33:39<2:03:14,  3.12it/s]
 94%|█████████▍| 348370/371472 [5:33:39<1:58:02,  3.26it/s]
 94%|█████████▍| 348371/371472 [5:33:40<2:00:51,  3.19it/s]
 94%|█████████▍| 348372/371472 [5:33:40<2:04:51,  3.08it/s]
 94%|█████████▍| 348373/371472 [5:33:40<1:59:27,  3.22it/s]
 94%|█████████▍| 348374/371472 [5:33:41<1:55:19,  3.34it/s]
 94%|█████████▍| 348375/371472 [5:33:41<2:06:24,  3.05it/s]
 94%|█████████▍| 348376/371472 [5:33:41<2:09:33,  2.97it/s]
 94%|█████��███▍| 348377/371472 [5:33:42<2:01:21,  3.17it/s]
 94%|█████████▍| 348378/371472 [5:33:42<2:01:40,  3.16it/s]
 94%|█████████▍| 348379/371472 [5:33:42<1:55:45,  3.32it/s]
 94%|█████████▍| 348380/371472 [5:33:43<1:59:22,  3.22it/s]
                                                           
{'loss': 2.632, 'learning_rate': 1.5597728888793117e-07, 'epoch': 15.01}
-
 94%|█████████▍| 348380/371472 [5:33:43<1:59:22,  3.22it/s]
 94%|█████████▍| 348381/371472 [5:33:43<2:09:33,  2.97it/s]
 94%|█████████▍| 348382/371472 [5:33:43<2:01:51,  3.16it/s]
 94%|█████████▍| 348383/371472 [5:33:43<1:55:53,  3.32it/s]
\ No newline at end of file
+
 94%|█████████▍| 348380/371472 [5:33:43<1:59:22,  3.22it/s]
 94%|█████████▍| 348381/371472 [5:33:43<2:09:33,  2.97it/s]
 94%|█████████▍| 348382/371472 [5:33:43<2:01:51,  3.16it/s]
 94%|█████████▍| 348383/371472 [5:33:43<1:55:53,  3.32it/s]
 94%|█████████▍| 348384/371472 [5:33:44<2:21:11,  2.73it/s]
 94%|█████████▍| 348385/371472 [5:33:44<2:12:48,  2.90it/s]
 94%|█████████▍| 348386/371472 [5:33:45<2:06:40,  3.04it/s]
 94%|█████████▍| 348387/371472 [5:33:45<1:58:22,  3.25it/s]
 94%|█████████▍| 348388/371472 [5:33:45<1:57:59,  3.26it/s]
 94%|█████████▍| 348389/371472 [5:33:45<1:53:28,  3.39it/s]
 94%|█████████▍| 348390/371472 [5:33:46<1:51:10,  3.46it/s]
 94%|█████████▍| 348391/371472 [5:33:46<1:46:54,  3.60it/s]
 94%|█████████▍| 348392/371472 [5:33:46<1:46:55,  3.60it/s]
 94%|█████████▍| 348393/371472 [5:33:47<1:53:35,  3.39it/s]
 94%|█████████▍| 348394/371472 [5:33:47<1:50:42,  3.47it/s]
 94%|█████████▍| 348395/371472 [5:33:47<1:50:42,  3.47it/s]
 94%|█████████▍| 348396/371472 [5:33:47<1:48:10,  3.56it/s]
 94%|█████████▍| 348397/371472 [5:33:48<1:49:57,  3.50it/s]
 94%|█████████▍| 348398/371472 [5:33:48<1:46:26,  3.61it/s]
 94%|█████████▍| 348399/371472 [5:33:48<1:52:18,  3.42it/s]
 94%|█████████▍| 348400/371472 [5:33:49<1:54:29,  3.36it/s]
                                                           
{'loss': 2.6343, 'learning_rate': 1.5592880691245232e-07, 'epoch': 15.01}
+
 94%|█████████▍| 348400/371472 [5:33:49<1:54:29,  3.36it/s]
 94%|█████████▍| 348401/371472 [5:33:49<1:56:48,  3.29it/s]
 94%|█████████▍| 348402/371472 [5:33:49<1:55:25,  3.33it/s]
 94%|█████████▍| 348403/371472 [5:33:50<2:05:05,  3.07it/s]
 94%|█████████▍| 348404/371472 [5:33:50<1:57:40,  3.27it/s]
 94%|█████████▍| 348405/371472 [5:33:50<1:50:24,  3.48it/s]
 94%|█████████▍| 348406/371472 [5:33:50<1:48:29,  3.54it/s]
 94%|█████████▍| 348407/371472 [5:33:51<1:53:25,  3.39it/s]
 94%|█████████▍| 348408/371472 [5:33:51<1:54:08,  3.37it/s]
 94%|█████████▍| 348409/371472 [5:33:51<1:49:41,  3.50it/s]
 94%|█████████▍| 348410/371472 [5:33:52<1:48:55,  3.53it/s]
 94%|█████████▍| 348411/371472 [5:33:52<1:49:25,  3.51it/s]
 94%|█████████▍| 348412/371472 [5:33:52<1:53:27,  3.39it/s]
 94%|█████████▍| 348413/371472 [5:33:52<2:00:11,  3.20it/s]
 94%|█████████▍| 348414/371472 [5:33:53<1:55:47,  3.32it/s]
 94%|█████████▍| 348415/371472 [5:33:53<1:58:09,  3.25it/s]
 94%|█████████▍| 348416/371472 [5:33:53<1:57:38,  3.27it/s]
 94%|█████████▍| 348417/371472 [5:33:54<1:53:54,  3.37it/s]
 94%|█████████▍| 348418/371472 [5:33:54<1:53:55,  3.37it/s]
 94%|█████████▍| 348419/371472 [5:33:54<1:50:50,  3.47it/s]
 94%|█████████▍| 348420/371472 [5:33:54<1:51:15,  3.45it/s]
                                                           
{'loss': 2.5431, 'learning_rate': 1.558803249369734e-07, 'epoch': 15.01}
+
 94%|█████████▍| 348420/371472 [5:33:54<1:51:15,  3.45it/s]
 94%|█████████▍| 348421/371472 [5:33:55<1:52:59,  3.40it/s]
 94%|█████████▍| 348422/371472 [5:33:55<1:55:26,  3.33it/s]
 94%|█████████▍| 348423/371472 [5:33:55<1:51:12,  3.45it/s]
 94%|█████████▍| 348424/371472 [5:33:56<1:48:17,  3.55it/s]
 94%|█████████▍| 348425/371472 [5:33:56<1:46:37,  3.60it/s]
 94%|█████████▍| 348426/371472 [5:33:56<1:49:16,  3.51it/s]
 94%|█████████▍| 348427/371472 [5:33:56<1:49:35,  3.50it/s]
 94%|█████████▍| 348428/371472 [5:33:57<1:49:42,  3.50it/s]
 94%|█████████▍| 348429/371472 [5:33:57<1:50:46,  3.47it/s]
 94%|█████████▍| 348430/371472 [5:33:57<1:47:10,  3.58it/s]
 94%|█████████▍| 348431/371472 [5:33:58<1:51:31,  3.44it/s]
 94%|█████████▍| 348432/371472 [5:33:58<1:52:28,  3.41it/s]
 94%|█████████▍| 348433/371472 [5:33:58<1:51:41,  3.44it/s]
 94%|█████████▍| 348434/371472 [5:33:59<1:50:34,  3.47it/s]
 94%|█████████▍| 348435/371472 [5:33:59<1:49:15,  3.51it/s]
 94%|█████████▍| 348436/371472 [5:33:59<1:44:35,  3.67it/s]
 94%|█████████▍| 348437/371472 [5:33:59<1:51:04,  3.46it/s]
 94%|█████████▍| 348438/371472 [5:34:00<1:56:07,  3.31it/s]
 94%|█████████▍| 348439/371472 [5:34:00<1:53:12,  3.39it/s]
 94%|█████████▍| 348440/371472 [5:34:00<1:49:15,  3.51it/s]
                                                           
{'loss': 2.5396, 'learning_rate': 1.5583184296149455e-07, 'epoch': 15.01}
+
 94%|█████████▍| 348440/371472 [5:34:00<1:49:15,  3.51it/s]
 94%|█████████▍| 348441/371472 [5:34:01<1:48:24,  3.54it/s]
 94%|█████████▍| 348442/371472 [5:34:01<1:49:36,  3.50it/s]
 94%|█████████▍| 348443/371472 [5:34:01<1:43:56,  3.69it/s]
 94%|█████████▍| 348444/371472 [5:34:01<1:46:00,  3.62it/s]
 94%|█████████▍| 348445/371472 [5:34:02<1:45:41,  3.63it/s]
 94%|█████████▍| 348446/371472 [5:34:02<1:53:02,  3.39it/s]
 94%|█████████▍| 348447/371472 [5:34:02<1:50:41,  3.47it/s]
 94%|█████████▍| 348448/371472 [5:34:02<1:47:10,  3.58it/s]
 94%|█████████▍| 348449/371472 [5:34:03<1:48:22,  3.54it/s]
 94%|█████████▍| 348450/371472 [5:34:03<1:46:25,  3.61it/s]
 94%|█████████▍| 348451/371472 [5:34:03<1:47:17,  3.58it/s]
 94%|█████████▍| 348452/371472 [5:34:04<1:42:37,  3.74it/s]
 94%|█████████▍| 348453/371472 [5:34:04<1:43:50,  3.69it/s]
 94%|█████████▍| 348454/371472 [5:34:04<1:44:40,  3.66it/s]
 94%|█████████▍| 348455/371472 [5:34:04<1:52:02,  3.42it/s]
 94%|█████████▍| 348456/371472 [5:34:05<1:49:12,  3.51it/s]
 94%|█████████▍| 348457/371472 [5:34:05<1:45:50,  3.62it/s]
 94%|█████████▍| 348458/371472 [5:34:05<1:44:48,  3.66it/s]
 94%|█████████▍| 348459/371472 [5:34:06<1:46:38,  3.60it/s]
 94%|█████████▍| 348460/371472 [5:34:06<1:46:31,  3.60it/s]
                                                           
{'loss': 2.7789, 'learning_rate': 1.5578336098601562e-07, 'epoch': 15.01}
+
 94%|█████████▍| 348460/371472 [5:34:06<1:46:31,  3.60it/s]
 94%|█████████▍| 348461/371472 [5:34:06<1:51:24,  3.44it/s]
 94%|█████████▍| 348462/371472 [5:34:06<1:46:17,  3.61it/s]
 94%|█████████▍| 348463/371472 [5:34:07<1:47:51,  3.56it/s]
 94%|█████████▍| 348464/371472 [5:34:07<1:46:46,  3.59it/s]
 94%|█████████▍| 348465/371472 [5:34:07<1:47:20,  3.57it/s]
 94%|█████████▍| 348466/371472 [5:34:08<1:47:09,  3.58it/s]
 94%|█████████▍| 348467/371472 [5:34:08<1:52:30,  3.41it/s]
 94%|█████████▍| 348468/371472 [5:34:08<1:53:34,  3.38it/s]
 94%|█████████▍| 348469/371472 [5:34:08<1:54:02,  3.36it/s]
 94%|█████████▍| 348470/371472 [5:34:09<2:02:13,  3.14it/s]
 94%|█████████▍| 348471/371472 [5:34:09<2:05:24,  3.06it/s]
 94%|█████████▍| 348472/371472 [5:34:10<2:13:06,  2.88it/s]
 94%|█████████▍| 348473/371472 [5:34:10<2:02:58,  3.12it/s]
 94%|█████████▍| 348474/371472 [5:34:10<1:59:12,  3.22it/s]
 94%|█████████▍| 348475/371472 [5:34:10<1:55:32,  3.32it/s]
 94%|█████████▍| 348476/371472 [5:34:11<1:54:30,  3.35it/s]
 94%|█████████▍| 348477/371472 [5:34:11<1:56:43,  3.28it/s]
 94%|█████████▍| 348478/371472 [5:34:11<1:53:58,  3.36it/s]
 94%|█████████▍| 348479/371472 [5:34:12<1:53:02,  3.39it/s]
 94%|█████████▍| 348480/371472 [5:34:12<1:55:58,  3.30it/s]
                                                           
{'loss': 2.8674, 'learning_rate': 1.5573487901053677e-07, 'epoch': 15.01}
+
 94%|█████████▍| 348480/371472 [5:34:12<1:55:58,  3.30it/s]
 94%|█████████▍| 348481/371472 [5:34:12<1:55:15,  3.32it/s]
 94%|█████████▍| 348482/371472 [5:34:13<2:00:11,  3.19it/s]
 94%|█████████▍| 348483/371472 [5:34:13<1:56:49,  3.28it/s]
 94%|█████████▍| 348484/371472 [5:34:13<1:55:32,  3.32it/s]
 94%|█████████▍| 348485/371472 [5:34:13<1:51:37,  3.43it/s]
 94%|█████████▍| 348486/371472 [5:34:14<1:49:19,  3.50it/s]
 94%|█████████▍| 348487/371472 [5:34:14<2:04:12,  3.08it/s]
 94%|█████████▍| 348488/371472 [5:34:14<2:02:51,  3.12it/s]
 94%|█████████▍| 348489/371472 [5:34:15<1:57:44,  3.25it/s]
 94%|█████████▍| 348490/371472 [5:34:15<1:52:53,  3.39it/s]
 94%|█████████▍| 348491/371472 [5:34:15<1:52:56,  3.39it/s]
 94%|█████████▍| 348492/371472 [5:34:15<1:51:40,  3.43it/s]
 94%|█████████▍| 348493/371472 [5:34:16<1:53:05,  3.39it/s]
 94%|█████████▍| 348494/371472 [5:34:16<1:56:41,  3.28it/s]
 94%|█████████▍| 348495/371472 [5:34:16<1:56:28,  3.29it/s]
 94%|█████████▍| 348496/371472 [5:34:17<1:56:06,  3.30it/s]
 94%|█████████▍| 348497/371472 [5:34:17<1:53:28,  3.37it/s]
 94%|█████████▍| 348498/371472 [5:34:17<1:54:57,  3.33it/s]
 94%|█████████▍| 348499/371472 [5:34:18<1:57:03,  3.27it/s]
 94%|█████████▍| 348500/371472 [5:34:18<1:58:45,  3.22it/s]
                                                           
{'loss': 2.5768, 'learning_rate': 1.5568639703505784e-07, 'epoch': 15.01}
+
 94%|█████████▍| 348500/371472 [5:34:18<1:58:45,  3.22it/s]
 94%|█████████▍| 348501/371472 [5:34:18<1:55:01,  3.33it/s]
 94%|█████████▍| 348502/371472 [5:34:18<1:48:29,  3.53it/s]
 94%|█████████▍| 348503/371472 [5:34:19<1:51:34,  3.43it/s]
 94%|█████████▍| 348504/371472 [5:34:19<1:49:39,  3.49it/s]
 94%|█████████▍| 348505/371472 [5:34:19<1:54:08,  3.35it/s]
 94%|█████████▍| 348506/371472 [5:34:20<1:56:48,  3.28it/s]
 94%|█████████▍| 348507/371472 [5:34:20<1:55:54,  3.30it/s]
 94%|█████████▍| 348508/371472 [5:34:20<1:52:42,  3.40it/s]
 94%|█████████▍| 348509/371472 [5:34:21<1:47:09,  3.57it/s]
 94%|█████████▍| 348510/371472 [5:34:21<1:46:35,  3.59it/s]
 94%|█████████▍| 348511/371472 [5:34:21<1:47:56,  3.55it/s]
 94%|█████████▍| 348512/371472 [5:34:22<2:12:11,  2.89it/s]
 94%|█████████▍| 348513/371472 [5:34:22<2:08:27,  2.98it/s]
 94%|█████████▍| 348514/371472 [5:34:22<2:05:26,  3.05it/s]
 94%|█████████▍| 348515/371472 [5:34:22<2:01:05,  3.16it/s]
 94%|█████████▍| 348516/371472 [5:34:23<1:57:50,  3.25it/s]
 94%|█████████▍| 348517/371472 [5:34:23<1:55:41,  3.31it/s]
 94%|█████████▍| 348518/371472 [5:34:23<1:52:34,  3.40it/s]
 94%|█████████▍| 348519/371472 [5:34:24<1:52:50,  3.39it/s]
 94%|█████████▍| 348520/371472 [5:34:24<1:55:23,  3.32it/s]
                                                           
{'loss': 2.5725, 'learning_rate': 1.55637915059579e-07, 'epoch': 15.01}
+
 94%|█████████▍| 348520/371472 [5:34:24<1:55:23,  3.32it/s]
 94%|█████████▍| 348521/371472 [5:34:24<1:55:40,  3.31it/s]
 94%|█████████▍| 348522/371472 [5:34:25<1:58:02,  3.24it/s]
 94%|█████████▍| 348523/371472 [5:34:25<1:50:26,  3.46it/s]
 94%|█████████▍| 348524/371472 [5:34:25<1:49:36,  3.49it/s]
 94%|█████████▍| 348525/371472 [5:34:25<1:49:07,  3.50it/s]
 94%|█████████▍| 348526/371472 [5:34:26<1:45:38,  3.62it/s]
 94%|█████████▍| 348527/371472 [5:34:26<1:46:24,  3.59it/s]
 94%|█████████▍| 348528/371472 [5:34:26<1:46:37,  3.59it/s]
 94%|█████████▍| 348529/371472 [5:34:26<1:47:40,  3.55it/s]
 94%|█████████▍| 348530/371472 [5:34:27<1:51:05,  3.44it/s]
 94%|█████████▍| 348531/371472 [5:34:27<1:51:47,  3.42it/s]
 94%|█████████▍| 348532/371472 [5:34:27<2:02:03,  3.13it/s]
 94%|█████████▍| 348533/371472 [5:34:28<2:02:38,  3.12it/s]
 94%|█████████▍| 348534/371472 [5:34:28<1:57:42,  3.25it/s]
 94%|█████████▍| 348535/371472 [5:34:28<1:53:38,  3.36it/s]
 94%|█████████▍| 348536/371472 [5:34:29<1:49:07,  3.50it/s]
 94%|█████████▍| 348537/371472 [5:34:29<1:46:33,  3.59it/s]
 94%|█████████▍| 348538/371472 [5:34:29<1:45:38,  3.62it/s]
 94%|█████████▍| 348539/371472 [5:34:29<1:47:56,  3.54it/s]
 94%|█████████▍| 348540/371472 [5:34:30<1:48:21,  3.53it/s]
                                                           
{'loss': 2.7716, 'learning_rate': 1.5558943308410004e-07, 'epoch': 15.01}
+
 94%|█████████▍| 348540/371472 [5:34:30<1:48:21,  3.53it/s]
 94%|█████████▍| 348541/371472 [5:34:30<1:50:34,  3.46it/s]
 94%|█████████▍| 348542/371472 [5:34:30<1:56:36,  3.28it/s]
 94%|█████████▍| 348543/371472 [5:34:31<2:00:08,  3.18it/s]
 94%|█████████▍| 348544/371472 [5:34:31<1:58:57,  3.21it/s]
 94%|█████████▍| 348545/371472 [5:34:31<1:54:56,  3.32it/s]
 94%|█████████▍| 348546/371472 [5:34:32<1:52:47,  3.39it/s]
 94%|█████████▍| 348547/371472 [5:34:32<2:01:06,  3.15it/s]
 94%|█████████▍| 348548/371472 [5:34:32<2:02:59,  3.11it/s]
 94%|█████████▍| 348549/371472 [5:34:33<2:00:42,  3.17it/s]
 94%|█████████▍| 348550/371472 [5:34:33<2:00:17,  3.18it/s]
 94%|█████████▍| 348551/371472 [5:34:33<2:04:50,  3.06it/s]
 94%|█████████▍| 348552/371472 [5:34:34<2:05:29,  3.04it/s]
 94%|█████████▍| 348553/371472 [5:34:34<2:02:29,  3.12it/s]
 94%|█████████▍| 348554/371472 [5:34:34<2:02:05,  3.13it/s]
 94%|█████████▍| 348555/371472 [5:34:35<2:09:50,  2.94it/s]
 94%|█████████▍| 348556/371472 [5:34:35<2:12:05,  2.89it/s]
 94%|█████████▍| 348557/371472 [5:34:35<2:11:25,  2.91it/s]
 94%|█████████▍| 348558/371472 [5:34:36<2:01:48,  3.14it/s]
 94%|█████████▍| 348559/371472 [5:34:36<1:56:14,  3.29it/s]
 94%|█████████▍| 348560/371472 [5:34:36<1:51:20,  3.43it/s]
                                                           
{'loss': 2.4726, 'learning_rate': 1.5554095110862119e-07, 'epoch': 15.01}
+
 94%|█████████▍| 348560/371472 [5:34:36<1:51:20,  3.43it/s]
 94%|█████████▍| 348561/371472 [5:34:36<1:50:08,  3.47it/s]
 94%|█████████▍| 348562/371472 [5:34:37<1:47:33,  3.55it/s]
 94%|█████████▍| 348563/371472 [5:34:37<1:46:00,  3.60it/s]
 94%|█████████▍| 348564/371472 [5:34:37<1:57:27,  3.25it/s]
 94%|█████████▍| 348565/371472 [5:34:38<1:56:12,  3.29it/s]
 94%|█████████▍| 348566/371472 [5:34:38<1:49:50,  3.48it/s]
 94%|█████████▍| 348567/371472 [5:34:38<1:51:12,  3.43it/s]
 94%|█████████▍| 348568/371472 [5:34:38<1:49:13,  3.50it/s]
 94%|█████████▍| 348569/371472 [5:34:39<1:48:32,  3.52it/s]
 94%|█████████▍| 348570/371472 [5:34:39<1:57:00,  3.26it/s]
 94%|█████████▍| 348571/371472 [5:34:39<1:56:02,  3.29it/s]
 94%|█████████▍| 348572/371472 [5:34:40<2:06:13,  3.02it/s]
 94%|█████████▍| 348573/371472 [5:34:40<1:58:51,  3.21it/s]
 94%|█████████▍| 348574/371472 [5:34:40<1:53:53,  3.35it/s]
 94%|█████████▍| 348575/371472 [5:34:41<2:00:20,  3.17it/s]
 94%|█████████▍| 348576/371472 [5:34:41<2:01:16,  3.15it/s]
 94%|█████████▍| 348577/371472 [5:34:41<1:56:54,  3.26it/s]
 94%|█████████▍| 348578/371472 [5:34:41<1:52:40,  3.39it/s]
 94%|█████████▍| 348579/371472 [5:34:42<1:51:20,  3.43it/s]
 94%|█████████▍| 348580/371472 [5:34:42<1:52:26,  3.39it/s]
                                                           
{'loss': 2.6004, 'learning_rate': 1.5549246913314226e-07, 'epoch': 15.01}
+
 94%|█████████▍| 348580/371472 [5:34:42<1:52:26,  3.39it/s]
 94%|█████████▍| 348581/371472 [5:34:42<1:55:47,  3.29it/s]
 94%|█████████▍| 348582/371472 [5:34:43<1:55:51,  3.29it/s]
 94%|█████████▍| 348583/371472 [5:34:43<1:54:03,  3.34it/s]
 94%|█████████▍| 348584/371472 [5:34:43<1:53:28,  3.36it/s]
 94%|█████████▍| 348585/371472 [5:34:44<1:56:23,  3.28it/s]
 94%|█████████▍| 348586/371472 [5:34:44<1:54:51,  3.32it/s]
 94%|█████████▍| 348587/371472 [5:34:44<1:52:19,  3.40it/s]
 94%|█████████▍| 348588/371472 [5:34:45<2:05:42,  3.03it/s]
 94%|█████████▍| 348589/371472 [5:34:45<1:58:25,  3.22it/s]
 94%|█████████▍| 348590/371472 [5:34:45<2:04:06,  3.07it/s]
 94%|█████████▍| 348591/371472 [5:34:45<1:58:10,  3.23it/s]
 94%|█████████▍| 348592/371472 [5:34:46<1:55:54,  3.29it/s]
 94%|█████████▍| 348593/371472 [5:34:46<2:13:38,  2.85it/s]
 94%|█████████▍| 348594/371472 [5:34:46<2:03:19,  3.09it/s]
 94%|█████████▍| 348595/371472 [5:34:47<1:58:57,  3.21it/s]
 94%|█████████▍| 348596/371472 [5:34:47<1:55:11,  3.31it/s]
 94%|█████████▍| 348597/371472 [5:34:47<2:02:05,  3.12it/s]
 94%|█████████▍| 348598/371472 [5:34:48<1:58:03,  3.23it/s]
 94%|█████████▍| 348599/371472 [5:34:48<1:52:59,  3.37it/s]
 94%|█████████▍| 348600/371472 [5:34:48<1:50:10,  3.46it/s]
                                                           
{'loss': 2.6019, 'learning_rate': 1.554439871576634e-07, 'epoch': 15.01}
+
 94%|█████████▍| 348600/371472 [5:34:48<1:50:10,  3.46it/s]
 94%|█████████▍| 348601/371472 [5:34:49<1:53:47,  3.35it/s]
 94%|█████████▍| 348602/371472 [5:34:49<1:56:19,  3.28it/s]
 94%|█████████▍| 348603/371472 [5:34:49<1:54:56,  3.32it/s]
 94%|█████████▍| 348604/371472 [5:34:49<1:54:24,  3.33it/s]
 94%|█████████▍| 348605/371472 [5:34:50<1:49:33,  3.48it/s]
 94%|█████████▍| 348606/371472 [5:34:50<1:50:14,  3.46it/s]
 94%|█████████▍| 348607/371472 [5:34:50<1:47:07,  3.56it/s]
 94%|█████████▍| 348608/371472 [5:34:51<1:50:54,  3.44it/s]
 94%|█████████▍| 348609/371472 [5:34:51<1:47:08,  3.56it/s]
 94%|█████████▍| 348610/371472 [5:34:51<1:49:24,  3.48it/s]
 94%|█████████▍| 348611/371472 [5:34:51<1:51:15,  3.42it/s]
 94%|█████████▍| 348612/371472 [5:34:52<1:50:38,  3.44it/s]
 94%|█████████▍| 348613/371472 [5:34:52<1:46:54,  3.56it/s]
 94%|█████████▍| 348614/371472 [5:34:52<1:47:08,  3.56it/s]
 94%|█████████▍| 348615/371472 [5:34:53<1:47:30,  3.54it/s]
 94%|█████████▍| 348616/371472 [5:34:53<1:50:31,  3.45it/s]
 94%|█████████▍| 348617/371472 [5:34:53<1:47:32,  3.54it/s]
 94%|█████████▍| 348618/371472 [5:34:53<1:46:08,  3.59it/s]
 94%|█████████▍| 348619/371472 [5:34:54<1:48:52,  3.50it/s]
 94%|█████████▍| 348620/371472 [5:34:54<1:51:27,  3.42it/s]
                                                           
{'loss': 2.5714, 'learning_rate': 1.5539550518218448e-07, 'epoch': 15.02}
+
 94%|█████████▍| 348620/371472 [5:34:54<1:51:27,  3.42it/s]
 94%|█████████▍| 348621/371472 [5:34:54<1:50:15,  3.45it/s]
 94%|█████████▍| 348622/371472 [5:34:55<1:53:19,  3.36it/s]
 94%|█████████▍| 348623/371472 [5:34:55<1:53:14,  3.36it/s]
 94%|█████████▍| 348624/371472 [5:34:55<1:54:00,  3.34it/s]
 94%|█████████▍| 348625/371472 [5:34:55<1:51:05,  3.43it/s]
 94%|█████████▍| 348626/371472 [5:34:56<1:53:28,  3.36it/s]
 94%|█████████▍| 348627/371472 [5:34:56<2:06:53,  3.00it/s]
 94%|█████████▍| 348628/371472 [5:34:57<2:02:26,  3.11it/s]
 94%|█████████▍| 348629/371472 [5:34:57<1:58:12,  3.22it/s]
 94%|█████████▍| 348630/371472 [5:34:57<1:59:33,  3.18it/s]
 94%|█████████▍| 348631/371472 [5:34:57<1:56:20,  3.27it/s]
 94%|█████████▍| 348632/371472 [5:34:58<1:54:43,  3.32it/s]
 94%|█████████▍| 348633/371472 [5:34:58<1:51:25,  3.42it/s]
 94%|█████████▍| 348634/371472 [5:34:58<1:51:41,  3.41it/s]
 94%|█████████▍| 348635/371472 [5:34:59<2:00:27,  3.16it/s]
 94%|█████████▍| 348636/371472 [5:34:59<1:57:02,  3.25it/s]
 94%|█████████▍| 348637/371472 [5:34:59<1:53:05,  3.37it/s]
 94%|█████████▍| 348638/371472 [5:34:59<1:51:50,  3.40it/s]
 94%|█████████▍| 348639/371472 [5:35:00<1:58:32,  3.21it/s]
 94%|█████████▍| 348640/371472 [5:35:00<1:53:57,  3.34it/s]
                                                           
{'loss': 2.4869, 'learning_rate': 1.5534702320670563e-07, 'epoch': 15.02}
+
 94%|█████████▍| 348640/371472 [5:35:00<1:53:57,  3.34it/s]
 94%|█████████▍| 348641/371472 [5:35:00<1:54:28,  3.32it/s]
 94%|█████████▍| 348642/371472 [5:35:01<1:55:39,  3.29it/s]
 94%|█████████▍| 348643/371472 [5:35:01<2:01:59,  3.12it/s]
 94%|█████████▍| 348644/371472 [5:35:01<2:08:58,  2.95it/s]
 94%|█████████▍| 348645/371472 [5:35:02<2:13:48,  2.84it/s]
 94%|█████████▍| 348646/371472 [5:35:02<2:02:21,  3.11it/s]
 94%|█████████▍| 348647/371472 [5:35:02<2:02:09,  3.11it/s]
 94%|█████████▍| 348648/371472 [5:35:03<1:57:16,  3.24it/s]
 94%|█████████▍| 348649/371472 [5:35:03<1:55:44,  3.29it/s]
 94%|█████████▍| 348650/371472 [5:35:03<1:50:42,  3.44it/s]
 94%|█████████▍| 348651/371472 [5:35:04<1:46:37,  3.57it/s]
 94%|█████████▍| 348652/371472 [5:35:04<1:51:58,  3.40it/s]
 94%|█████████▍| 348653/371472 [5:35:04<2:03:42,  3.07it/s]
 94%|█████████▍| 348654/371472 [5:35:04<1:55:43,  3.29it/s]
 94%|██��██████▍| 348655/371472 [5:35:05<1:56:20,  3.27it/s]
 94%|█████████▍| 348656/371472 [5:35:05<1:56:11,  3.27it/s]
 94%|█████████▍| 348657/371472 [5:35:05<1:56:46,  3.26it/s]
 94%|█████████▍| 348658/371472 [5:35:06<1:56:36,  3.26it/s]
 94%|█████████▍| 348659/371472 [5:35:06<1:56:22,  3.27it/s]
 94%|█████████▍| 348660/371472 [5:35:06<1:55:46,  3.28it/s]
                                                           
{'loss': 2.4539, 'learning_rate': 1.5529854123122668e-07, 'epoch': 15.02}
+
 94%|█████████▍| 348660/371472 [5:35:06<1:55:46,  3.28it/s]
 94%|█████████▍| 348661/371472 [5:35:07<1:52:27,  3.38it/s]
 94%|█████████▍| 348662/371472 [5:35:07<1:46:57,  3.55it/s]
 94%|█████████▍| 348663/371472 [5:35:07<1:45:27,  3.60it/s]
 94%|█████████▍| 348664/371472 [5:35:07<1:45:04,  3.62it/s]
 94%|█████████▍| 348665/371472 [5:35:08<1:49:23,  3.48it/s]
 94%|█████████▍| 348666/371472 [5:35:08<1:55:44,  3.28it/s]
 94%|█████████▍| 348667/371472 [5:35:08<1:54:06,  3.33it/s]
 94%|█████████▍| 348668/371472 [5:35:09<1:51:41,  3.40it/s]
 94%|█████████▍| 348669/371472 [5:35:09<1:52:11,  3.39it/s]
 94%|█████████▍| 348670/371472 [5:35:09<1:48:24,  3.51it/s]
 94%|█████████▍| 348671/371472 [5:35:09<1:46:03,  3.58it/s]
 94%|█████████▍| 348672/371472 [5:35:10<1:44:54,  3.62it/s]
 94%|█████████▍| 348673/371472 [5:35:10<1:47:01,  3.55it/s]
 94%|█████████▍| 348674/371472 [5:35:10<1:44:00,  3.65it/s]
 94%|█████████▍| 348675/371472 [5:35:11<1:43:12,  3.68it/s]
 94%|█████████▍| 348676/371472 [5:35:11<1:50:10,  3.45it/s]
 94%|█████████▍| 348677/371472 [5:35:11<2:05:01,  3.04it/s]
 94%|█████████▍| 348678/371472 [5:35:12<2:10:33,  2.91it/s]
 94%|█████████▍| 348679/371472 [5:35:12<2:02:56,  3.09it/s]
 94%|█████████▍| 348680/371472 [5:35:12<1:57:11,  3.24it/s]
                                                           
{'loss': 2.7332, 'learning_rate': 1.5525005925574785e-07, 'epoch': 15.02}
+
 94%|█████████▍| 348680/371472 [5:35:12<1:57:11,  3.24it/s]
 94%|█████████▍| 348681/371472 [5:35:13<1:59:28,  3.18it/s]
 94%|█████████▍| 348682/371472 [5:35:13<1:56:45,  3.25it/s]
 94%|█████████▍| 348683/371472 [5:35:13<2:06:28,  3.00it/s]
 94%|█████████▍| 348684/371472 [5:35:14<2:01:57,  3.11it/s]
 94%|█████████▍| 348685/371472 [5:35:14<1:58:58,  3.19it/s]
 94%|█████████▍| 348686/371472 [5:35:14<2:00:20,  3.16it/s]
 94%|█████████▍| 348687/371472 [5:35:14<1:53:58,  3.33it/s]
 94%|█████████▍| 348688/371472 [5:35:15<1:58:35,  3.20it/s]
 94%|█████████▍| 348689/371472 [5:35:15<1:55:33,  3.29it/s]
 94%|█████████▍| 348690/371472 [5:35:15<1:59:27,  3.18it/s]
 94%|█████████▍| 348691/371472 [5:35:16<2:00:43,  3.14it/s]
 94%|█████████▍| 348692/371472 [5:35:16<1:53:56,  3.33it/s]
 94%|█████████▍| 348693/371472 [5:35:16<1:50:43,  3.43it/s]
 94%|█████████▍| 348694/371472 [5:35:17<1:52:31,  3.37it/s]
 94%|█████████▍| 348695/371472 [5:35:17<1:54:01,  3.33it/s]
 94%|█████████▍| 348696/371472 [5:35:17<1:53:23,  3.35it/s]
 94%|█████████▍| 348697/371472 [5:35:17<1:51:32,  3.40it/s]
 94%|█████████▍| 348698/371472 [5:35:18<1:50:40,  3.43it/s]
 94%|█████████▍| 348699/371472 [5:35:18<1:47:22,  3.53it/s]
 94%|█████████▍| 348700/371472 [5:35:18<1:49:07,  3.48it/s]
                                                           
{'loss': 2.4427, 'learning_rate': 1.552015772802689e-07, 'epoch': 15.02}
+
 94%|█████████▍| 348700/371472 [5:35:18<1:49:07,  3.48it/s]
 94%|█████████▍| 348701/371472 [5:35:19<1:50:25,  3.44it/s]
 94%|█████████▍| 348702/371472 [5:35:19<2:08:10,  2.96it/s]
 94%|█████████▍| 348703/371472 [5:35:19<2:11:12,  2.89it/s]
 94%|█████████▍| 348704/371472 [5:35:20<2:09:01,  2.94it/s]
 94%|█████████▍| 348705/371472 [5:35:20<2:03:32,  3.07it/s]
 94%|█████████▍| 348706/371472 [5:35:20<1:57:29,  3.23it/s]
 94%|█████████▍| 348707/371472 [5:35:21<1:54:05,  3.33it/s]
 94%|█████████▍| 348708/371472 [5:35:21<1:53:45,  3.33it/s]
 94%|█████████▍| 348709/371472 [5:35:21<2:20:14,  2.71it/s]
 94%|█████████▍| 348710/371472 [5:35:22<2:14:52,  2.81it/s]
 94%|���████████▍| 348711/371472 [5:35:22<2:13:03,  2.85it/s]
 94%|█████████▍| 348712/371472 [5:35:22<2:09:05,  2.94it/s]
 94%|█████████▍| 348713/371472 [5:35:23<2:06:35,  3.00it/s]
 94%|█████████▍| 348714/371472 [5:35:23<2:02:39,  3.09it/s]
 94%|█████████▍| 348715/371472 [5:35:23<2:00:16,  3.15it/s]
 94%|█████████▍| 348716/371472 [5:35:24<1:58:53,  3.19it/s]
 94%|█████████▍| 348717/371472 [5:35:24<1:56:17,  3.26it/s]
 94%|█████████▍| 348718/371472 [5:35:24<1:56:23,  3.26it/s]
 94%|█████████▍| 348719/371472 [5:35:24<1:51:53,  3.39it/s]
 94%|█████████▍| 348720/371472 [5:35:25<1:50:33,  3.43it/s]
                                                           
{'loss': 2.7404, 'learning_rate': 1.5515309530479005e-07, 'epoch': 15.02}
+
 94%|█████████▍| 348720/371472 [5:35:25<1:50:33,  3.43it/s]
 94%|█████████▍| 348721/371472 [5:35:25<1:45:59,  3.58it/s]
 94%|█████████▍| 348722/371472 [5:35:25<1:46:39,  3.56it/s]
 94%|█████████▍| 348723/371472 [5:35:26<1:50:02,  3.45it/s]
 94%|█████████▍| 348724/371472 [5:35:26<1:50:08,  3.44it/s]
 94%|█████████▍| 348725/371472 [5:35:26<1:59:33,  3.17it/s]
 94%|█████████▍| 348726/371472 [5:35:27<1:58:38,  3.20it/s]
 94%|█████████▍| 348727/371472 [5:35:27<2:00:31,  3.15it/s]
 94%|█████████▍| 348728/371472 [5:35:27<1:57:57,  3.21it/s]
 94%|█████████▍| 348729/371472 [5:35:27<1:53:40,  3.33it/s]
 94%|█████████▍| 348730/371472 [5:35:28<1:50:24,  3.43it/s]
 94%|█████████▍| 348731/371472 [5:35:28<1:47:51,  3.51it/s]
 94%|█████████▍| 348732/371472 [5:35:28<1:48:44,  3.49it/s]
 94%|█████████▍| 348733/371472 [5:35:29<1:43:55,  3.65it/s]
 94%|█████████▍| 348734/371472 [5:35:29<1:52:12,  3.38it/s]
 94%|█████████▍| 348735/371472 [5:35:29<1:49:04,  3.47it/s]
 94%|█████████▍| 348736/371472 [5:35:29<1:55:10,  3.29it/s]
 94%|█████████▍| 348737/371472 [5:35:30<1:53:49,  3.33it/s]
 94%|█████████▍| 348738/371472 [5:35:30<1:52:40,  3.36it/s]
 94%|█████████▍| 348739/371472 [5:35:30<1:50:39,  3.42it/s]
 94%|█████████▍| 348740/371472 [5:35:31<1:44:18,  3.63it/s]
                                                           
{'loss': 2.6401, 'learning_rate': 1.5510461332931112e-07, 'epoch': 15.02}
+
 94%|█████████▍| 348740/371472 [5:35:31<1:44:18,  3.63it/s]
 94%|█████████▍| 348741/371472 [5:35:31<1:42:25,  3.70it/s]
 94%|█████████▍| 348742/371472 [5:35:31<1:44:22,  3.63it/s]
 94%|█████████▍| 348743/371472 [5:35:31<1:41:57,  3.72it/s]
 94%|█████████▍| 348744/371472 [5:35:32<1:43:08,  3.67it/s]
 94%|█████████▍| 348745/371472 [5:35:32<1:42:23,  3.70it/s]
 94%|█████████▍| 348746/371472 [5:35:32<1:40:39,  3.76it/s]
 94%|█████████▍| 348747/371472 [5:35:32<1:43:11,  3.67it/s]
 94%|█████████▍| 348748/371472 [5:35:33<1:42:04,  3.71it/s]
 94%|█████████▍| 348749/371472 [5:35:33<1:43:43,  3.65it/s]
 94%|█████████▍| 348750/371472 [5:35:33<1:45:23,  3.59it/s]
 94%|█████████▍| 348751/371472 [5:35:34<1:49:12,  3.47it/s]
 94%|█████████▍| 348752/371472 [5:35:34<1:46:33,  3.55it/s]
 94%|█████████▍| 348753/371472 [5:35:34<1:43:49,  3.65it/s]
 94%|█████████▍| 348754/371472 [5:35:34<1:50:58,  3.41it/s]
 94%|█████████▍| 348755/371472 [5:35:35<1:50:42,  3.42it/s]
 94%|█████████▍| 348756/371472 [5:35:35<1:51:55,  3.38it/s]
 94%|█████████▍| 348757/371472 [5:35:35<1:47:38,  3.52it/s]
 94%|█████████▍| 348758/371472 [5:35:36<1:44:33,  3.62it/s]
 94%|█████████▍| 348759/371472 [5:35:36<1:43:06,  3.67it/s]
 94%|█████████▍| 348760/371472 [5:35:36<1:44:53,  3.61it/s]
                                                           
{'loss': 2.5081, 'learning_rate': 1.5505613135383217e-07, 'epoch': 15.02}
+
 94%|█████████▍| 348760/371472 [5:35:36<1:44:53,  3.61it/s]
 94%|█████████▍| 348761/371472 [5:35:37<2:02:43,  3.08it/s]
 94%|█████████▍| 348762/371472 [5:35:37<1:54:59,  3.29it/s]
 94%|█████████▍| 348763/371472 [5:35:37<1:47:50,  3.51it/s]
 94%|█████████▍| 348764/371472 [5:35:37<1:46:00,  3.57it/s]
 94%|█████████▍| 348765/371472 [5:35:38<1:42:48,  3.68it/s]
 94%|█████████▍| 348766/371472 [5:35:38<1:42:34,  3.69it/s]
 94%|█████████▍| 348767/371472 [5:35:38<1:44:30,  3.62it/s]
 94%|█████████▍| 348768/371472 [5:35:38<1:44:59,  3.60it/s]
 94%|█████████▍| 348769/371472 [5:35:39<1:41:56,  3.71it/s]
 94%|█████████▍| 348770/371472 [5:35:39<1:41:47,  3.72it/s]
 94%|█████████▍| 348771/371472 [5:35:39<1:50:43,  3.42it/s]
 94%|█████████▍| 348772/371472 [5:35:40<1:48:05,  3.50it/s]
 94%|█████████▍| 348773/371472 [5:35:40<1:45:35,  3.58it/s]
 94%|█████████▍| 348774/371472 [5:35:40<1:45:03,  3.60it/s]
 94%|█████████▍| 348775/371472 [5:35:40<1:51:14,  3.40it/s]
 94%|█████████▍| 348776/371472 [5:35:41<1:52:24,  3.36it/s]
 94%|█████████▍| 348777/371472 [5:35:41<1:51:54,  3.38it/s]
 94%|█████████▍| 348778/371472 [5:35:41<1:50:10,  3.43it/s]
 94%|█████████▍| 348779/371472 [5:35:42<1:47:05,  3.53it/s]
 94%|█████████▍| 348780/371472 [5:35:42<1:46:12,  3.56it/s]
                                                           
{'loss': 2.5531, 'learning_rate': 1.5500764937835334e-07, 'epoch': 15.02}
+
 94%|█████████▍| 348780/371472 [5:35:42<1:46:12,  3.56it/s]
 94%|█████████▍| 348781/371472 [5:35:42<1:43:45,  3.65it/s]
 94%|█████████▍| 348782/371472 [5:35:42<1:45:52,  3.57it/s]
 94%|█████████▍| 348783/371472 [5:35:43<1:48:46,  3.48it/s]
 94%|█████████▍| 348784/371472 [5:35:43<1:45:52,  3.57it/s]
 94%|█████████▍| 348785/371472 [5:35:43<1:46:25,  3.55it/s]
 94%|█████████▍| 348786/371472 [5:35:44<1:46:02,  3.57it/s]
 94%|█████████▍| 348787/371472 [5:35:44<1:47:25,  3.52it/s]
 94%|█████████▍| 348788/371472 [5:35:44<1:49:49,  3.44it/s]
 94%|█████████▍| 348789/371472 [5:35:44<1:51:13,  3.40it/s]
 94%|█████████▍| 348790/371472 [5:35:45<1:50:50,  3.41it/s]
 94%|█████████▍| 348791/371472 [5:35:45<2:07:23,  2.97it/s]
 94%|█████████▍| 348792/371472 [5:35:45<2:01:42,  3.11it/s]
 94%|█████████▍| 348793/371472 [5:35:46<2:04:03,  3.05it/s]
 94%|█████████▍| 348794/371472 [5:35:46<1:55:50,  3.26it/s]
 94%|█████████▍| 348795/371472 [5:35:46<1:50:25,  3.42it/s]
 94%|█████████▍| 348796/371472 [5:35:47<1:44:44,  3.61it/s]
 94%|█████████▍| 348797/371472 [5:35:47<1:47:04,  3.53it/s]
 94%|█████████▍| 348798/371472 [5:35:47<1:45:45,  3.57it/s]
 94%|█████████▍| 348799/371472 [5:35:47<1:43:19,  3.66it/s]
 94%|█████████▍| 348800/371472 [5:35:48<1:48:46,  3.47it/s]
                                                           
{'loss': 2.5296, 'learning_rate': 1.5495916740287436e-07, 'epoch': 15.02}
+
 94%|█████████▍| 348800/371472 [5:35:48<1:48:46,  3.47it/s]
 94%|█████████▍| 348801/371472 [5:35:48<1:52:42,  3.35it/s]
 94%|█████████▍| 348802/371472 [5:35:48<1:47:50,  3.50it/s]
 94%|█████████▍| 348803/371472 [5:35:49<1:45:11,  3.59it/s]
 94%|█████████▍| 348804/371472 [5:35:49<1:50:52,  3.41it/s]
 94%|█████████▍| 348805/371472 [5:35:49<1:59:57,  3.15it/s]
 94%|█████████▍| 348806/371472 [5:35:50<2:00:30,  3.13it/s]
 94%|█████████▍| 348807/371472 [5:35:50<1:59:41,  3.16it/s]
 94%|█████████▍| 348808/371472 [5:35:50<1:53:58,  3.31it/s]
 94%|█████████▍| 348809/371472 [5:35:50<1:51:47,  3.38it/s]
 94%|█████████▍| 348810/371472 [5:35:51<1:51:35,  3.38it/s]
 94%|█████████▍| 348811/371472 [5:35:51<1:51:17,  3.39it/s]
 94%|█████████▍| 348812/371472 [5:35:51<1:46:06,  3.56it/s]
 94%|█████████▍| 348813/371472 [5:35:52<1:49:32,  3.45it/s]
 94%|█████████▍| 348814/371472 [5:35:52<1:49:13,  3.46it/s]
 94%|█████████▍| 348815/371472 [5:35:52<1:46:31,  3.54it/s]
 94%|█████████▍| 348816/371472 [5:35:52<1:50:09,  3.43it/s]
 94%|█████████▍| 348817/371472 [5:35:53<1:49:05,  3.46it/s]
 94%|█████████▍| 348818/371472 [5:35:53<1:47:17,  3.52it/s]
 94%|█████████▍| 348819/371472 [5:35:53<1:45:04,  3.59it/s]
 94%|█████████▍| 348820/371472 [5:35:54<1:44:03,  3.63it/s]
                                                           
{'loss': 2.6226, 'learning_rate': 1.5491068542739554e-07, 'epoch': 15.02}
+
 94%|█████████▍| 348820/371472 [5:35:54<1:44:03,  3.63it/s]
 94%|█████████▍| 348821/371472 [5:35:54<1:43:59,  3.63it/s]
 94%|█████████▍| 348822/371472 [5:35:54<1:45:19,  3.58it/s]
 94%|█████████▍| 348823/371472 [5:35:54<1:44:46,  3.60it/s]
 94%|█████████▍| 348824/371472 [5:35:55<1:48:02,  3.49it/s]
 94%|█████████▍| 348825/371472 [5:35:55<1:46:58,  3.53it/s]
 94%|█████████▍| 348826/371472 [5:35:55<1:47:42,  3.50it/s]
 94%|█████████▍| 348827/371472 [5:35:55<1:45:20,  3.58it/s]
 94%|█████████▍| 348828/371472 [5:35:56<1:48:49,  3.47it/s]
 94%|█████████▍| 348829/371472 [5:35:56<1:51:16,  3.39it/s]
 94%|█████████▍| 348830/371472 [5:35:56<1:47:28,  3.51it/s]
 94%|█████████▍| 348831/371472 [5:35:57<1:42:04,  3.70it/s]
 94%|█████████▍| 348832/371472 [5:35:57<1:43:58,  3.63it/s]
 94%|█████████▍| 348833/371472 [5:35:57<1:45:28,  3.58it/s]
 94%|█████████▍| 348834/371472 [5:35:57<1:46:21,  3.55it/s]
 94%|█████████▍| 348835/371472 [5:35:58<1:44:19,  3.62it/s]
 94%|█████████▍| 348836/371472 [5:35:58<1:45:09,  3.59it/s]
 94%|█████████▍| 348837/371472 [5:35:58<1:46:53,  3.53it/s]
 94%|█████████▍| 348838/371472 [5:35:59<1:45:48,  3.57it/s]
 94%|█████████▍| 348839/371472 [5:35:59<1:48:51,  3.47it/s]
 94%|█████████▍| 348840/371472 [5:35:59<1:46:02,  3.56it/s]
                                                           
{'loss': 2.589, 'learning_rate': 1.548622034519166e-07, 'epoch': 15.03}
+
 94%|█████████▍| 348840/371472 [5:35:59<1:46:02,  3.56it/s]
 94%|█████████▍| 348841/371472 [5:35:59<1:45:46,  3.57it/s]
 94%|█████████▍| 348842/371472 [5:36:00<1:46:32,  3.54it/s]
 94%|█████████▍| 348843/371472 [5:36:00<1:49:14,  3.45it/s]
 94%|█████████▍| 348844/371472 [5:36:00<1:48:20,  3.48it/s]
 94%|█████████▍| 348845/371472 [5:36:01<1:49:27,  3.45it/s]
 94%|█████████▍| 348846/371472 [5:36:01<1:45:49,  3.56it/s]
 94%|█████████▍| 348847/371472 [5:36:01<1:44:52,  3.60it/s]
 94%|█████████▍| 348848/371472 [5:36:02<1:54:05,  3.30it/s]
 94%|█████████▍| 348849/371472 [5:36:02<1:52:17,  3.36it/s]
 94%|█████████▍| 348850/371472 [5:36:02<1:50:05,  3.42it/s]
 94%|█████████▍| 348851/371472 [5:36:02<1:51:02,  3.40it/s]
 94%|█████████▍| 348852/371472 [5:36:03<1:50:05,  3.42it/s]
 94%|█████████▍| 348853/371472 [5:36:03<1:48:57,  3.46it/s]
 94%|█████████▍| 348854/371472 [5:36:03<1:53:52,  3.31it/s]
 94%|█████████▍| 348855/371472 [5:36:04<1:54:57,  3.28it/s]
 94%|█████████▍| 348856/371472 [5:36:04<1:50:53,  3.40it/s]
 94%|█████████▍| 348857/371472 [5:36:04<1:56:57,  3.22it/s]
 94%|█████████▍| 348858/371472 [5:36:05<1:55:39,  3.26it/s]
 94%|█████████▍| 348859/371472 [5:36:05<1:51:53,  3.37it/s]
 94%|█████████▍| 348860/371472 [5:36:05<1:50:20,  3.42it/s]
                                                           
{'loss': 2.6768, 'learning_rate': 1.5481372147643776e-07, 'epoch': 15.03}
+
 94%|█████████▍| 348860/371472 [5:36:05<1:50:20,  3.42it/s]
 94%|█████████▍| 348861/371472 [5:36:05<1:46:28,  3.54it/s]
 94%|█████████▍| 348862/371472 [5:36:06<1:53:58,  3.31it/s]
 94%|█████████▍| 348863/371472 [5:36:06<1:51:24,  3.38it/s]
 94%|█████████▍| 348864/371472 [5:36:06<1:50:02,  3.42it/s]
 94%|█████████▍| 348865/371472 [5:36:06<1:46:28,  3.54it/s]
 94%|█████████▍| 348866/371472 [5:36:07<1:51:49,  3.37it/s]
 94%|█████████▍| 348867/371472 [5:36:07<2:00:06,  3.14it/s]
 94%|█████████▍| 348868/371472 [5:36:07<1:53:57,  3.31it/s]
 94%|█████████▍| 348869/371472 [5:36:08<1:52:27,  3.35it/s]
 94%|█████████▍| 348870/371472 [5:36:08<1:48:47,  3.46it/s]
 94%|█████████▍| 348871/371472 [5:36:08<1:47:20,  3.51it/s]
 94%|█████████▍| 348872/371472 [5:36:09<1:44:13,  3.61it/s]
 94%|█████████▍| 348873/371472 [5:36:09<1:45:02,  3.59it/s]
 94%|█████████▍| 348874/371472 [5:36:09<1:48:38,  3.47it/s]
 94%|█████████▍| 348875/371472 [5:36:09<1:45:41,  3.56it/s]
 94%|█████████▍| 348876/371472 [5:36:10<1:48:45,  3.46it/s]
 94%|█████████▍| 348877/371472 [5:36:10<1:47:39,  3.50it/s]
 94%|█████████▍| 348878/371472 [5:36:10<1:52:10,  3.36it/s]
 94%|█████████▍| 348879/371472 [5:36:11<1:49:08,  3.45it/s]
 94%|█████████▍| 348880/371472 [5:36:11<2:00:48,  3.12it/s]
                                                           
{'loss': 2.566, 'learning_rate': 1.5476523950095883e-07, 'epoch': 15.03}
+
 94%|█████████▍| 348880/371472 [5:36:11<2:00:48,  3.12it/s]
 94%|█████████▍| 348881/371472 [5:36:11<1:57:34,  3.20it/s]
 94%|█████████▍| 348882/371472 [5:36:12<1:53:14,  3.32it/s]
 94%|█████████▍| 348883/371472 [5:36:12<1:48:20,  3.47it/s]
 94%|█████████▍| 348884/371472 [5:36:12<1:47:42,  3.50it/s]
 94%|█████████▍| 348885/371472 [5:36:12<1:45:04,  3.58it/s]
 94%|█████████▍| 348886/371472 [5:36:13<1:46:11,  3.54it/s]
 94%|█████████▍| 348887/371472 [5:36:13<1:43:14,  3.65it/s]
 94%|█████████▍| 348888/371472 [5:36:13<1:51:18,  3.38it/s]
 94%|█████████▍| 348889/371472 [5:36:14<1:50:25,  3.41it/s]
 94%|█████████▍| 348890/371472 [5:36:14<1:55:10,  3.27it/s]
 94%|█████████▍| 348891/371472 [5:36:14<1:53:12,  3.32it/s]
 94%|█████████▍| 348892/371472 [5:36:14<1:48:30,  3.47it/s]
 94%|█████████▍| 348893/371472 [5:36:15<1:55:54,  3.25it/s]
 94%|█████████▍| 348894/371472 [5:36:15<1:54:27,  3.29it/s]
 94%|█████████▍| 348895/371472 [5:36:15<1:52:25,  3.35it/s]
 94%|█████████▍| 348896/371472 [5:36:16<1:51:50,  3.36it/s]
 94%|█████████▍| 348897/371472 [5:36:16<1:55:04,  3.27it/s]
 94%|█████████▍| 348898/371472 [5:36:16<1:56:11,  3.24it/s]
 94%|█████████▍| 348899/371472 [5:36:17<1:56:08,  3.24it/s]
 94%|█████████▍| 348900/371472 [5:36:17<1:53:06,  3.33it/s]
                                                           
{'loss': 2.6027, 'learning_rate': 1.5471675752547998e-07, 'epoch': 15.03}
+
 94%|█████████▍| 348900/371472 [5:36:17<1:53:06,  3.33it/s]
 94%|█████████▍| 348901/371472 [5:36:17<1:48:32,  3.47it/s]
 94%|█████████▍| 348902/371472 [5:36:17<1:46:54,  3.52it/s]
 94%|█████████▍| 348903/371472 [5:36:18<1:44:44,  3.59it/s]
 94%|█████████▍| 348904/371472 [5:36:18<1:42:58,  3.65it/s]
 94%|█████████▍| 348905/371472 [5:36:18<1:40:11,  3.75it/s]
 94%|█████████▍| 348906/371472 [5:36:18<1:38:02,  3.84it/s]
 94%|█████████▍| 348907/371472 [5:36:19<1:41:16,  3.71it/s]
 94%|█████████▍| 348908/371472 [5:36:19<1:45:31,  3.56it/s]
 94%|█████████▍| 348909/371472 [5:36:19<1:43:39,  3.63it/s]
 94%|█████████▍| 348910/371472 [5:36:20<1:44:12,  3.61it/s]
 94%|█████████▍| 348911/371472 [5:36:20<1:46:55,  3.52it/s]
 94%|█████████▍| 348912/371472 [5:36:20<1:50:35,  3.40it/s]
 94%|█████████▍| 348913/371472 [5:36:20<1:52:08,  3.35it/s]
 94%|█████████▍| 348914/371472 [5:36:21<1:54:35,  3.28it/s]
 94%|█████████▍| 348915/371472 [5:36:21<1:53:13,  3.32it/s]
 94%|█████████▍| 348916/371472 [5:36:21<1:49:53,  3.42it/s]
 94%|█████████▍| 348917/371472 [5:36:22<1:48:54,  3.45it/s]
 94%|█████████▍| 348918/371472 [5:36:22<1:53:07,  3.32it/s]
 94%|█████████▍| 348919/371472 [5:36:22<1:55:38,  3.25it/s]
 94%|█████████▍| 348920/371472 [5:36:23<1:55:25,  3.26it/s]
                                                           
{'loss': 2.7091, 'learning_rate': 1.5466827555000103e-07, 'epoch': 15.03}
+
 94%|█████████▍| 348920/371472 [5:36:23<1:55:25,  3.26it/s]
 94%|█████████▍| 348921/371472 [5:36:23<1:50:14,  3.41it/s]
 94%|█████████▍| 348922/371472 [5:36:23<1:54:46,  3.27it/s]
 94%|█████████▍| 348923/371472 [5:36:23<1:51:07,  3.38it/s]
 94%|█████████▍| 348924/371472 [5:36:24<1:48:03,  3.48it/s]
 94%|█████████▍| 348925/371472 [5:36:24<1:46:58,  3.51it/s]
 94%|█████████▍| 348926/371472 [5:36:24<1:42:04,  3.68it/s]
 94%|█████████▍| 348927/371472 [5:36:25<1:43:33,  3.63it/s]
 94%|█████████▍| 348928/371472 [5:36:25<1:42:30,  3.67it/s]
 94%|█████████▍| 348929/371472 [5:36:25<1:52:21,  3.34it/s]
 94%|█████████▍| 348930/371472 [5:36:25<1:53:54,  3.30it/s]
 94%|█████████▍| 348931/371472 [5:36:26<1:53:30,  3.31it/s]
 94%|█████████▍| 348932/371472 [5:36:26<1:57:30,  3.20it/s]
 94%|█████████▍| 348933/371472 [5:36:26<1:52:14,  3.35it/s]
 94%|█████████▍| 348934/371472 [5:36:27<1:55:54,  3.24it/s]
 94%|█████████▍| 348935/371472 [5:36:27<1:54:56,  3.27it/s]
 94%|█████████▍| 348936/371472 [5:36:27<1:50:15,  3.41it/s]
 94%|█████████���| 348937/371472 [5:36:28<1:55:46,  3.24it/s]
 94%|█████████▍| 348938/371472 [5:36:28<1:49:51,  3.42it/s]
 94%|█████████▍| 348939/371472 [5:36:28<1:53:00,  3.32it/s]
 94%|█████████▍| 348940/371472 [5:36:29<1:54:26,  3.28it/s]
                                                           
{'loss': 2.5573, 'learning_rate': 1.546197935745222e-07, 'epoch': 15.03}
+
 94%|█████████▍| 348940/371472 [5:36:29<1:54:26,  3.28it/s]
 94%|█████████▍| 348941/371472 [5:36:29<1:49:58,  3.41it/s]
 94%|█████████▍| 348942/371472 [5:36:29<1:51:11,  3.38it/s]
 94%|█████████▍| 348943/371472 [5:36:29<1:56:59,  3.21it/s]
 94%|█████████▍| 348944/371472 [5:36:30<2:03:05,  3.05it/s]
 94%|█████████▍| 348945/371472 [5:36:30<2:05:18,  3.00it/s]
 94%|█████████▍| 348946/371472 [5:36:31<2:07:25,  2.95it/s]
 94%|█████████▍| 348947/371472 [5:36:31<2:04:05,  3.03it/s]
 94%|█████████▍| 348948/371472 [5:36:31<2:01:03,  3.10it/s]
 94%|█████████▍| 348949/371472 [5:36:31<1:54:42,  3.27it/s]
 94%|█████████▍| 348950/371472 [5:36:32<1:57:45,  3.19it/s]
 94%|█████████▍| 348951/371472 [5:36:32<1:52:43,  3.33it/s]
 94%|█████████▍| 348952/371472 [5:36:32<1:53:26,  3.31it/s]
 94%|█████████▍| 348953/371472 [5:36:33<1:49:31,  3.43it/s]
 94%|█████████▍| 348954/371472 [5:36:33<1:49:57,  3.41it/s]
 94%|█████████▍| 348955/371472 [5:36:33<1:45:38,  3.55it/s]
 94%|█████████▍| 348956/371472 [5:36:33<1:45:36,  3.55it/s]
 94%|█████████▍| 348957/371472 [5:36:34<1:49:31,  3.43it/s]
 94%|█████████▍| 348958/371472 [5:36:34<1:46:57,  3.51it/s]
 94%|█████████▍| 348959/371472 [5:36:34<1:53:48,  3.30it/s]
 94%|█████████▍| 348960/371472 [5:36:35<1:52:28,  3.34it/s]
                                                           
{'loss': 2.5873, 'learning_rate': 1.5457131159904325e-07, 'epoch': 15.03}
+
 94%|█████████▍| 348960/371472 [5:36:35<1:52:28,  3.34it/s]
 94%|█████████▍| 348961/371472 [5:36:35<1:46:07,  3.54it/s]
 94%|█████████▍| 348962/371472 [5:36:35<1:43:10,  3.64it/s]
 94%|█████████▍| 348963/371472 [5:36:35<1:48:10,  3.47it/s]
 94%|█████████▍| 348964/371472 [5:36:36<1:54:40,  3.27it/s]
 94%|█████████▍| 348965/371472 [5:36:36<1:56:20,  3.22it/s]
 94%|█████████▍| 348966/371472 [5:36:36<1:50:16,  3.40it/s]
 94%|█████████▍| 348967/371472 [5:36:37<1:55:54,  3.24it/s]
 94%|█████████▍| 348968/371472 [5:36:37<1:52:40,  3.33it/s]
 94%|█████████▍| 348969/371472 [5:36:37<1:47:53,  3.48it/s]
 94%|█████████▍| 348970/371472 [5:36:38<1:50:47,  3.39it/s]
 94%|█████████▍| 348971/371472 [5:36:38<1:49:49,  3.41it/s]
 94%|█████████▍| 348972/371472 [5:36:38<1:46:56,  3.51it/s]
 94%|█████████▍| 348973/371472 [5:36:38<1:46:02,  3.54it/s]
 94%|█████████▍| 348974/371472 [5:36:39<1:42:54,  3.64it/s]
 94%|█████████▍| 348975/371472 [5:36:39<1:43:59,  3.61it/s]
 94%|█████████▍| 348976/371472 [5:36:39<1:54:23,  3.28it/s]
 94%|█████████▍| 348977/371472 [5:36:40<1:48:19,  3.46it/s]
 94%|█████████▍| 348978/371472 [5:36:40<1:56:43,  3.21it/s]
 94%|█████████▍| 348979/371472 [5:36:40<1:53:15,  3.31it/s]
 94%|█████████▍| 348980/371472 [5:36:41<1:57:04,  3.20it/s]
                                                           
{'loss': 2.5214, 'learning_rate': 1.5452282962356443e-07, 'epoch': 15.03}
+
 94%|█████████▍| 348980/371472 [5:36:41<1:57:04,  3.20it/s]
 94%|█████████▍| 348981/371472 [5:36:41<1:54:04,  3.29it/s]
 94%|█████████▍| 348982/371472 [5:36:41<2:05:15,  2.99it/s]
 94%|█████████▍| 348983/371472 [5:36:41<1:55:38,  3.24it/s]
 94%|█████████▍| 348984/371472 [5:36:42<1:53:24,  3.30it/s]
 94%|█████████▍| 348985/371472 [5:36:42<1:57:11,  3.20it/s]
 94%|█████████▍| 348986/371472 [5:36:42<1:56:14,  3.22it/s]
 94%|█████████▍| 348987/371472 [5:36:43<1:59:08,  3.15it/s]
 94%|█████████▍| 348988/371472 [5:36:43<1:52:21,  3.34it/s]
 94%|█████████▍| 348989/371472 [5:36:43<1:48:09,  3.46it/s]
 94%|█████████▍| 348990/371472 [5:36:44<1:45:13,  3.56it/s]
 94%|█████████▍| 348991/371472 [5:36:44<1:52:05,  3.34it/s]
 94%|█████████▍| 348992/371472 [5:36:44<1:50:19,  3.40it/s]
 94%|█████████▍| 348993/371472 [5:36:44<1:47:09,  3.50it/s]
 94%|█████████▍| 348994/371472 [5:36:45<1:46:24,  3.52it/s]
 94%|█████████▍| 348995/371472 [5:36:45<1:44:44,  3.58it/s]
 94%|█████████▍| 348996/371472 [5:36:45<1:47:13,  3.49it/s]
 94%|█████████▍| 348997/371472 [5:36:46<1:44:47,  3.57it/s]
 94%|█████████▍| 348998/371472 [5:36:46<1:44:45,  3.58it/s]
 94%|█████████▍| 348999/371472 [5:36:46<1:45:02,  3.57it/s]
 94%|█████████▍| 349000/371472 [5:36:46<1:41:28,  3.69it/s]
                                                           
{'loss': 2.6873, 'learning_rate': 1.5447434764808547e-07, 'epoch': 15.03}
+
 94%|█████████▍| 349000/371472 [5:36:46<1:41:28,  3.69it/s]
 94%|█████████▍| 349001/371472 [5:36:47<1:45:02,  3.57it/s]
 94%|█████████▍| 349002/371472 [5:36:47<1:44:35,  3.58it/s]
 94%|█████████▍| 349003/371472 [5:36:47<1:48:12,  3.46it/s]
 94%|█████████▍| 349004/371472 [5:36:48<1:51:06,  3.37it/s]
 94%|█████████▍| 349005/371472 [5:36:48<1:49:34,  3.42it/s]
 94%|█████████▍| 349006/371472 [5:36:48<1:44:00,  3.60it/s]
 94%|█████████▍| 349007/371472 [5:36:48<1:42:39,  3.65it/s]
 94%|█████████▍| 349008/371472 [5:36:49<1:43:26,  3.62it/s]
 94%|█████████▍| 349009/371472 [5:36:49<1:45:40,  3.54it/s]
 94%|█████████▍| 349010/371472 [5:36:49<1:46:16,  3.52it/s]
 94%|█████████▍| 349011/371472 [5:36:49<1:45:10,  3.56it/s]
 94%|█████████▍| 349012/371472 [5:36:50<1:47:25,  3.48it/s]
 94%|█████████▍| 349013/371472 [5:36:50<1:48:32,  3.45it/s]
 94%|█████████▍| 349014/371472 [5:36:50<1:49:05,  3.43it/s]
 94%|█████████▍| 349015/371472 [5:36:51<1:45:06,  3.56it/s]
 94%|█████████▍| 349016/371472 [5:36:51<1:45:37,  3.54it/s]
 94%|█████████▍| 349017/371472 [5:36:51<1:43:29,  3.62it/s]
 94%|█████████▍| 349018/371472 [5:36:51<1:42:21,  3.66it/s]
 94%|█████████▍| 349019/371472 [5:36:52<1:52:17,  3.33it/s]
 94%|█████████▍| 349020/371472 [5:36:52<1:48:58,  3.43it/s]
                                                           
{'loss': 2.5452, 'learning_rate': 1.5442586567260662e-07, 'epoch': 15.03}
+
 94%|█████████▍| 349020/371472 [5:36:52<1:48:58,  3.43it/s]
 94%|█████████▍| 349021/371472 [5:36:52<1:52:58,  3.31it/s]
 94%|█████████▍| 349022/371472 [5:36:53<1:50:45,  3.38it/s]
 94%|█████████▍| 349023/371472 [5:36:53<1:51:36,  3.35it/s]
 94%|█████████▍| 349024/371472 [5:36:53<1:53:39,  3.29it/s]
 94%|█████████▍| 349025/371472 [5:36:54<1:49:18,  3.42it/s]
 94%|█████████▍| 349026/371472 [5:36:54<1:54:54,  3.26it/s]
 94%|█████████▍| 349027/371472 [5:36:54<1:50:40,  3.38it/s]
 94%|█████████▍| 349028/371472 [5:36:55<1:55:38,  3.23it/s]
 94%|█████████▍| 349029/371472 [5:36:55<1:52:54,  3.31it/s]
 94%|█████████▍| 349030/371472 [5:36:55<1:50:03,  3.40it/s]
 94%|█████████▍| 349031/371472 [5:36:55<1:47:21,  3.48it/s]
 94%|█████████▍| 349032/371472 [5:36:56<1:53:05,  3.31it/s]
 94%|█████████▍| 349033/371472 [5:36:56<1:52:12,  3.33it/s]
 94%|█████████▍| 349034/371472 [5:36:56<1:52:47,  3.32it/s]
 94%|█████████▍| 349035/371472 [5:36:57<1:50:54,  3.37it/s]
 94%|█████████▍| 349036/371472 [5:36:57<1:48:54,  3.43it/s]
 94%|█████████▍| 349037/371472 [5:36:57<1:45:04,  3.56it/s]
 94%|█████████▍| 349038/371472 [5:36:57<1:44:11,  3.59it/s]
 94%|█████████▍| 349039/371472 [5:36:58<1:48:15,  3.45it/s]
 94%|█████████▍| 349040/371472 [5:36:58<1:44:17,  3.58it/s]
                                                           
{'loss': 2.6069, 'learning_rate': 1.5437738369712767e-07, 'epoch': 15.03}
+
 94%|█████████▍| 349040/371472 [5:36:58<1:44:17,  3.58it/s]
 94%|█████████▍| 349041/371472 [5:36:58<1:47:55,  3.46it/s]
 94%|█████████▍| 349042/371472 [5:36:59<1:43:10,  3.62it/s]
 94%|█████████▍| 349043/371472 [5:36:59<1:41:56,  3.67it/s]
 94%|█████████▍| 349044/371472 [5:36:59<1:43:35,  3.61it/s]
 94%|█████████▍| 349045/371472 [5:36:59<1:45:01,  3.56it/s]
 94%|█████████▍| 349046/371472 [5:37:00<1:56:26,  3.21it/s]
 94%|█████████▍| 349047/371472 [5:37:00<1:49:11,  3.42it/s]
 94%|█████████▍| 349048/371472 [5:37:00<1:52:32,  3.32it/s]
 94%|████���████▍| 349049/371472 [5:37:01<1:48:34,  3.44it/s]
 94%|█████████▍| 349050/371472 [5:37:01<1:49:09,  3.42it/s]
 94%|█████████▍| 349051/371472 [5:37:01<1:47:41,  3.47it/s]
 94%|█████████▍| 349052/371472 [5:37:01<1:53:45,  3.28it/s]
 94%|█████████▍| 349053/371472 [5:37:02<1:52:06,  3.33it/s]
 94%|█████████▍| 349054/371472 [5:37:02<1:49:02,  3.43it/s]
 94%|█████████▍| 349055/371472 [5:37:02<1:45:01,  3.56it/s]
 94%|█████████▍| 349056/371472 [5:37:03<1:44:35,  3.57it/s]
 94%|█████████▍| 349057/371472 [5:37:03<1:42:11,  3.66it/s]
 94%|█████████▍| 349058/371472 [5:37:03<1:49:56,  3.40it/s]
 94%|█████████▍| 349059/371472 [5:37:03<1:45:14,  3.55it/s]
 94%|█████████▍| 349060/371472 [5:37:04<1:41:07,  3.69it/s]
                                                           
{'loss': 2.502, 'learning_rate': 1.5432890172164885e-07, 'epoch': 15.03}
+
 94%|█████████▍| 349060/371472 [5:37:04<1:41:07,  3.69it/s]
 94%|█████████▍| 349061/371472 [5:37:04<1:38:54,  3.78it/s]
 94%|█████████▍| 349062/371472 [5:37:04<1:41:11,  3.69it/s]
 94%|█████████▍| 349063/371472 [5:37:05<1:45:19,  3.55it/s]
 94%|█████████▍| 349064/371472 [5:37:05<1:43:29,  3.61it/s]
 94%|█████████▍| 349065/371472 [5:37:05<1:49:55,  3.40it/s]
 94%|█████████▍| 349066/371472 [5:37:05<1:58:08,  3.16it/s]
 94%|█████████▍| 349067/371472 [5:37:06<1:55:10,  3.24it/s]
 94%|█████████▍| 349068/371472 [5:37:06<1:49:21,  3.41it/s]
 94%|█████████▍| 349069/371472 [5:37:06<1:47:55,  3.46it/s]
 94%|█████████▍| 349070/371472 [5:37:07<1:46:20,  3.51it/s]
 94%|█████████▍| 349071/371472 [5:37:07<1:41:44,  3.67it/s]
 94%|█████████▍| 349072/371472 [5:37:07<1:41:45,  3.67it/s]
 94%|█████████▍| 349073/371472 [5:37:07<1:41:09,  3.69it/s]
 94%|█████████▍| 349074/371472 [5:37:08<1:43:29,  3.61it/s]
 94%|█████████▍| 349075/371472 [5:37:08<1:38:47,  3.78it/s]
 94%|█████████▍| 349076/371472 [5:37:08<1:38:08,  3.80it/s]
 94%|█████████▍| 349077/371472 [5:37:08<1:36:32,  3.87it/s]
 94%|█████████▍| 349078/371472 [5:37:09<1:41:05,  3.69it/s]
 94%|█████████▍| 349079/371472 [5:37:09<1:39:17,  3.76it/s]
 94%|█████████▍| 349080/371472 [5:37:09<1:40:39,  3.71it/s]
                                                           
{'loss': 2.5976, 'learning_rate': 1.542804197461699e-07, 'epoch': 15.04}
+
 94%|█████████▍| 349080/371472 [5:37:09<1:40:39,  3.71it/s]
 94%|█████████▍| 349081/371472 [5:37:10<1:41:18,  3.68it/s]
 94%|█████████▍| 349082/371472 [5:37:10<1:43:00,  3.62it/s]
 94%|█████████▍| 349083/371472 [5:37:10<1:46:17,  3.51it/s]
 94%|█████████▍| 349084/371472 [5:37:10<1:44:10,  3.58it/s]
 94%|█████████▍| 349085/371472 [5:37:11<1:42:10,  3.65it/s]
 94%|█████████▍| 349086/371472 [5:37:11<1:50:11,  3.39it/s]
 94%|█████████▍| 349087/371472 [5:37:11<1:46:04,  3.52it/s]
 94%|█████████▍| 349088/371472 [5:37:12<1:50:16,  3.38it/s]
 94%|█████████▍| 349089/371472 [5:37:12<1:48:51,  3.43it/s]
 94%|█████████▍| 349090/371472 [5:37:12<1:45:42,  3.53it/s]
 94%|█████████▍| 349091/371472 [5:37:12<1:56:43,  3.20it/s]
 94%|█████████▍| 349092/371472 [5:37:13<1:57:31,  3.17it/s]
 94%|█████████▍| 349093/371472 [5:37:13<1:55:18,  3.23it/s]
 94%|█████████▍| 349094/371472 [5:37:13<1:49:36,  3.40it/s]
 94%|█████████▍| 349095/371472 [5:37:14<1:47:38,  3.46it/s]
 94%|█████████▍| 349096/371472 [5:37:14<1:52:05,  3.33it/s]
 94%|█████████▍| 349097/371472 [5:37:14<1:47:28,  3.47it/s]
 94%|█████████▍| 349098/371472 [5:37:15<1:48:44,  3.43it/s]
 94%|█████████▍| 349099/371472 [5:37:15<1:48:49,  3.43it/s]
 94%|█████████▍| 349100/371472 [5:37:15<1:46:55,  3.49it/s]
                                                           
{'loss': 2.6143, 'learning_rate': 1.5423193777069104e-07, 'epoch': 15.04}
+
 94%|█████████▍| 349100/371472 [5:37:15<1:46:55,  3.49it/s]
 94%|█████████▍| 349101/371472 [5:37:15<1:46:22,  3.51it/s]
 94%|█████████▍| 349102/371472 [5:37:16<1:45:03,  3.55it/s]
 94%|█████████▍| 349103/371472 [5:37:16<1:43:52,  3.59it/s]
 94%|█████████▍| 349104/371472 [5:37:16<1:45:07,  3.55it/s]
 94%|██���██████▍| 349105/371472 [5:37:17<1:46:31,  3.50it/s]
 94%|█████████▍| 349106/371472 [5:37:17<1:46:51,  3.49it/s]
 94%|█████████▍| 349107/371472 [5:37:17<1:49:38,  3.40it/s]
 94%|█████████▍| 349108/371472 [5:37:17<1:50:45,  3.37it/s]
 94%|█████████▍| 349109/371472 [5:37:18<1:47:14,  3.48it/s]
 94%|█████████▍| 349110/371472 [5:37:18<1:44:31,  3.57it/s]
 94%|█████████▍| 349111/371472 [5:37:18<1:49:12,  3.41it/s]
 94%|█████████▍| 349112/371472 [5:37:19<1:51:01,  3.36it/s]
 94%|█████████▍| 349113/371472 [5:37:19<1:50:18,  3.38it/s]
 94%|█████████▍| 349114/371472 [5:37:19<1:46:36,  3.50it/s]
 94%|█████████▍| 349115/371472 [5:37:19<1:46:36,  3.50it/s]
 94%|█████████▍| 349116/371472 [5:37:20<1:46:43,  3.49it/s]
 94%|█████████▍| 349117/371472 [5:37:20<1:43:52,  3.59it/s]
 94%|█████████▍| 349118/371472 [5:37:20<2:02:33,  3.04it/s]
 94%|█████████▍| 349119/371472 [5:37:21<1:55:54,  3.21it/s]
 94%|█████████▍| 349120/371472 [5:37:21<1:51:52,  3.33it/s]
                                                           
{'loss': 2.6239, 'learning_rate': 1.5418345579521211e-07, 'epoch': 15.04}
+
 94%|█████████▍| 349120/371472 [5:37:21<1:51:52,  3.33it/s]
 94%|█████████▍| 349121/371472 [5:37:21<1:52:29,  3.31it/s]
 94%|█████████▍| 349122/371472 [5:37:22<1:46:37,  3.49it/s]
 94%|█████████▍| 349123/371472 [5:37:22<1:44:07,  3.58it/s]
 94%|█████████▍| 349124/371472 [5:37:22<1:47:43,  3.46it/s]
 94%|█████████▍| 349125/371472 [5:37:22<1:46:02,  3.51it/s]
 94%|█████████▍| 349126/371472 [5:37:23<1:44:32,  3.56it/s]
 94%|█████████▍| 349127/371472 [5:37:23<1:50:52,  3.36it/s]
 94%|█████████▍| 349128/371472 [5:37:23<1:48:08,  3.44it/s]
 94%|█████████▍| 349129/371472 [5:37:24<1:45:59,  3.51it/s]
 94%|█████████▍| 349130/371472 [5:37:24<1:46:19,  3.50it/s]
 94%|█████████▍| 349131/371472 [5:37:24<1:45:25,  3.53it/s]
 94%|█████████▍| 349132/371472 [5:37:24<1:45:48,  3.52it/s]
 94%|█████████▍| 349133/371472 [5:37:25<1:46:14,  3.50it/s]
 94%|█████████▍| 349134/371472 [5:37:25<1:46:51,  3.48it/s]
 94%|█████████▍| 349135/371472 [5:37:25<1:52:42,  3.30it/s]
 94%|█████████▍| 349136/371472 [5:37:26<1:46:29,  3.50it/s]
 94%|█████████▍| 349137/371472 [5:37:26<1:44:15,  3.57it/s]
 94%|█████████▍| 349138/371472 [5:37:26<1:44:06,  3.58it/s]
 94%|█████████▍| 349139/371472 [5:37:26<1:42:49,  3.62it/s]
 94%|█████████▍| 349140/371472 [5:37:27<1:39:55,  3.72it/s]
                                                           
{'loss': 2.6629, 'learning_rate': 1.5413497381973327e-07, 'epoch': 15.04}
+
 94%|█████████▍| 349140/371472 [5:37:27<1:39:55,  3.72it/s]
 94%|█████████▍| 349141/371472 [5:37:27<1:39:22,  3.75it/s]
 94%|█████████▍| 349142/371472 [5:37:27<1:40:04,  3.72it/s]
 94%|█████████▍| 349143/371472 [5:37:27<1:41:24,  3.67it/s]
 94%|█████████▍| 349144/371472 [5:37:28<1:42:48,  3.62it/s]
 94%|█████████▍| 349145/371472 [5:37:28<1:40:16,  3.71it/s]
 94%|█████████▍| 349146/371472 [5:37:28<1:38:56,  3.76it/s]
 94%|█████████▍| 349147/371472 [5:37:28<1:39:08,  3.75it/s]
 94%|█████████▍| 349148/371472 [5:37:29<1:36:57,  3.84it/s]
 94%|█████████▍| 349149/371472 [5:37:29<1:55:37,  3.22it/s]
 94%|█████████▍| 349150/371472 [5:37:29<1:52:43,  3.30it/s]
 94%|█████████▍| 349151/371472 [5:37:30<1:48:37,  3.42it/s]
 94%|█████████▍| 349152/371472 [5:37:30<1:44:12,  3.57it/s]
 94%|█████████▍| 349153/371472 [5:37:30<1:42:04,  3.64it/s]
 94%|█████████▍| 349154/371472 [5:37:30<1:41:54,  3.65it/s]
 94%|█████████▍| 349155/371472 [5:37:31<1:43:15,  3.60it/s]
 94%|█████████▍| 349156/371472 [5:37:31<1:46:29,  3.49it/s]
 94%|█████████▍| 349157/371472 [5:37:31<1:50:21,  3.37it/s]
 94%|█████████▍| 349158/371472 [5:37:32<1:51:18,  3.34it/s]
 94%|█████████▍| 349159/371472 [5:37:32<1:47:39,  3.45it/s]
 94%|█████████▍| 349160/371472 [5:37:32<1:45:04,  3.54it/s]
                                                           
{'loss': 2.5854, 'learning_rate': 1.5408649184425434e-07, 'epoch': 15.04}
+
 94%|█████████▍| 349160/371472 [5:37:32<1:45:04,  3.54it/s]
 94%|█████████▍| 349161/371472 [5:37:33<1:52:03,  3.32it/s]
 94%|█████████▍| 349162/371472 [5:37:33<1:47:45,  3.45it/s]
 94%|█████████▍| 349163/371472 [5:37:33<1:53:19,  3.28it/s]
 94%|█████████▍| 349164/371472 [5:37:34<1:58:57,  3.13it/s]
 94%|█████████▍| 349165/371472 [5:37:34<1:57:10,  3.17it/s]
 94%|█████████▍| 349166/371472 [5:37:34<1:54:27,  3.25it/s]
 94%|█████████▍| 349167/371472 [5:37:34<1:48:58,  3.41it/s]
 94%|█████████▍| 349168/371472 [5:37:35<1:47:54,  3.44it/s]
 94%|█████████▍| 349169/371472 [5:37:35<1:43:14,  3.60it/s]
 94%|█████████▍| 349170/371472 [5:37:35<1:44:19,  3.56it/s]
 94%|█████████▍| 349171/371472 [5:37:35<1:43:15,  3.60it/s]
 94%|█████████▍| 349172/371472 [5:37:36<1:41:33,  3.66it/s]
 94%|█████████▍| 349173/371472 [5:37:36<1:46:27,  3.49it/s]
 94%|█████████▍| 349174/371472 [5:37:36<1:43:32,  3.59it/s]
 94%|█████████▍| 349175/371472 [5:37:37<1:41:32,  3.66it/s]
 94%|█████████▍| 349176/371472 [5:37:37<1:40:27,  3.70it/s]
 94%|█████████▍| 349177/371472 [5:37:37<1:40:11,  3.71it/s]
 94%|█████████▍| 349178/371472 [5:37:37<1:48:33,  3.42it/s]
 94%|█████████▍| 349179/371472 [5:37:38<1:53:43,  3.27it/s]
 94%|█████████▍| 349180/371472 [5:37:38<1:48:50,  3.41it/s]
                                                           
{'loss': 2.6154, 'learning_rate': 1.540380098687755e-07, 'epoch': 15.04}
+
 94%|█████████▍| 349180/371472 [5:37:38<1:48:50,  3.41it/s]
 94%|█████████▍| 349181/371472 [5:37:38<1:50:48,  3.35it/s]
 94%|█████████▍| 349182/371472 [5:37:39<1:51:53,  3.32it/s]
 94%|█████████▍| 349183/371472 [5:37:39<1:50:54,  3.35it/s]
 94%|█████████▍| 349184/371472 [5:37:39<1:53:36,  3.27it/s]
 94%|█████████▍| 349185/371472 [5:37:40<1:53:22,  3.28it/s]
 94%|█████████▍| 349186/371472 [5:37:40<1:49:58,  3.38it/s]
 94%|█████████▍| 349187/371472 [5:37:40<1:45:49,  3.51it/s]
 94%|█████████▍| 349188/371472 [5:37:40<1:42:21,  3.63it/s]
 94%|█████████▍| 349189/371472 [5:37:41<1:45:44,  3.51it/s]
 94%|█████████▍| 349190/371472 [5:37:41<1:44:11,  3.56it/s]
 94%|█████████▍| 349191/371472 [5:37:41<1:41:43,  3.65it/s]
 94%|█████████▍| 349192/371472 [5:37:42<1:41:51,  3.65it/s]
 94%|█████████▍| 349193/371472 [5:37:42<1:44:24,  3.56it/s]
 94%|█████████▍| 349194/371472 [5:37:42<1:46:28,  3.49it/s]
 94%|█████████▍| 349195/371472 [5:37:42<1:43:51,  3.57it/s]
 94%|█████████▍| 349196/371472 [5:37:43<1:41:58,  3.64it/s]
 94%|█████████▍| 349197/371472 [5:37:43<1:38:42,  3.76it/s]
 94%|█████████▍| 349198/371472 [5:37:43<1:42:01,  3.64it/s]
 94%|█████████▍| 349199/371472 [5:37:43<1:41:09,  3.67it/s]
 94%|█████████▍| 349200/371472 [5:37:44<1:44:23,  3.56it/s]
                                                           
{'loss': 2.6273, 'learning_rate': 1.5398952789329656e-07, 'epoch': 15.04}
+
 94%|█████████▍| 349200/371472 [5:37:44<1:44:23,  3.56it/s]
 94%|█████████▍| 349201/371472 [5:37:44<1:41:44,  3.65it/s]
 94%|█████████▍| 349202/371472 [5:37:44<1:39:50,  3.72it/s]
 94%|█████████▍| 349203/371472 [5:37:45<1:38:35,  3.76it/s]
 94%|█████████▍| 349204/371472 [5:37:45<1:40:41,  3.69it/s]
 94%|█████████▍| 349205/371472 [5:37:45<1:41:53,  3.64it/s]
 94%|█████████▍| 349206/371472 [5:37:45<1:47:59,  3.44it/s]
 94%|█████████▍| 349207/371472 [5:37:46<1:45:02,  3.53it/s]
 94%|█████████▍| 349208/371472 [5:37:46<1:56:16,  3.19it/s]
 94%|█████████▍| 349209/371472 [5:37:46<1:49:34,  3.39it/s]
 94%|█████████▍| 349210/371472 [5:37:47<1:46:30,  3.48it/s]
 94%|█████████▍| 349211/371472 [5:37:47<1:47:00,  3.47it/s]
 94%|█████████▍| 349212/371472 [5:37:47<1:45:21,  3.52it/s]
 94%|█████████▍| 349213/371472 [5:37:47<1:51:04,  3.34it/s]
 94%|█████████▍| 349214/371472 [5:37:48<1:49:20,  3.39it/s]
 94%|█████████▍| 349215/371472 [5:37:48<1:44:30,  3.55it/s]
 94%|█████████▍| 349216/371472 [5:37:48<1:44:09,  3.56it/s]
 94%|█████████▍| 349217/371472 [5:37:49<1:41:45,  3.64it/s]
 94%|█████████▍| 349218/371472 [5:37:49<1:37:33,  3.80it/s]
 94%|█████████▍| 349219/371472 [5:37:49<1:37:02,  3.82it/s]
 94%|█████████▍| 349220/371472 [5:37:49<1:36:47,  3.83it/s]
                                                           
{'loss': 2.6398, 'learning_rate': 1.539410459178177e-07, 'epoch': 15.04}
+
 94%|█████████▍| 349220/371472 [5:37:49<1:36:47,  3.83it/s]
 94%|█████████▍| 349221/371472 [5:37:50<1:38:54,  3.75it/s]
 94%|█████████▍| 349222/371472 [5:37:50<1:39:08,  3.74it/s]
 94%|█████████▍| 349223/371472 [5:37:50<1:44:06,  3.56it/s]
 94%|█████████▍| 349224/371472 [5:37:50<1:44:50,  3.54it/s]
 94%|█████████▍| 349225/371472 [5:37:51<1:46:51,  3.47it/s]
 94%|█████████▍| 349226/371472 [5:37:51<1:46:19,  3.49it/s]
 94%|█████████▍| 349227/371472 [5:37:51<1:46:10,  3.49it/s]
 94%|█████████▍| 349228/371472 [5:37:52<1:53:04,  3.28it/s]
 94%|█████████▍| 349229/371472 [5:37:52<1:53:32,  3.27it/s]
 94%|█████████▍| 349230/371472 [5:37:52<1:56:53,  3.17it/s]
 94%|█████████▍| 349231/371472 [5:37:53<1:50:57,  3.34it/s]
 94%|█████████▍| 349232/371472 [5:37:53<1:57:46,  3.15it/s]
 94%|█████████▍| 349233/371472 [5:37:53<1:55:32,  3.21it/s]
 94%|█████████▍| 349234/371472 [5:37:54<1:51:57,  3.31it/s]
 94%|█████████▍| 349235/371472 [5:37:54<1:48:02,  3.43it/s]
 94%|█████████▍| 349236/371472 [5:37:54<1:46:10,  3.49it/s]
 94%|█████████▍| 349237/371472 [5:37:54<1:54:27,  3.24it/s]
 94%|█████████▍| 349238/371472 [5:37:55<1:51:40,  3.32it/s]
 94%|█████████▍| 349239/371472 [5:37:55<1:49:28,  3.38it/s]
 94%|█████████▍| 349240/371472 [5:37:55<1:42:31,  3.61it/s]
                                                           
{'loss': 2.5581, 'learning_rate': 1.5389256394233876e-07, 'epoch': 15.04}
+
 94%|█████████▍| 349240/371472 [5:37:55<1:42:31,  3.61it/s]
 94%|█████████▍| 349241/371472 [5:37:56<1:46:30,  3.48it/s]
 94%|█████████▍| 349242/371472 [5:37:56<1:45:23,  3.52it/s]
 94%|█████████▍| 349243/371472 [5:37:56<1:45:25,  3.51it/s]
 94%|█████████▍| 349244/371472 [5:37:56<1:42:33,  3.61it/s]
 94%|█████████▍| 349245/371472 [5:37:57<1:44:48,  3.53it/s]
 94%|█████████▍| 349246/371472 [5:37:57<1:47:10,  3.46it/s]
 94%|█████████▍| 349247/371472 [5:37:57<1:44:43,  3.54it/s]
 94%|█████████▍| 349248/371472 [5:37:58<1:45:46,  3.50it/s]
 94%|█████████▍| 349249/371472 [5:37:58<1:48:32,  3.41it/s]
 94%|█████████▍| 349250/371472 [5:37:58<1:49:08,  3.39it/s]
 94%|█████████▍| 349251/371472 [5:37:58<1:48:32,  3.41it/s]
 94%|█████████▍| 349252/371472 [5:37:59<1:42:45,  3.60it/s]
 94%|█████████▍| 349253/371472 [5:37:59<1:43:18,  3.58it/s]
 94%|█████████▍| 349254/371472 [5:37:59<1:42:20,  3.62it/s]
 94%|█████████▍| 349255/371472 [5:37:59<1:41:10,  3.66it/s]
 94%|█████████▍| 349256/371472 [5:38:00<1:41:03,  3.66it/s]
 94%|█████████▍| 349257/371472 [5:38:00<1:42:42,  3.60it/s]
 94%|█████████▍| 349258/371472 [5:38:00<1:40:09,  3.70it/s]
 94%|█████████▍| 349259/371472 [5:38:01<1:40:02,  3.70it/s]
 94%|█████████▍| 349260/371472 [5:38:01<1:40:04,  3.70it/s]
                                                           
{'loss': 2.4779, 'learning_rate': 1.5384408196685993e-07, 'epoch': 15.04}
+
 94%|█████████▍| 349260/371472 [5:38:01<1:40:04,  3.70it/s]
 94%|█████████▍| 349261/371472 [5:38:01<1:43:33,  3.57it/s]
 94%|█████████▍| 349262/371472 [5:38:01<1:40:52,  3.67it/s]
 94%|█████████▍| 349263/371472 [5:38:02<1:39:59,  3.70it/s]
 94%|█████████▍| 349264/371472 [5:38:02<1:38:49,  3.75it/s]
 94%|█████████▍| 349265/371472 [5:38:02<1:39:05,  3.74it/s]
 94%|█████████▍| 349266/371472 [5:38:02<1:37:17,  3.80it/s]
 94%|█████████▍| 349267/371472 [5:38:03<1:37:38,  3.79it/s]
 94%|█████████▍| 349268/371472 [5:38:03<1:37:22,  3.80it/s]
 94%|█████████▍| 349269/371472 [5:38:03<1:51:35,  3.32it/s]
 94%|█████████▍| 349270/371472 [5:38:04<1:53:16,  3.27it/s]
 94%|█████████▍| 349271/371472 [5:38:04<1:57:25,  3.15it/s]
 94%|█████████▍| 349272/371472 [5:38:04<1:52:20,  3.29it/s]
 94%|█████████▍| 349273/371472 [5:38:05<1:49:43,  3.37it/s]
 94%|█████████▍| 349274/371472 [5:38:05<1:45:29,  3.51it/s]
 94%|█████████▍| 349275/371472 [5:38:05<1:43:09,  3.59it/s]
 94%|█████████▍| 349276/371472 [5:38:05<1:45:43,  3.50it/s]
 94%|█████████▍| 349277/371472 [5:38:06<1:41:49,  3.63it/s]
 94%|█████████▍| 349278/371472 [5:38:06<1:45:46,  3.50it/s]
 94%|█████████▍| 349279/371472 [5:38:06<1:45:06,  3.52it/s]
 94%|█████████▍| 349280/371472 [5:38:07<1:44:48,  3.53it/s]
                                                           
{'loss': 2.6787, 'learning_rate': 1.5379559999138098e-07, 'epoch': 15.04}
+
 94%|█████████▍| 349280/371472 [5:38:07<1:44:48,  3.53it/s]
 94%|█████████▍| 349281/371472 [5:38:07<1:44:46,  3.53it/s]
 94%|█████████▍| 349282/371472 [5:38:07<1:43:09,  3.59it/s]
 94%|█████████▍| 349283/371472 [5:38:07<1:44:28,  3.54it/s]
 94%|█████████▍| 349284/371472 [5:38:08<1:48:39,  3.40it/s]
 94%|█████████▍| 349285/371472 [5:38:08<1:47:15,  3.45it/s]
 94%|█████████▍| 349286/371472 [5:38:08<1:43:21,  3.58it/s]
 94%|█████████▍| 349287/371472 [5:38:09<1:54:22,  3.23it/s]
 94%|█████████▍| 349288/371472 [5:38:09<1:48:20,  3.41it/s]
 94%|█████████▍| 349289/371472 [5:38:09<2:00:02,  3.08it/s]
 94%|█████████▍| 349290/371472 [5:38:10<1:59:26,  3.10it/s]
 94%|█████████▍| 349291/371472 [5:38:10<1:54:28,  3.23it/s]
 94%|█████████▍| 349292/371472 [5:38:10<1:49:16,  3.38it/s]
 94%|█████████▍| 349293/371472 [5:38:10<1:44:33,  3.54it/s]
 94%|█████████▍| 349294/371472 [5:38:11<1:52:49,  3.28it/s]
 94%|█████████▍| 349295/371472 [5:38:11<1:45:33,  3.50it/s]
 94%|█████████▍| 349296/371472 [5:38:11<1:47:09,  3.45it/s]
 94%|█████████▍| 349297/371472 [5:38:12<1:43:34,  3.57it/s]
 94%|█████████▍| 349298/371472 [5:38:12<1:48:44,  3.40it/s]
 94%|█████████▍| 349299/371472 [5:38:12<1:48:54,  3.39it/s]
 94%|█████████▍| 349300/371472 [5:38:12<1:47:20,  3.44it/s]
                                                           
{'loss': 2.7193, 'learning_rate': 1.5374711801590202e-07, 'epoch': 15.05}
+
 94%|█████████▍| 349300/371472 [5:38:12<1:47:20,  3.44it/s]
 94%|█████████▍| 349301/371472 [5:38:13<1:44:01,  3.55it/s]
 94%|█████████▍| 349302/371472 [5:38:13<1:46:41,  3.46it/s]
 94%|█████████▍| 349303/371472 [5:38:13<1:49:39,  3.37it/s]
 94%|█████████▍| 349304/371472 [5:38:14<1:47:16,  3.44it/s]
 94%|█████████▍| 349305/371472 [5:38:14<1:53:37,  3.25it/s]
 94%|█████████▍| 349306/371472 [5:38:14<1:47:56,  3.42it/s]
 94%|█████████▍| 349307/371472 [5:38:14<1:45:06,  3.51it/s]
 94%|█████████▍| 349308/371472 [5:38:15<1:50:33,  3.34it/s]
 94%|█████████▍| 349309/371472 [5:38:15<1:47:00,  3.45it/s]
 94%|█████████▍| 349310/371472 [5:38:15<1:43:55,  3.55it/s]
 94%|█████████▍| 349311/371472 [5:38:16<1:46:56,  3.45it/s]
 94%|█████████▍| 349312/371472 [5:38:16<1:43:07,  3.58it/s]
 94%|█████████▍| 349313/371472 [5:38:16<1:44:56,  3.52it/s]
 94%|█████████▍| 349314/371472 [5:38:16<1:41:19,  3.64it/s]
 94%|█████████▍| 349315/371472 [5:38:17<1:39:48,  3.70it/s]
 94%|█████████▍| 349316/371472 [5:38:17<1:40:36,  3.67it/s]
 94%|█████████▍| 349317/371472 [5:38:17<1:41:24,  3.64it/s]
 94%|█████████▍| 349318/371472 [5:38:18<1:45:16,  3.51it/s]
 94%|█████████▍| 349319/371472 [5:38:18<1:45:55,  3.49it/s]
 94%|█████████▍| 349320/371472 [5:38:18<1:46:05,  3.48it/s]
                                                           
{'loss': 2.6037, 'learning_rate': 1.536986360404232e-07, 'epoch': 15.05}
+
 94%|█████████▍| 349320/371472 [5:38:18<1:46:05,  3.48it/s]
 94%|█████████▍| 349321/371472 [5:38:18<1:48:02,  3.42it/s]
 94%|█████████▍| 349322/371472 [5:38:19<1:44:01,  3.55it/s]
 94%|█████████▍| 349323/371472 [5:38:19<1:44:22,  3.54it/s]
 94%|█████████▍| 349324/371472 [5:38:19<1:46:49,  3.46it/s]
 94%|█████████▍| 349325/371472 [5:38:20<1:46:11,  3.48it/s]
 94%|█████████▍| 349326/371472 [5:38:20<1:45:46,  3.49it/s]
 94%|█████████▍| 349327/371472 [5:38:20<1:43:43,  3.56it/s]
 94%|█████████▍| 349328/371472 [5:38:20<1:45:10,  3.51it/s]
 94%|█████████▍| 349329/371472 [5:38:21<1:43:17,  3.57it/s]
 94%|█████████▍| 349330/371472 [5:38:21<1:56:19,  3.17it/s]
 94%|█████████▍| 349331/371472 [5:38:21<1:49:43,  3.36it/s]
 94%|█████████▍| 349332/371472 [5:38:22<1:48:34,  3.40it/s]
 94%|█████████▍| 349333/371472 [5:38:22<1:48:02,  3.42it/s]
 94%|█████████▍| 349334/371472 [5:38:22<1:46:57,  3.45it/s]
 94%|█████████▍| 349335/371472 [5:38:22<1:49:25,  3.37it/s]
 94%|█████████▍| 349336/371472 [5:38:23<1:47:33,  3.43it/s]
 94%|█████████▍| 349337/371472 [5:38:23<1:51:14,  3.32it/s]
 94%|█████████▍| 349338/371472 [5:38:23<1:52:40,  3.27it/s]
 94%|█████████▍| 349339/371472 [5:38:24<1:56:10,  3.18it/s]
 94%|█████████▍| 349340/371472 [5:38:24<1:51:23,  3.31it/s]
                                                           
{'loss': 2.6163, 'learning_rate': 1.5365015406494425e-07, 'epoch': 15.05}
+
 94%|█████████▍| 349340/371472 [5:38:24<1:51:23,  3.31it/s]
 94%|█████████▍| 349341/371472 [5:38:24<1:54:00,  3.24it/s]
 94%|█████████▍| 349342/371472 [5:38:25<1:50:06,  3.35it/s]
 94%|█████████▍| 349343/371472 [5:38:25<1:49:12,  3.38it/s]
 94%|█████████▍| 349344/371472 [5:38:25<1:47:24,  3.43it/s]
 94%|█████████▍| 349345/371472 [5:38:25<1:45:57,  3.48it/s]
 94%|█████████▍| 349346/371472 [5:38:26<1:48:14,  3.41it/s]
 94%|█████████▍| 349347/371472 [5:38:26<1:46:16,  3.47it/s]
 94%|█████████▍| 349348/371472 [5:38:26<1:46:58,  3.45it/s]
 94%|█████████▍| 349349/371472 [5:38:27<2:00:18,  3.06it/s]
 94%|█████████▍| 349350/371472 [5:38:27<1:56:19,  3.17it/s]
 94%|█████████▍| 349351/371472 [5:38:27<1:50:24,  3.34it/s]
 94%|█████████▍| 349352/371472 [5:38:28<1:46:14,  3.47it/s]
 94%|█████████▍| 349353/371472 [5:38:28<1:43:56,  3.55it/s]
 94%|█████████▍| 349354/371472 [5:38:28<1:44:06,  3.54it/s]
 94%|█████████▍| 349355/371472 [5:38:28<1:45:18,  3.50it/s]
 94%|█████████▍| 349356/371472 [5:38:29<1:42:35,  3.59it/s]
 94%|█████████▍| 349357/371472 [5:38:29<1:49:05,  3.38it/s]
 94%|█████████▍| 349358/371472 [5:38:29<1:46:24,  3.46it/s]
 94%|█████████▍| 349359/371472 [5:38:30<1:42:06,  3.61it/s]
 94%|█████████▍| 349360/371472 [5:38:30<1:43:50,  3.55it/s]
                                                           
{'loss': 2.4882, 'learning_rate': 1.536016720894654e-07, 'epoch': 15.05}
+
 94%|█████████▍| 349360/371472 [5:38:30<1:43:50,  3.55it/s]
 94%|█████████▍| 349361/371472 [5:38:30<1:45:57,  3.48it/s]
 94%|█████████▍| 349362/371472 [5:38:30<1:46:47,  3.45it/s]
 94%|█████████▍| 349363/371472 [5:38:31<1:43:18,  3.57it/s]
 94%|█████████▍| 349364/371472 [5:38:31<1:45:50,  3.48it/s]
 94%|█████████▍| 349365/371472 [5:38:31<1:44:06,  3.54it/s]
 94%|█████████▍| 349366/371472 [5:38:32<1:45:39,  3.49it/s]
 94%|█████████▍| 349367/371472 [5:38:32<1:43:40,  3.55it/s]
 94%|█████████▍| 349368/371472 [5:38:32<1:40:36,  3.66it/s]
 94%|█████████▍| 349369/371472 [5:38:32<1:39:56,  3.69it/s]
 94%|█████████▍| 349370/371472 [5:38:33<1:42:07,  3.61it/s]
 94%|█████████▍| 349371/371472 [5:38:33<1:39:50,  3.69it/s]
 94%|█████████▍| 349372/371472 [5:38:33<1:41:42,  3.62it/s]
 94%|█████████▍| 349373/371472 [5:38:33<1:42:17,  3.60it/s]
 94%|█████████▍| 349374/371472 [5:38:34<1:50:47,  3.32it/s]
 94%|█████████▍| 349375/371472 [5:38:34<1:50:13,  3.34it/s]
 94%|█████████▍| 349376/371472 [5:38:34<1:47:59,  3.41it/s]
 94%|█████████▍| 349377/371472 [5:38:35<1:43:39,  3.55it/s]
 94%|█████████▍| 349378/371472 [5:38:35<1:55:20,  3.19it/s]
 94%|█████████▍| 349379/371472 [5:38:35<1:50:14,  3.34it/s]
 94%|█████████▍| 349380/371472 [5:38:36<1:51:13,  3.31it/s]
                                                           
{'loss': 2.4446, 'learning_rate': 1.5355319011398647e-07, 'epoch': 15.05}
+
 94%|█████████▍| 349380/371472 [5:38:36<1:51:13,  3.31it/s]
 94%|█████████▍| 349381/371472 [5:38:36<1:50:21,  3.34it/s]
 94%|█████████▍| 349382/371472 [5:38:36<1:56:18,  3.17it/s]
 94%|█████████▍| 349383/371472 [5:38:37<1:49:44,  3.35it/s]
 94%|█████████▍| 349384/371472 [5:38:37<1:46:44,  3.45it/s]
 94%|█████████▍| 349385/371472 [5:38:37<1:53:22,  3.25it/s]
 94%|█████████▍| 349386/371472 [5:38:37<1:53:29,  3.24it/s]
 94%|█████████▍| 349387/371472 [5:38:38<1:48:35,  3.39it/s]
 94%|█████████▍| 349388/371472 [5:38:38<1:51:17,  3.31it/s]
 94%|█████████▍| 349389/371472 [5:38:38<1:50:20,  3.34it/s]
 94%|█████████▍| 349390/371472 [5:38:39<1:51:00,  3.32it/s]
 94%|█████████▍| 349391/371472 [5:38:39<2:01:04,  3.04it/s]
 94%|█████████▍| 349392/371472 [5:38:39<1:53:32,  3.24it/s]
 94%|█████████▍| 349393/371472 [5:38:40<1:57:41,  3.13it/s]
 94%|█████████▍| 349394/371472 [5:38:40<1:55:49,  3.18it/s]
 94%|█████████▍| 349395/371472 [5:38:40<1:51:54,  3.29it/s]
 94%|█████████▍| 349396/371472 [5:38:40<1:49:25,  3.36it/s]
 94%|█████████▍| 349397/371472 [5:38:41<1:48:10,  3.40it/s]
 94%|█████████▍| 349398/371472 [5:38:41<1:46:38,  3.45it/s]
 94%|█████████▍| 349399/371472 [5:38:41<1:49:22,  3.36it/s]
 94%|█████████▍| 349400/371472 [5:38:42<1:48:43,  3.38it/s]
                                                           
{'loss': 2.5655, 'learning_rate': 1.535047081385076e-07, 'epoch': 15.05}
+
 94%|█████████▍| 349400/371472 [5:38:42<1:48:43,  3.38it/s]
 94%|█████████▍| 349401/371472 [5:38:42<1:46:58,  3.44it/s]
 94%|█████████▍| 349402/371472 [5:38:42<1:52:55,  3.26it/s]
 94%|█████████▍| 349403/371472 [5:38:43<1:49:25,  3.36it/s]
 94%|█████████▍| 349404/371472 [5:38:43<1:48:09,  3.40it/s]
 94%|█████████▍| 349405/371472 [5:38:43<1:51:12,  3.31it/s]
 94%|█████████▍| 349406/371472 [5:38:43<1:46:36,  3.45it/s]
 94%|█████████▍| 349407/371472 [5:38:44<1:47:24,  3.42it/s]
 94%|█████████▍| 349408/371472 [5:38:44<1:44:01,  3.54it/s]
 94%|█████████▍| 349409/371472 [5:38:44<1:45:48,  3.48it/s]
 94%|█████████▍| 349410/371472 [5:38:45<1:40:54,  3.64it/s]
 94%|█████████▍| 349411/371472 [5:38:45<1:45:00,  3.50it/s]
 94%|█████████▍| 349412/371472 [5:38:45<1:45:04,  3.50it/s]
 94%|█████████▍| 349413/371472 [5:38:45<1:43:05,  3.57it/s]
 94%|█████████▍| 349414/371472 [5:38:46<1:48:51,  3.38it/s]
 94%|█████████▍| 349415/371472 [5:38:46<1:48:19,  3.39it/s]
 94%|█████████▍| 349416/371472 [5:38:46<1:48:31,  3.39it/s]
 94%|█████████▍| 349417/371472 [5:38:47<1:52:25,  3.27it/s]
 94%|█████████▍| 349418/371472 [5:38:47<1:49:57,  3.34it/s]
 94%|█████████▍| 349419/371472 [5:38:47<1:49:00,  3.37it/s]
 94%|█████████▍| 349420/371472 [5:38:47<1:46:05,  3.46it/s]
                                                           
{'loss': 2.5647, 'learning_rate': 1.5345622616302866e-07, 'epoch': 15.05}
+
 94%|█████████▍| 349420/371472 [5:38:47<1:46:05,  3.46it/s]
 94%|█████████▍| 349421/371472 [5:38:48<1:53:38,  3.23it/s]
 94%|█████████▍| 349422/371472 [5:38:48<1:49:52,  3.34it/s]
 94%|█████████▍| 349423/371472 [5:38:48<1:48:33,  3.39it/s]
 94%|█████████▍| 349424/371472 [5:38:49<1:45:18,  3.49it/s]
 94%|█████████▍| 349425/371472 [5:38:49<1:43:55,  3.54it/s]
 94%|█████████▍| 349426/371472 [5:38:49<1:41:17,  3.63it/s]
 94%|█████████▍| 349427/371472 [5:38:49<1:42:22,  3.59it/s]
 94%|█████████▍| 349428/371472 [5:38:50<1:39:13,  3.70it/s]
 94%|█████████▍| 349429/371472 [5:38:50<1:38:27,  3.73it/s]
 94%|█████████▍| 349430/371472 [5:38:50<1:39:03,  3.71it/s]
 94%|█████████▍| 349431/371472 [5:38:51<1:40:52,  3.64it/s]
 94%|█████████▍| 349432/371472 [5:38:51<1:42:00,  3.60it/s]
 94%|█████████▍| 349433/371472 [5:38:51<1:41:17,  3.63it/s]
 94%|█████████▍| 349434/371472 [5:38:51<1:43:57,  3.53it/s]
 94%|█████████▍| 349435/371472 [5:38:52<1:43:18,  3.56it/s]
 94%|█████████▍| 349436/371472 [5:38:52<1:41:05,  3.63it/s]
 94%|█████████▍| 349437/371472 [5:38:52<1:45:57,  3.47it/s]
 94%|█████████▍| 349438/371472 [5:38:53<1:49:08,  3.36it/s]
 94%|█████████▍| 349439/371472 [5:38:53<1:47:59,  3.40it/s]
 94%|█████████▍| 349440/371472 [5:38:53<1:44:12,  3.52it/s]
                                                           
{'loss': 2.5786, 'learning_rate': 1.5340774418754984e-07, 'epoch': 15.05}
+
 94%|█████████▍| 349440/371472 [5:38:53<1:44:12,  3.52it/s]
 94%|█████████▍| 349441/371472 [5:38:53<1:40:35,  3.65it/s]
 94%|█████████▍| 349442/371472 [5:38:54<1:46:34,  3.45it/s]
 94%|██████��██▍| 349443/371472 [5:38:54<1:42:57,  3.57it/s]
 94%|█████████▍| 349444/371472 [5:38:54<1:43:46,  3.54it/s]
 94%|█████████▍| 349445/371472 [5:38:55<1:41:31,  3.62it/s]
 94%|█████████▍| 349446/371472 [5:38:55<1:46:20,  3.45it/s]
 94%|█████████▍| 349447/371472 [5:38:55<1:44:03,  3.53it/s]
 94%|█████████▍| 349448/371472 [5:38:55<1:41:28,  3.62it/s]
 94%|█████████▍| 349449/371472 [5:38:56<1:43:33,  3.54it/s]
 94%|█████████▍| 349450/371472 [5:38:56<1:50:20,  3.33it/s]
 94%|█████████▍| 349451/371472 [5:38:56<1:52:46,  3.25it/s]
 94%|█████████▍| 349452/371472 [5:38:57<1:49:25,  3.35it/s]
 94%|█████████▍| 349453/371472 [5:38:57<1:45:45,  3.47it/s]
 94%|█████████▍| 349454/371472 [5:38:57<1:47:00,  3.43it/s]
 94%|█████████▍| 349455/371472 [5:38:57<1:47:06,  3.43it/s]
 94%|█████████▍| 349456/371472 [5:38:58<1:50:32,  3.32it/s]
 94%|█████████▍| 349457/371472 [5:38:58<1:47:49,  3.40it/s]
 94%|█████████▍| 349458/371472 [5:38:58<1:44:28,  3.51it/s]
 94%|█████████▍| 349459/371472 [5:38:59<1:47:16,  3.42it/s]
 94%|█████████▍| 349460/371472 [5:38:59<1:45:28,  3.48it/s]
                                                           
{'loss': 2.5515, 'learning_rate': 1.5335926221207089e-07, 'epoch': 15.05}
+
 94%|█████████▍| 349460/371472 [5:38:59<1:45:28,  3.48it/s]
 94%|█████████▍| 349461/371472 [5:38:59<1:49:09,  3.36it/s]
 94%|█████████▍| 349462/371472 [5:39:00<1:47:24,  3.42it/s]
 94%|█████████▍| 349463/371472 [5:39:00<1:57:57,  3.11it/s]
 94%|█████████▍| 349464/371472 [5:39:00<1:51:37,  3.29it/s]
 94%|█████████▍| 349465/371472 [5:39:00<1:53:03,  3.24it/s]
 94%|█████████▍| 349466/371472 [5:39:01<1:51:25,  3.29it/s]
 94%|█████████▍| 349467/371472 [5:39:01<1:53:18,  3.24it/s]
 94%|█████████▍| 349468/371472 [5:39:01<1:55:29,  3.18it/s]
 94%|█████████▍| 349469/371472 [5:39:02<1:48:17,  3.39it/s]
 94%|█████████▍| 349470/371472 [5:39:02<1:48:02,  3.39it/s]
 94%|█████████▍| 349471/371472 [5:39:02<1:42:25,  3.58it/s]
 94%|█████████▍| 349472/371472 [5:39:03<1:45:22,  3.48it/s]
 94%|█████████▍| 349473/371472 [5:39:03<1:47:41,  3.40it/s]
 94%|█████████▍| 349474/371472 [5:39:03<1:44:07,  3.52it/s]
 94%|█████████▍| 349475/371472 [5:39:03<1:48:26,  3.38it/s]
 94%|█████████▍| 349476/371472 [5:39:04<1:42:06,  3.59it/s]
 94%|█████████▍| 349477/371472 [5:39:04<1:53:58,  3.22it/s]
 94%|█████████▍| 349478/371472 [5:39:04<1:49:02,  3.36it/s]
 94%|█████████▍| 349479/371472 [5:39:05<1:44:45,  3.50it/s]
 94%|█████████▍| 349480/371472 [5:39:05<1:44:22,  3.51it/s]
                                                           
{'loss': 2.6034, 'learning_rate': 1.5331078023659204e-07, 'epoch': 15.05}
+
 94%|█████████▍| 349480/371472 [5:39:05<1:44:22,  3.51it/s]
 94%|█████████▍| 349481/371472 [5:39:05<1:42:13,  3.59it/s]
 94%|█████████▍| 349482/371472 [5:39:05<1:47:29,  3.41it/s]
 94%|█████████▍| 349483/371472 [5:39:06<1:46:32,  3.44it/s]
 94%|█████████▍| 349484/371472 [5:39:06<1:48:29,  3.38it/s]
 94%|█████████▍| 349485/371472 [5:39:06<1:53:42,  3.22it/s]
 94%|█████████▍| 349486/371472 [5:39:07<1:49:37,  3.34it/s]
 94%|█████████▍| 349487/371472 [5:39:07<1:47:36,  3.41it/s]
 94%|█████████▍| 349488/371472 [5:39:07<1:46:51,  3.43it/s]
 94%|█████████▍| 349489/371472 [5:39:08<1:53:26,  3.23it/s]
 94%|█████████▍| 349490/371472 [5:39:08<1:51:12,  3.29it/s]
 94%|█████████▍| 349491/371472 [5:39:08<1:45:06,  3.49it/s]
 94%|█████████▍| 349492/371472 [5:39:08<1:48:24,  3.38it/s]
 94%|█████████▍| 349493/371472 [5:39:09<1:49:11,  3.35it/s]
 94%|█████████▍| 349494/371472 [5:39:09<1:55:35,  3.17it/s]
 94%|█████████▍| 349495/371472 [5:39:09<1:57:34,  3.12it/s]
 94%|█████████▍| 349496/371472 [5:39:10<1:50:15,  3.32it/s]
 94%|█████████▍| 349497/371472 [5:39:10<1:52:26,  3.26it/s]
 94%|█████████▍| 349498/371472 [5:39:10<1:52:27,  3.26it/s]
 94%|█████████▍| 349499/371472 [5:39:11<1:46:50,  3.43it/s]
 94%|█████████▍| 349500/371472 [5:39:11<1:40:32,  3.64it/s]
                                                           
{'loss': 2.7043, 'learning_rate': 1.532622982611131e-07, 'epoch': 15.05}
+
 94%|█████████▍| 349500/371472 [5:39:11<1:40:32,  3.64it/s]
 94%|█████████▍| 349501/371472 [5:39:11<1:40:23,  3.65it/s]
 94%|█████████▍| 349502/371472 [5:39:11<1:39:00,  3.70it/s]
 94%|█████████▍| 349503/371472 [5:39:12<1:35:00,  3.85it/s]
 94%|█████████▍| 349504/371472 [5:39:12<1:37:44,  3.75it/s]
 94%|█████████▍| 349505/371472 [5:39:12<1:46:21,  3.44it/s]
 94%|█████████▍| 349506/371472 [5:39:12<1:42:26,  3.57it/s]
 94%|█████████▍| 349507/371472 [5:39:13<1:42:48,  3.56it/s]
 94%|█████████▍| 349508/371472 [5:39:13<1:49:45,  3.34it/s]
 94%|█████████▍| 349509/371472 [5:39:13<1:45:10,  3.48it/s]
 94%|█████████▍| 349510/371472 [5:39:14<1:41:56,  3.59it/s]
 94%|█████████▍| 349511/371472 [5:39:14<1:46:26,  3.44it/s]
 94%|█████████▍| 349512/371472 [5:39:14<1:45:58,  3.45it/s]
 94%|█████████▍| 349513/371472 [5:39:14<1:41:47,  3.60it/s]
 94%|█████████▍| 349514/371472 [5:39:15<1:44:15,  3.51it/s]
 94%|█████████▍| 349515/371472 [5:39:15<1:48:19,  3.38it/s]
 94%|█████████▍| 349516/371472 [5:39:15<1:51:10,  3.29it/s]
 94%|█████████▍| 349517/371472 [5:39:16<1:46:19,  3.44it/s]
 94%|█████████▍| 349518/371472 [5:39:16<1:46:05,  3.45it/s]
 94%|█████████▍| 349519/371472 [5:39:16<1:44:19,  3.51it/s]
 94%|█████████▍| 349520/371472 [5:39:16<1:42:29,  3.57it/s]
                                                           
{'loss': 2.6381, 'learning_rate': 1.5321381628563426e-07, 'epoch': 15.05}
+
 94%|█████████▍| 349520/371472 [5:39:16<1:42:29,  3.57it/s]
 94%|█████████▍| 349521/371472 [5:39:17<1:45:17,  3.47it/s]
 94%|█████████▍| 349522/371472 [5:39:17<1:42:01,  3.59it/s]
 94%|█████████▍| 349523/371472 [5:39:17<1:37:12,  3.76it/s]
 94%|█████████▍| 349524/371472 [5:39:18<1:34:39,  3.86it/s]
 94%|█████████▍| 349525/371472 [5:39:18<1:34:47,  3.86it/s]
 94%|█████████▍| 349526/371472 [5:39:18<1:34:17,  3.88it/s]
 94%|█████████▍| 349527/371472 [5:39:18<1:34:50,  3.86it/s]
 94%|█████████▍| 349528/371472 [5:39:19<1:38:02,  3.73it/s]
 94%|█████████▍| 349529/371472 [5:39:19<1:38:35,  3.71it/s]
 94%|█████████▍| 349530/371472 [5:39:19<1:39:02,  3.69it/s]
 94%|█████████▍| 349531/371472 [5:39:19<1:39:35,  3.67it/s]
 94%|█████████▍| 349532/371472 [5:39:20<1:40:02,  3.65it/s]
 94%|█████████▍| 349533/371472 [5:39:20<1:40:38,  3.63it/s]
 94%|█████████▍| 349534/371472 [5:39:20<1:38:58,  3.69it/s]
 94%|█████████▍| 349535/371472 [5:39:21<1:39:35,  3.67it/s]
 94%|█████████▍| 349536/371472 [5:39:21<1:37:11,  3.76it/s]
 94%|█████████▍| 349537/371472 [5:39:21<1:36:39,  3.78it/s]
 94%|█████████▍| 349538/371472 [5:39:21<1:37:39,  3.74it/s]
 94%|█████████▍| 349539/371472 [5:39:22<1:33:16,  3.92it/s]
 94%|█████████▍| 349540/371472 [5:39:22<1:33:56,  3.89it/s]
                                                           
{'loss': 2.5839, 'learning_rate': 1.5316533431015533e-07, 'epoch': 15.06}
+
 94%|█████████▍| 349540/371472 [5:39:22<1:33:56,  3.89it/s]
 94%|█████████▍| 349541/371472 [5:39:22<1:35:36,  3.82it/s]
 94%|█████████▍| 349542/371472 [5:39:22<1:32:50,  3.94it/s]
 94%|█████████▍| 349543/371472 [5:39:23<1:39:21,  3.68it/s]
 94%|█████████▍| 349544/371472 [5:39:23<1:38:40,  3.70it/s]
 94%|█████████▍| 349545/371472 [5:39:23<1:39:48,  3.66it/s]
 94%|█████████▍| 349546/371472 [5:39:23<1:40:33,  3.63it/s]
 94%|█████████▍| 349547/371472 [5:39:24<1:37:42,  3.74it/s]
 94%|█████████▍| 349548/371472 [5:39:24<1:38:59,  3.69it/s]
 94%|█████████▍| 349549/371472 [5:39:24<1:45:31,  3.46it/s]
 94%|█████████▍| 349550/371472 [5:39:25<1:43:36,  3.53it/s]
 94%|█████████▍| 349551/371472 [5:39:25<1:40:13,  3.65it/s]
 94%|█████████▍| 349552/371472 [5:39:25<1:38:08,  3.72it/s]
 94%|█████████▍| 349553/371472 [5:39:25<1:38:30,  3.71it/s]
 94%|█████████▍| 349554/371472 [5:39:26<1:37:33,  3.74it/s]
 94%|█████████▍| 349555/371472 [5:39:26<1:36:54,  3.77it/s]
 94%|█████████▍| 349556/371472 [5:39:26<1:45:27,  3.46it/s]
 94%|█████████▍| 349557/371472 [5:39:26<1:42:35,  3.56it/s]
 94%|█████████▍| 349558/371472 [5:39:27<1:41:47,  3.59it/s]
 94%|█████████▍| 349559/371472 [5:39:27<1:41:42,  3.59it/s]
 94%|█████████▍| 349560/371472 [5:39:27<1:39:50,  3.66it/s]
                                                           
{'loss': 2.5019, 'learning_rate': 1.5311685233467648e-07, 'epoch': 15.06}
+
 94%|█████████▍| 349560/371472 [5:39:27<1:39:50,  3.66it/s]
 94%|█████████▍| 349561/371472 [5:39:28<1:39:42,  3.66it/s]
 94%|█████████▍| 349562/371472 [5:39:28<1:36:46,  3.77it/s]
 94%|█████████▍| 349563/371472 [5:39:28<1:39:10,  3.68it/s]
 94%|█████████▍| 349564/371472 [5:39:28<1:38:05,  3.72it/s]
 94%|█████████▍| 349565/371472 [5:39:29<1:41:39,  3.59it/s]
 94%|█████████▍| 349566/371472 [5:39:29<1:38:45,  3.70it/s]
 94%|█████████▍| 349567/371472 [5:39:29<1:35:58,  3.80it/s]
 94%|█████████▍| 349568/371472 [5:39:29<1:42:49,  3.55it/s]
 94%|█████████▍| 349569/371472 [5:39:30<1:43:46,  3.52it/s]
 94%|█████████▍| 349570/371472 [5:39:30<1:48:25,  3.37it/s]
 94%|█████████▍| 349571/371472 [5:39:30<1:52:24,  3.25it/s]
 94%|█████████▍| 349572/371472 [5:39:31<1:48:48,  3.35it/s]
 94%|█████████▍| 349573/371472 [5:39:31<1:52:06,  3.26it/s]
 94%|█████████▍| 349574/371472 [5:39:31<1:48:32,  3.36it/s]
 94%|█████████▍| 349575/371472 [5:39:32<1:50:55,  3.29it/s]
 94%|█████████▍| 349576/371472 [5:39:32<1:52:23,  3.25it/s]
 94%|█████████▍| 349577/371472 [5:39:32<1:47:35,  3.39it/s]
 94%|█████████▍| 349578/371472 [5:39:32<1:47:25,  3.40it/s]
 94%|█████████▍| 349579/371472 [5:39:33<1:50:20,  3.31it/s]
 94%|█████████▍| 349580/371472 [5:39:33<1:47:18,  3.40it/s]
                                                           
{'loss': 2.5579, 'learning_rate': 1.5306837035919755e-07, 'epoch': 15.06}
+
 94%|█████████▍| 349580/371472 [5:39:33<1:47:18,  3.40it/s]
 94%|█████████▍| 349581/371472 [5:39:33<1:49:38,  3.33it/s]
 94%|█████████▍| 349582/371472 [5:39:34<1:45:33,  3.46it/s]
 94%|█████████▍| 349583/371472 [5:39:34<1:43:13,  3.53it/s]
 94%|█████████▍| 349584/371472 [5:39:34<1:45:51,  3.45it/s]
 94%|█████████▍| 349585/371472 [5:39:35<1:58:17,  3.08it/s]
 94%|█████████▍| 349586/371472 [5:39:35<1:50:23,  3.30it/s]
 94%|█████████▍| 349587/371472 [5:39:35<1:46:31,  3.42it/s]
 94%|█████████▍| 349588/371472 [5:39:35<1:46:48,  3.42it/s]
 94%|█████████▍| 349589/371472 [5:39:36<1:45:24,  3.46it/s]
 94%|█████████▍| 349590/371472 [5:39:36<1:44:25,  3.49it/s]
 94%|█████████▍| 349591/371472 [5:39:36<1:42:34,  3.56it/s]
 94%|█████████▍| 349592/371472 [5:39:37<1:42:38,  3.55it/s]
 94%|█████████▍| 349593/371472 [5:39:37<1:44:21,  3.49it/s]
 94%|█████████▍| 349594/371472 [5:39:37<1:48:12,  3.37it/s]
 94%|█████████▍| 349595/371472 [5:39:37<1:44:49,  3.48it/s]
 94%|█████████▍| 349596/371472 [5:39:38<1:43:05,  3.54it/s]
 94%|█████████▍| 349597/371472 [5:39:38<1:41:44,  3.58it/s]
 94%|█████████▍| 349598/371472 [5:39:38<1:44:12,  3.50it/s]
 94%|█████████▍| 349599/371472 [5:39:39<1:45:36,  3.45it/s]
 94%|█████████▍| 349600/371472 [5:39:39<1:47:49,  3.38it/s]
                                                           
{'loss': 2.5047, 'learning_rate': 1.530198883837187e-07, 'epoch': 15.06}
+
 94%|█████████▍| 349600/371472 [5:39:39<1:47:49,  3.38it/s]
 94%|█████████▍| 349601/371472 [5:39:39<1:44:52,  3.48it/s]
 94%|█████████▍| 349602/371472 [5:39:39<1:44:20,  3.49it/s]
 94%|█████████▍| 349603/371472 [5:39:40<1:42:47,  3.55it/s]
 94%|█████████▍| 349604/371472 [5:39:40<1:46:25,  3.42it/s]
 94%|█████████▍| 349605/371472 [5:39:40<1:47:58,  3.38it/s]
 94%|█████████▍| 349606/371472 [5:39:41<1:45:08,  3.47it/s]
 94%|█████████▍| 349607/371472 [5:39:41<1:45:08,  3.47it/s]
 94%|█████████▍| 349608/371472 [5:39:41<1:40:47,  3.62it/s]
 94%|█████████▍| 349609/371472 [5:39:41<1:37:16,  3.75it/s]
 94%|█████████▍| 349610/371472 [5:39:42<1:37:51,  3.72it/s]
 94%|█████████▍| 349611/371472 [5:39:42<1:42:36,  3.55it/s]
 94%|█████████▍| 349612/371472 [5:39:42<1:46:42,  3.41it/s]
 94%|█████████▍| 349613/371472 [5:39:43<1:40:47,  3.61it/s]
 94%|█████████▍| 349614/371472 [5:39:43<1:43:02,  3.54it/s]
 94%|█████████▍| 349615/371472 [5:39:43<1:37:34,  3.73it/s]
 94%|█████████▍| 349616/371472 [5:39:43<1:41:33,  3.59it/s]
 94%|█████████▍| 349617/371472 [5:39:44<1:40:37,  3.62it/s]
 94%|█████████▍| 349618/371472 [5:39:44<1:43:33,  3.52it/s]
 94%|█████████▍| 349619/371472 [5:39:44<1:48:20,  3.36it/s]
 94%|█████████▍| 349620/371472 [5:39:45<1:42:28,  3.55it/s]
                                                           
{'loss': 2.6704, 'learning_rate': 1.5297140640823975e-07, 'epoch': 15.06}
+
 94%|█████████▍| 349620/371472 [5:39:45<1:42:28,  3.55it/s]
 94%|█████████▍| 349621/371472 [5:39:45<1:38:57,  3.68it/s]
 94%|█████████▍| 349622/371472 [5:39:45<1:44:31,  3.48it/s]
 94%|█████████▍| 349623/371472 [5:39:45<1:46:56,  3.41it/s]
 94%|█████████▍| 349624/371472 [5:39:46<1:44:53,  3.47it/s]
 94%|█████████▍| 349625/371472 [5:39:46<1:49:01,  3.34it/s]
 94%|█████████▍| 349626/371472 [5:39:46<1:47:16,  3.39it/s]
 94%|█████████▍| 349627/371472 [5:39:47<1:42:04,  3.57it/s]
 94%|█████████▍| 349628/371472 [5:39:47<1:49:06,  3.34it/s]
 94%|█████████▍| 349629/371472 [5:39:47<1:53:32,  3.21it/s]
 94%|█████████▍| 349630/371472 [5:39:48<1:48:47,  3.35it/s]
 94%|█████████▍| 349631/371472 [5:39:48<1:45:55,  3.44it/s]
 94%|█████████▍| 349632/371472 [5:39:48<1:44:18,  3.49it/s]
 94%|█████████▍| 349633/371472 [5:39:48<1:41:59,  3.57it/s]
 94%|█████████▍| 349634/371472 [5:39:49<1:43:48,  3.51it/s]
 94%|█████████▍| 349635/371472 [5:39:49<1:41:15,  3.59it/s]
 94%|█████████▍| 349636/371472 [5:39:49<1:40:41,  3.61it/s]
 94%|█████████▍| 349637/371472 [5:39:49<1:37:38,  3.73it/s]
 94%|█████████▍| 349638/371472 [5:39:50<1:41:16,  3.59it/s]
 94%|█████████▍| 349639/371472 [5:39:50<1:43:20,  3.52it/s]
 94%|█████████▍| 349640/371472 [5:39:50<1:39:28,  3.66it/s]
                                                           
{'loss': 2.5967, 'learning_rate': 1.5292292443276093e-07, 'epoch': 15.06}
+
 94%|█████████▍| 349640/371472 [5:39:50<1:39:28,  3.66it/s]
 94%|█████████▍| 349641/371472 [5:39:50<1:37:29,  3.73it/s]
 94%|█████████▍| 349642/371472 [5:39:51<1:42:06,  3.56it/s]
 94%|█████████▍| 349643/371472 [5:39:51<1:43:16,  3.52it/s]
 94%|█████████▍| 349644/371472 [5:39:51<1:44:55,  3.47it/s]
 94%|█████████▍| 349645/371472 [5:39:52<1:39:34,  3.65it/s]
 94%|█████████▍| 349646/371472 [5:39:52<1:37:05,  3.75it/s]
 94%|█████████▍| 349647/371472 [5:39:52<1:45:20,  3.45it/s]
 94%|█████████▍| 349648/371472 [5:39:53<1:44:20,  3.49it/s]
 94%|█████████▍| 349649/371472 [5:39:53<1:45:30,  3.45it/s]
 94%|█████████▍| 349650/371472 [5:39:53<1:43:43,  3.51it/s]
 94%|█████████▍| 349651/371472 [5:39:53<1:42:31,  3.55it/s]
 94%|█████████▍| 349652/371472 [5:39:54<1:38:25,  3.69it/s]
 94%|█████████▍| 349653/371472 [5:39:54<1:49:33,  3.32it/s]
 94%|█████████▍| 349654/371472 [5:39:54<1:47:52,  3.37it/s]
 94%|█████████▍| 349655/371472 [5:39:55<1:47:22,  3.39it/s]
 94%|█████████▍| 349656/371472 [5:39:55<1:46:26,  3.42it/s]
 94%|█████████▍| 349657/371472 [5:39:55<1:47:48,  3.37it/s]
 94%|█████████▍| 349658/371472 [5:39:55<1:47:51,  3.37it/s]
 94%|█████████▍| 349659/371472 [5:39:56<1:43:36,  3.51it/s]
 94%|█████████▍| 349660/371472 [5:39:56<1:46:39,  3.41it/s]
                                                           
{'loss': 2.622, 'learning_rate': 1.5287444245728197e-07, 'epoch': 15.06}
+
 94%|█████████▍| 349660/371472 [5:39:56<1:46:39,  3.41it/s]
 94%|█████████▍| 349661/371472 [5:39:56<1:44:37,  3.47it/s]
 94%|█████████▍| 349662/371472 [5:39:57<1:42:46,  3.54it/s]
 94%|█████████▍| 349663/371472 [5:39:57<1:39:45,  3.64it/s]
 94%|█████████▍| 349664/371472 [5:39:57<1:39:55,  3.64it/s]
 94%|█████████▍| 349665/371472 [5:39:57<1:40:40,  3.61it/s]
 94%|█████████▍| 349666/371472 [5:39:58<1:39:32,  3.65it/s]
 94%|█████████▍| 349667/371472 [5:39:58<1:37:45,  3.72it/s]
 94%|█████████▍| 349668/371472 [5:39:58<1:38:27,  3.69it/s]
 94%|█████████▍| 349669/371472 [5:39:59<1:45:31,  3.44it/s]
 94%|█████████▍| 349670/371472 [5:39:59<1:47:36,  3.38it/s]
 94%|█████████▍| 349671/371472 [5:39:59<1:42:25,  3.55it/s]
 94%|█████████▍| 349672/371472 [5:39:59<1:39:40,  3.65it/s]
 94%|█████████▍| 349673/371472 [5:40:00<1:41:45,  3.57it/s]
 94%|█████████▍| 349674/371472 [5:40:00<1:51:10,  3.27it/s]
 94%|█████████▍| 349675/371472 [5:40:00<1:45:28,  3.44it/s]
 94%|█████████▍| 349676/371472 [5:40:01<1:45:05,  3.46it/s]
 94%|█████████▍| 349677/371472 [5:40:01<1:42:37,  3.54it/s]
 94%|█████████▍| 349678/371472 [5:40:01<1:56:03,  3.13it/s]
 94%|█████████▍| 349679/371472 [5:40:01<1:49:05,  3.33it/s]
 94%|█████████▍| 349680/371472 [5:40:02<1:45:17,  3.45it/s]
                                                           
{'loss': 2.5388, 'learning_rate': 1.5282596048180312e-07, 'epoch': 15.06}
+
 94%|█████████▍| 349680/371472 [5:40:02<1:45:17,  3.45it/s]
 94%|█████████▍| 349681/371472 [5:40:02<1:44:21,  3.48it/s]
 94%|█████████▍| 349682/371472 [5:40:02<1:47:42,  3.37it/s]
 94%|█████████▍| 349683/371472 [5:40:03<1:56:54,  3.11it/s]
 94%|█████████▍| 349684/371472 [5:40:03<1:48:52,  3.34it/s]
 94%|█████████▍| 349685/371472 [5:40:03<1:43:20,  3.51it/s]
 94%|█████████▍| 349686/371472 [5:40:03<1:41:56,  3.56it/s]
 94%|█████████▍| 349687/371472 [5:40:04<1:38:36,  3.68it/s]
 94%|█████████▍| 349688/371472 [5:40:04<1:41:40,  3.57it/s]
 94%|█████████▍| 349689/371472 [5:40:04<1:43:10,  3.52it/s]
 94%|█████████▍| 349690/371472 [5:40:05<1:42:17,  3.55it/s]
 94%|█████████▍| 349691/371472 [5:40:05<1:42:28,  3.54it/s]
 94%|█████████▍| 349692/371472 [5:40:05<1:42:03,  3.56it/s]
 94%|█████████▍| 349693/371472 [5:40:05<1:46:02,  3.42it/s]
 94%|█████████▍| 349694/371472 [5:40:06<1:41:46,  3.57it/s]
 94%|█████████▍| 349695/371472 [5:40:06<1:42:10,  3.55it/s]
 94%|█████████▍| 349696/371472 [5:40:06<1:44:54,  3.46it/s]
 94%|█████████▍| 349697/371472 [5:40:07<1:49:43,  3.31it/s]
 94%|█████████▍| 349698/371472 [5:40:07<1:46:24,  3.41it/s]
 94%|█████████▍| 349699/371472 [5:40:07<1:47:21,  3.38it/s]
 94%|█████████▍| 349700/371472 [5:40:07<1:41:36,  3.57it/s]
                                                           
{'loss': 2.6993, 'learning_rate': 1.527774785063242e-07, 'epoch': 15.06}
+
 94%|█████████▍| 349700/371472 [5:40:07<1:41:36,  3.57it/s]
 94%|█████████▍| 349701/371472 [5:40:08<1:46:39,  3.40it/s]
 94%|█████████▍| 349702/371472 [5:40:08<1:42:50,  3.53it/s]
 94%|█████████▍| 349703/371472 [5:40:08<1:41:54,  3.56it/s]
 94%|█████████▍| 349704/371472 [5:40:09<1:44:50,  3.46it/s]
 94%|█████████▍| 349705/371472 [5:40:09<1:56:04,  3.13it/s]
 94%|█████████▍| 349706/371472 [5:40:09<1:57:35,  3.08it/s]
 94%|█████████▍| 349707/371472 [5:40:10<1:58:16,  3.07it/s]
 94%|█████████▍| 349708/371472 [5:40:10<1:51:57,  3.24it/s]
 94%|█████████▍| 349709/371472 [5:40:10<1:50:28,  3.28it/s]
 94%|█████████▍| 349710/371472 [5:40:11<1:44:47,  3.46it/s]
 94%|█████████▍| 349711/371472 [5:40:11<1:42:05,  3.55it/s]
 94%|█████████▍| 349712/371472 [5:40:11<1:41:46,  3.56it/s]
 94%|█████████▍| 349713/371472 [5:40:11<1:40:25,  3.61it/s]
 94%|█████████▍| 349714/371472 [5:40:12<1:41:15,  3.58it/s]
 94%|█████████▍| 349715/371472 [5:40:12<1:42:02,  3.55it/s]
 94%|█████████▍| 349716/371472 [5:40:12<1:39:19,  3.65it/s]
 94%|█████████▍| 349717/371472 [5:40:12<1:38:42,  3.67it/s]
 94%|█████████▍| 349718/371472 [5:40:13<1:45:40,  3.43it/s]
 94%|█████████▍| 349719/371472 [5:40:13<1:47:34,  3.37it/s]
 94%|█████████▍| 349720/371472 [5:40:13<1:51:42,  3.25it/s]
                                                           
{'loss': 2.538, 'learning_rate': 1.5272899653084534e-07, 'epoch': 15.06}
+
 94%|█████████▍| 349720/371472 [5:40:13<1:51:42,  3.25it/s]
 94%|█████████▍| 349721/371472 [5:40:14<1:45:23,  3.44it/s]
 94%|█████████▍| 349722/371472 [5:40:14<1:48:14,  3.35it/s]
 94%|█████████▍| 349723/371472 [5:40:14<1:50:09,  3.29it/s]
 94%|█████████▍| 349724/371472 [5:40:15<1:46:54,  3.39it/s]
 94%|█████████▍| 349725/371472 [5:40:15<1:46:02,  3.42it/s]
 94%|█████████▍| 349726/371472 [5:40:15<1:44:58,  3.45it/s]
 94%|█████████▍| 349727/371472 [5:40:15<1:40:29,  3.61it/s]
 94%|█████████▍| 349728/371472 [5:40:16<1:38:47,  3.67it/s]
 94%|█████████▍| 349729/371472 [5:40:16<1:41:47,  3.56it/s]
 94%|█████████▍| 349730/371472 [5:40:16<1:44:11,  3.48it/s]
 94%|█████████▍| 349731/371472 [5:40:16<1:39:51,  3.63it/s]
 94%|█████████▍| 349732/371472 [5:40:17<1:37:19,  3.72it/s]
 94%|█████████▍| 349733/371472 [5:40:17<1:37:24,  3.72it/s]
 94%|█████████▍| 349734/371472 [5:40:17<1:37:39,  3.71it/s]
 94%|█████████▍| 349735/371472 [5:40:18<1:40:22,  3.61it/s]
 94%|█████████▍| 349736/371472 [5:40:18<1:42:14,  3.54it/s]
 94%|█████████▍| 349737/371472 [5:40:18<1:40:19,  3.61it/s]
 94%|█████████▍| 349738/371472 [5:40:18<1:42:07,  3.55it/s]
 94%|█████████▍| 349739/371472 [5:40:19<1:41:57,  3.55it/s]
 94%|█████████▍| 349740/371472 [5:40:19<1:39:23,  3.64it/s]
                                                           
{'loss': 2.6006, 'learning_rate': 1.526805145553664e-07, 'epoch': 15.06}
+
 94%|█████████▍| 349740/371472 [5:40:19<1:39:23,  3.64it/s]
 94%|█████████▍| 349741/371472 [5:40:19<1:37:38,  3.71it/s]
 94%|█████████▍| 349742/371472 [5:40:19<1:36:49,  3.74it/s]
 94%|█████████▍| 349743/371472 [5:40:20<1:46:06,  3.41it/s]
 94%|█████████▍| 349744/371472 [5:40:20<1:50:21,  3.28it/s]
 94%|█████████▍| 349745/371472 [5:40:21<1:56:14,  3.12it/s]
 94%|█████████▍| 349746/371472 [5:40:21<1:47:11,  3.38it/s]
 94%|█████████▍| 349747/371472 [5:40:21<1:43:10,  3.51it/s]
 94%|█████████▍| 349748/371472 [5:40:21<1:45:39,  3.43it/s]
 94%|█████████▍| 349749/371472 [5:40:22<1:45:02,  3.45it/s]
 94%|█████████▍| 349750/371472 [5:40:22<1:40:27,  3.60it/s]
 94%|█████████▍| 349751/371472 [5:40:22<1:40:59,  3.58it/s]
 94%|█████████▍| 349752/371472 [5:40:23<1:55:54,  3.12it/s]
 94%|█████████▍| 349753/371472 [5:40:23<1:51:26,  3.25it/s]
 94%|█████████▍| 349754/371472 [5:40:23<1:46:59,  3.38it/s]
 94%|█████████▍| 349755/371472 [5:40:23<1:47:43,  3.36it/s]
 94%|█████████▍| 349756/371472 [5:40:24<1:42:39,  3.53it/s]
 94%|█████████▍| 349757/371472 [5:40:24<1:52:41,  3.21it/s]
 94%|█████████▍| 349758/371472 [5:40:24<1:57:19,  3.08it/s]
 94%|█████████▍| 349759/371472 [5:40:25<1:50:21,  3.28it/s]
 94%|█████████▍| 349760/371472 [5:40:25<1:50:58,  3.26it/s]
                                                           
{'loss': 2.6051, 'learning_rate': 1.5263203257988757e-07, 'epoch': 15.06}
+
 94%|█████████▍| 349760/371472 [5:40:25<1:50:58,  3.26it/s]
 94%|█████████▍| 349761/371472 [5:40:25<1:46:23,  3.40it/s]
 94%|█████████▍| 349762/371472 [5:40:26<1:45:25,  3.43it/s]
 94%|█████████▍| 349763/371472 [5:40:26<1:40:29,  3.60it/s]
 94%|█████████▍| 349764/371472 [5:40:26<1:45:55,  3.42it/s]
 94%|█████████▍| 349765/371472 [5:40:26<1:43:59,  3.48it/s]
 94%|█████████▍| 349766/371472 [5:40:27<1:40:41,  3.59it/s]
 94%|█████████▍| 349767/371472 [5:40:27<1:45:04,  3.44it/s]
 94%|█████████▍| 349768/371472 [5:40:27<1:48:12,  3.34it/s]
 94%|█████████▍| 349769/371472 [5:40:28<1:49:08,  3.31it/s]
 94%|█████████▍| 349770/371472 [5:40:28<1:43:54,  3.48it/s]
 94%|█████████▍| 349771/371472 [5:40:28<1:41:36,  3.56it/s]
 94%|█████████▍| 349772/371472 [5:40:28<1:39:54,  3.62it/s]
 94%|█████████▍| 349773/371472 [5:40:29<1:43:28,  3.49it/s]
 94%|█████████▍| 349774/371472 [5:40:29<1:41:29,  3.56it/s]
 94%|█████████▍| 349775/371472 [5:40:29<1:41:37,  3.56it/s]
 94%|█████████▍| 349776/371472 [5:40:29<1:41:32,  3.56it/s]
 94%|█████████▍| 349777/371472 [5:40:30<1:43:46,  3.48it/s]
 94%|█████████▍| 349778/371472 [5:40:30<1:42:54,  3.51it/s]
 94%|█████████▍| 349779/371472 [5:40:30<1:45:45,  3.42it/s]
 94%|█████████▍| 349780/371472 [5:40:31<1:53:28,  3.19it/s]
                                                           
{'loss': 2.5891, 'learning_rate': 1.525835506044086e-07, 'epoch': 15.07}
+
 94%|█████████▍| 349780/371472 [5:40:31<1:53:28,  3.19it/s]
 94%|█████████▍| 349781/371472 [5:40:31<1:48:29,  3.33it/s]
 94%|█████████▍| 349782/371472 [5:40:31<1:50:42,  3.27it/s]
 94%|█████████▍| 349783/371472 [5:40:32<1:49:27,  3.30it/s]
 94%|█████████▍| 349784/371472 [5:40:32<1:51:43,  3.24it/s]
 94%|█████████▍| 349785/371472 [5:40:32<1:54:14,  3.16it/s]
 94%|█████████▍| 349786/371472 [5:40:33<1:59:41,  3.02it/s]
 94%|█████████▍| 349787/371472 [5:40:33<1:49:32,  3.30it/s]
 94%|█████████▍| 349788/371472 [5:40:33<1:45:02,  3.44it/s]
 94%|█████████▍| 349789/371472 [5:40:33<1:45:34,  3.42it/s]
 94%|█████████▍| 349790/371472 [5:40:34<1:44:50,  3.45it/s]
 94%|█████████▍| 349791/371472 [5:40:34<1:44:32,  3.46it/s]
 94%|█████████▍| 349792/371472 [5:40:34<1:42:59,  3.51it/s]
 94%|█████████▍| 349793/371472 [5:40:35<1:42:19,  3.53it/s]
 94%|█████████▍| 349794/371472 [5:40:35<1:41:09,  3.57it/s]
 94%|█████████▍| 349795/371472 [5:40:35<1:40:50,  3.58it/s]
 94%|█████████▍| 349796/371472 [5:40:35<1:40:05,  3.61it/s]
 94%|█████████▍| 349797/371472 [5:40:36<1:39:03,  3.65it/s]
 94%|█████████▍| 349798/371472 [5:40:36<1:38:13,  3.68it/s]
 94%|█████████▍| 349799/371472 [5:40:36<1:39:38,  3.62it/s]
 94%|█████████▍| 349800/371472 [5:40:36<1:40:57,  3.58it/s]
                                                           
{'loss': 2.5886, 'learning_rate': 1.5253506862892976e-07, 'epoch': 15.07}
+
 94%|█████████▍| 349800/371472 [5:40:36<1:40:57,  3.58it/s]
 94%|█████████▍| 349801/371472 [5:40:37<1:37:05,  3.72it/s]
 94%|█████████▍| 349802/371472 [5:40:37<1:39:16,  3.64it/s]
 94%|█████████▍| 349803/371472 [5:40:37<1:35:51,  3.77it/s]
 94%|█████████▍| 349804/371472 [5:40:38<1:37:38,  3.70it/s]
 94%|█████████▍| 349805/371472 [5:40:38<1:41:47,  3.55it/s]
 94%|█████████▍| 349806/371472 [5:40:38<1:37:42,  3.70it/s]
 94%|█████████▍| 349807/371472 [5:40:38<1:40:09,  3.60it/s]
 94%|█████████▍| 349808/371472 [5:40:39<1:46:55,  3.38it/s]
 94%|█████████▍| 349809/371472 [5:40:39<1:47:45,  3.35it/s]
 94%|█████████▍| 349810/371472 [5:40:39<1:51:01,  3.25it/s]
 94%|█████████▍| 349811/371472 [5:40:40<1:46:04,  3.40it/s]
 94%|█████████▍| 349812/371472 [5:40:40<1:40:24,  3.60it/s]
 94%|█████████▍| 349813/371472 [5:40:40<1:41:07,  3.57it/s]
 94%|█████████▍| 349814/371472 [5:40:40<1:39:56,  3.61it/s]
 94%|█████████▍| 349815/371472 [5:40:41<1:41:41,  3.55it/s]
 94%|█████████▍| 349816/371472 [5:40:41<1:43:29,  3.49it/s]
 94%|█████████▍| 349817/371472 [5:40:41<1:46:20,  3.39it/s]
 94%|█████████▍| 349818/371472 [5:40:42<1:49:47,  3.29it/s]
 94%|█████████▍| 349819/371472 [5:40:42<1:45:54,  3.41it/s]
 94%|█████████▍| 349820/371472 [5:40:42<1:53:32,  3.18it/s]
                                                           
{'loss': 2.5356, 'learning_rate': 1.5248658665345083e-07, 'epoch': 15.07}
+
 94%|█████████▍| 349820/371472 [5:40:42<1:53:32,  3.18it/s]
 94%|█████████▍| 349821/371472 [5:40:43<1:47:37,  3.35it/s]
 94%|█████████▍| 349822/371472 [5:40:43<1:55:32,  3.12it/s]
 94%|█████████▍| 349823/371472 [5:40:43<1:52:58,  3.19it/s]
 94%|█████████▍| 349824/371472 [5:40:44<1:50:42,  3.26it/s]
 94%|█████████▍| 349825/371472 [5:40:44<1:50:28,  3.27it/s]
 94%|█████████▍| 349826/371472 [5:40:44<1:47:04,  3.37it/s]
 94%|█████████▍| 349827/371472 [5:40:44<1:41:34,  3.55it/s]
 94%|█████████▍| 349828/371472 [5:40:45<1:40:56,  3.57it/s]
 94%|█████████▍| 349829/371472 [5:40:45<1:42:42,  3.51it/s]
 94%|█████████▍| 349830/371472 [5:40:45<1:39:08,  3.64it/s]
 94%|█████████▍| 349831/371472 [5:40:45<1:35:45,  3.77it/s]
 94%|█████████▍| 349832/371472 [5:40:46<1:37:56,  3.68it/s]
 94%|█████████▍| 349833/371472 [5:40:46<1:47:10,  3.37it/s]
 94%|█████████▍| 349834/371472 [5:40:46<1:44:56,  3.44it/s]
 94%|█████████▍| 349835/371472 [5:40:47<1:40:08,  3.60it/s]
 94%|█████████▍| 349836/371472 [5:40:47<1:38:21,  3.67it/s]
 94%|█████████▍| 349837/371472 [5:40:47<1:35:52,  3.76it/s]
 94%|█████████▍| 349838/371472 [5:40:47<1:35:00,  3.80it/s]
 94%|█████████▍| 349839/371472 [5:40:48<1:31:39,  3.93it/s]
 94%|█████████▍| 349840/371472 [5:40:48<1:33:48,  3.84it/s]
                                                           
{'loss': 2.5449, 'learning_rate': 1.524381046779719e-07, 'epoch': 15.07}
+
 94%|█████████▍| 349840/371472 [5:40:48<1:33:48,  3.84it/s]
 94%|█████████▍| 349841/371472 [5:40:48<1:32:13,  3.91it/s]
 94%|█████████▍| 349842/371472 [5:40:48<1:32:19,  3.90it/s]
 94%|█████████▍| 349843/371472 [5:40:49<1:39:51,  3.61it/s]
 94%|█████████▍| 349844/371472 [5:40:49<1:36:48,  3.72it/s]
 94%|█████████▍| 349845/371472 [5:40:49<1:36:44,  3.73it/s]
 94%|█████████▍| 349846/371472 [5:40:49<1:35:58,  3.76it/s]
 94%|█████████▍| 349847/371472 [5:40:50<1:36:14,  3.75it/s]
 94%|█████████▍| 349848/371472 [5:40:50<1:43:22,  3.49it/s]
 94%|█████████▍| 349849/371472 [5:40:50<1:45:39,  3.41it/s]
 94%|█████████▍| 349850/371472 [5:40:51<1:44:44,  3.44it/s]
 94%|█████████▍| 349851/371472 [5:40:51<1:44:47,  3.44it/s]
 94%|█████████▍| 349852/371472 [5:40:51<1:45:01,  3.43it/s]
 94%|█████████▍| 349853/371472 [5:40:52<1:51:52,  3.22it/s]
 94%|█████████▍| 349854/371472 [5:40:52<1:52:16,  3.21it/s]
 94%|█████████▍| 349855/371472 [5:40:52<1:45:19,  3.42it/s]
 94%|█████████▍| 349856/371472 [5:40:52<1:41:37,  3.55it/s]
 94%|█████████▍| 349857/371472 [5:40:53<1:44:36,  3.44it/s]
 94%|█████████▍| 349858/371472 [5:40:53<1:45:09,  3.43it/s]
 94%|█████████▍| 349859/371472 [5:40:53<1:43:11,  3.49it/s]
 94%|█████████▍| 349860/371472 [5:40:54<1:44:52,  3.43it/s]
                                                           
{'loss': 2.8186, 'learning_rate': 1.5238962270249306e-07, 'epoch': 15.07}
+
 94%|█████████▍| 349860/371472 [5:40:54<1:44:52,  3.43it/s]
 94%|█████████▍| 349861/371472 [5:40:54<1:49:44,  3.28it/s]
 94%|█████████▍| 349862/371472 [5:40:54<1:46:25,  3.38it/s]
 94%|█████████▍| 349863/371472 [5:40:54<1:45:46,  3.40it/s]
 94%|█████████▍| 349864/371472 [5:40:55<1:44:44,  3.44it/s]
 94%|█████████▍| 349865/371472 [5:40:55<1:44:07,  3.46it/s]
 94%|█████████▍| 349866/371472 [5:40:55<1:42:26,  3.52it/s]
 94%|█████████▍| 349867/371472 [5:40:56<1:39:54,  3.60it/s]
 94%|█████████▍| 349868/371472 [5:40:56<1:38:52,  3.64it/s]
 94%|█████████▍| 349869/371472 [5:40:56<1:44:04,  3.46it/s]
 94%|█████████▍| 349870/371472 [5:40:56<1:45:47,  3.40it/s]
 94%|█████████▍| 349871/371472 [5:40:57<1:48:40,  3.31it/s]
 94%|█████████▍| 349872/371472 [5:40:57<1:46:10,  3.39it/s]
 94%|█████████▍| 349873/371472 [5:40:57<1:42:01,  3.53it/s]
 94%|█████████▍| 349874/371472 [5:40:58<1:38:56,  3.64it/s]
 94%|█████████▍| 349875/371472 [5:40:58<1:38:23,  3.66it/s]
 94%|█████████▍| 349876/371472 [5:40:58<1:37:29,  3.69it/s]
 94%|█████████▍| 349877/371472 [5:40:58<1:38:43,  3.65it/s]
 94%|█████████▍| 349878/371472 [5:40:59<1:38:22,  3.66it/s]
 94%|█████████▍| 349879/371472 [5:40:59<1:36:30,  3.73it/s]
 94%|█████████▍| 349880/371472 [5:40:59<1:35:27,  3.77it/s]
                                                           
{'loss': 2.6495, 'learning_rate': 1.523411407270141e-07, 'epoch': 15.07}
+
 94%|█████████▍| 349880/371472 [5:40:59<1:35:27,  3.77it/s]
 94%|█████████▍| 349881/371472 [5:41:00<1:44:52,  3.43it/s]
 94%|█████████▍| 349882/371472 [5:41:00<1:43:11,  3.49it/s]
 94%|█████████▍| 349883/371472 [5:41:00<1:47:56,  3.33it/s]
 94%|█████████▍| 349884/371472 [5:41:00<1:44:46,  3.43it/s]
 94%|█████████▍| 349885/371472 [5:41:01<1:52:13,  3.21it/s]
 94%|█████████▍| 349886/371472 [5:41:01<1:59:13,  3.02it/s]
 94%|█████████▍| 349887/371472 [5:41:01<1:53:54,  3.16it/s]
 94%|█████████▍| 349888/371472 [5:41:02<1:52:30,  3.20it/s]
 94%|█████████▍| 349889/371472 [5:41:02<2:07:30,  2.82it/s]
 94%|█████████▍| 349890/371472 [5:41:02<1:57:54,  3.05it/s]
 94%|█████████▍| 349891/371472 [5:41:03<1:50:44,  3.25it/s]
 94%|█████████▍| 349892/371472 [5:41:03<1:43:04,  3.49it/s]
 94%|█████████▍| 349893/371472 [5:41:03<1:40:57,  3.56it/s]
 94%|█████████▍| 349894/371472 [5:41:04<1:43:30,  3.47it/s]
 94%|█████████▍| 349895/371472 [5:41:04<1:51:02,  3.24it/s]
 94%|█████████▍| 349896/371472 [5:41:04<1:50:01,  3.27it/s]
 94%|█████████▍| 349897/371472 [5:41:04<1:47:38,  3.34it/s]
 94%|█████████▍| 349898/371472 [5:41:05<1:45:46,  3.40it/s]
 94%|█████████▍| 349899/371472 [5:41:05<1:43:46,  3.46it/s]
 94%|█████████▍| 349900/371472 [5:41:05<1:40:28,  3.58it/s]
                                                           
{'loss': 2.5028, 'learning_rate': 1.5229265875153528e-07, 'epoch': 15.07}
+
 94%|█████████▍| 349900/371472 [5:41:05<1:40:28,  3.58it/s]
 94%|█████████▍| 349901/371472 [5:41:06<1:45:19,  3.41it/s]
 94%|█████████▍| 349902/371472 [5:41:06<1:58:27,  3.03it/s]
 94%|█████████▍| 349903/371472 [5:41:06<1:53:13,  3.17it/s]
 94%|█████████▍| 349904/371472 [5:41:07<1:49:54,  3.27it/s]
 94%|█████████▍| 349905/371472 [5:41:07<1:46:51,  3.36it/s]
 94%|█████████▍| 349906/371472 [5:41:07<1:44:53,  3.43it/s]
 94%|█████████▍| 349907/371472 [5:41:07<1:40:23,  3.58it/s]
 94%|█████████▍| 349908/371472 [5:41:08<1:39:09,  3.62it/s]
 94%|█████████▍| 349909/371472 [5:41:08<1:39:14,  3.62it/s]
 94%|█████████▍| 349910/371472 [5:41:08<1:42:54,  3.49it/s]
 94%|█████████▍| 349911/371472 [5:41:09<1:42:55,  3.49it/s]
 94%|█████████▍| 349912/371472 [5:41:09<1:41:24,  3.54it/s]
 94%|█████████▍| 349913/371472 [5:41:09<1:46:43,  3.37it/s]
 94%|█████████▍| 349914/371472 [5:41:09<1:48:42,  3.31it/s]
 94%|█████████▍| 349915/371472 [5:41:10<1:45:21,  3.41it/s]
 94%|█████████▍| 349916/371472 [5:41:10<1:41:48,  3.53it/s]
 94%|█████████▍| 349917/371472 [5:41:10<1:40:55,  3.56it/s]
 94%|█████████▍| 349918/371472 [5:41:11<1:47:51,  3.33it/s]
 94%|█████████▍| 349919/371472 [5:41:11<1:42:30,  3.50it/s]
 94%|█████████▍| 349920/371472 [5:41:11<1:43:13,  3.48it/s]
                                                           
{'loss': 2.5592, 'learning_rate': 1.5224417677605632e-07, 'epoch': 15.07}
+
 94%|█████████▍| 349920/371472 [5:41:11<1:43:13,  3.48it/s]
 94%|█████████▍| 349921/371472 [5:41:11<1:44:51,  3.43it/s]
 94%|█████████▍| 349922/371472 [5:41:12<1:58:01,  3.04it/s]
 94%|█████████▍| 349923/371472 [5:41:12<2:02:07,  2.94it/s]
 94%|█████████▍| 349924/371472 [5:41:13<1:55:15,  3.12it/s]
 94%|█████████▍| 349925/371472 [5:41:13<1:49:39,  3.27it/s]
 94%|█████████▍| 349926/371472 [5:41:13<1:43:49,  3.46it/s]
 94%|█████████▍| 349927/371472 [5:41:13<1:41:48,  3.53it/s]
 94%|█████████▍| 349928/371472 [5:41:14<1:41:02,  3.55it/s]
 94%|█████████▍| 349929/371472 [5:41:14<1:44:35,  3.43it/s]
 94%|█████████▍| 349930/371472 [5:41:14<1:46:28,  3.37it/s]
 94%|█████████▍| 349931/371472 [5:41:14<1:41:20,  3.54it/s]
 94%|█████████▍| 349932/371472 [5:41:15<1:42:04,  3.52it/s]
 94%|█████████▍| 349933/371472 [5:41:15<1:44:24,  3.44it/s]
 94%|█████████▍| 349934/371472 [5:41:15<1:56:02,  3.09it/s]
 94%|█████████▍| 349935/371472 [5:41:16<1:52:09,  3.20it/s]
 94%|█████████▍| 349936/371472 [5:41:16<1:46:23,  3.37it/s]
 94%|█████████▍| 349937/371472 [5:41:16<1:45:04,  3.42it/s]
 94%|█████████▍| 349938/371472 [5:41:17<1:41:07,  3.55it/s]
 94%|█████████▍| 349939/371472 [5:41:17<1:45:24,  3.40it/s]
 94%|█████████▍| 349940/371472 [5:41:17<1:46:44,  3.36it/s]
                                                           
{'loss': 2.5802, 'learning_rate': 1.5219569480057748e-07, 'epoch': 15.07}
+
 94%|█████████▍| 349940/371472 [5:41:17<1:46:44,  3.36it/s]
 94%|█████████▍| 349941/371472 [5:41:17<1:44:35,  3.43it/s]
 94%|█████████▍| 349942/371472 [5:41:18<1:40:11,  3.58it/s]
 94%|█████████▍| 349943/371472 [5:41:18<1:42:09,  3.51it/s]
 94%|█████████▍| 349944/371472 [5:41:18<1:47:11,  3.35it/s]
 94%|█████████▍| 349945/371472 [5:41:19<1:43:03,  3.48it/s]
 94%|█████████▍| 349946/371472 [5:41:19<1:38:55,  3.63it/s]
 94%|█████████▍| 349947/371472 [5:41:19<1:39:49,  3.59it/s]
 94%|█████████▍| 349948/371472 [5:41:19<1:43:46,  3.46it/s]
 94%|█████████▍| 349949/371472 [5:41:20<1:43:27,  3.47it/s]
 94%|█████████▍| 349950/371472 [5:41:20<1:44:02,  3.45it/s]
 94%|█████████▍| 349951/371472 [5:41:20<1:40:02,  3.59it/s]
 94%|█████████▍| 349952/371472 [5:41:21<1:36:13,  3.73it/s]
 94%|█████████▍| 349953/371472 [5:41:21<1:46:09,  3.38it/s]
 94%|█████████▍| 349954/371472 [5:41:21<1:42:52,  3.49it/s]
 94%|█████████▍| 349955/371472 [5:41:21<1:38:17,  3.65it/s]
 94%|█████████▍| 349956/371472 [5:41:22<1:43:37,  3.46it/s]
 94%|█████████▍| 349957/371472 [5:41:22<1:41:15,  3.54it/s]
 94%|█████████▍| 349958/371472 [5:41:22<1:38:26,  3.64it/s]
 94%|█████████▍| 349959/371472 [5:41:23<1:40:36,  3.56it/s]
 94%|█████████▍| 349960/371472 [5:41:23<1:41:05,  3.55it/s]
                                                           
{'loss': 2.528, 'learning_rate': 1.5214721282509855e-07, 'epoch': 15.07}
+
 94%|█████████▍| 349960/371472 [5:41:23<1:41:05,  3.55it/s]
 94%|█████████▍| 349961/371472 [5:41:23<1:37:52,  3.66it/s]
 94%|█████████▍| 349962/371472 [5:41:23<1:53:06,  3.17it/s]
 94%|█████████▍| 349963/371472 [5:41:24<1:47:57,  3.32it/s]
 94%|█████████▍| 349964/371472 [5:41:24<1:55:12,  3.11it/s]
 94%|█████████▍| 349965/371472 [5:41:24<1:49:51,  3.26it/s]
 94%|█████████▍| 349966/371472 [5:41:25<1:42:57,  3.48it/s]
 94%|█████████▍| 349967/371472 [5:41:25<1:43:09,  3.47it/s]
 94%|█████████▍| 349968/371472 [5:41:25<1:39:19,  3.61it/s]
 94%|█████████▍| 349969/371472 [5:41:25<1:44:26,  3.43it/s]
 94%|█████████▍| 349970/371472 [5:41:26<1:40:57,  3.55it/s]
 94%|█████████▍| 349971/371472 [5:41:26<1:45:27,  3.40it/s]
 94%|█████████▍| 349972/371472 [5:41:26<1:41:20,  3.54it/s]
 94%|█████████▍| 349973/371472 [5:41:27<1:39:10,  3.61it/s]
 94%|█████████▍| 349974/371472 [5:41:27<1:38:22,  3.64it/s]
 94%|█████████▍| 349975/371472 [5:41:27<1:36:41,  3.71it/s]
 94%|█████████▍| 349976/371472 [5:41:27<1:34:20,  3.80it/s]
 94%|█████████▍| 349977/371472 [5:41:28<1:34:09,  3.80it/s]
 94%|█████████▍| 349978/371472 [5:41:28<1:35:29,  3.75it/s]
 94%|█████████▍| 349979/371472 [5:41:28<1:33:57,  3.81it/s]
 94%|█████████▍| 349980/371472 [5:41:28<1:32:08,  3.89it/s]
                                                           
{'loss': 2.5758, 'learning_rate': 1.520987308496197e-07, 'epoch': 15.07}
+
 94%|█████████▍| 349980/371472 [5:41:28<1:32:08,  3.89it/s]
 94%|█████████▍| 349981/371472 [5:41:29<1:41:35,  3.53it/s]
 94%|█████████▍| 349982/371472 [5:41:29<1:44:38,  3.42it/s]
 94%|█████████▍| 349983/371472 [5:41:29<1:49:32,  3.27it/s]
 94%|█████████▍| 349984/371472 [5:41:30<1:47:33,  3.33it/s]
 94%|█████████▍| 349985/371472 [5:41:30<1:50:37,  3.24it/s]
 94%|█████████▍| 349986/371472 [5:41:30<1:51:11,  3.22it/s]
 94%|█████████▍| 349987/371472 [5:41:31<1:48:00,  3.32it/s]
 94%|█████████▍| 349988/371472 [5:41:31<1:44:49,  3.42it/s]
 94%|█████████▍| 349989/371472 [5:41:31<1:40:31,  3.56it/s]
 94%|█████████▍| 349990/371472 [5:41:31<1:44:51,  3.41it/s]
 94%|█████████▍| 349991/371472 [5:41:32<1:52:48,  3.17it/s]
 94%|█████████▍| 349992/371472 [5:41:32<1:50:22,  3.24it/s]
 94%|█████████▍| 349993/371472 [5:41:32<1:49:27,  3.27it/s]
 94%|█████████▍| 349994/371472 [5:41:33<1:44:32,  3.42it/s]
 94%|█████████▍| 349995/371472 [5:41:33<1:41:45,  3.52it/s]
 94%|█████████▍| 349996/371472 [5:41:33<1:38:52,  3.62it/s]
 94%|█████████▍| 349997/371472 [5:41:34<2:05:47,  2.85it/s]
 94%|█████████▍| 349998/371472 [5:41:34<1:56:06,  3.08it/s]
 94%|█████████▍| 349999/371472 [5:41:34<1:56:51,  3.06it/s]
 94%|█████████▍| 350000/371472 [5:41:35<1:48:24,  3.30it/s]
                                                           
{'loss': 2.5553, 'learning_rate': 1.5205024887414074e-07, 'epoch': 15.08}
+
 94%|█████████▍| 350000/371472 [5:41:35<1:48:24,  3.30it/s]
 94%|█████████▍| 350001/371472 [5:41:35<1:51:21,  3.21it/s]
 94%|█████████▍| 350002/371472 [5:41:35<1:44:54,  3.41it/s]
 94%|█████████▍| 350003/371472 [5:41:35<1:42:40,  3.48it/s]
 94%|█████████▍| 350004/371472 [5:41:36<1:39:27,  3.60it/s]
 94%|█████████▍| 350005/371472 [5:41:36<1:46:41,  3.35it/s]
 94%|█████████▍| 350006/371472 [5:41:36<1:43:11,  3.47it/s]
 94%|█████████▍| 350007/371472 [5:41:37<1:40:15,  3.57it/s]
 94%|█████████▍| 350008/371472 [5:41:37<1:38:43,  3.62it/s]
 94%|█████████▍| 350009/371472 [5:41:37<1:36:44,  3.70it/s]
 94%|█████████▍| 350010/371472 [5:41:37<1:35:51,  3.73it/s]
 94%|█████████▍| 350011/371472 [5:41:38<1:38:42,  3.62it/s]
 94%|█████████▍| 350012/371472 [5:41:38<1:38:31,  3.63it/s]
 94%|█████████▍| 350013/371472 [5:41:38<1:46:14,  3.37it/s]
 94%|█████████▍| 350014/371472 [5:41:39<1:49:47,  3.26it/s]
 94%|█████████▍| 350015/371472 [5:41:39<1:42:40,  3.48it/s]
 94%|█████████▍| 350016/371472 [5:41:39<1:49:18,  3.27it/s]
 94%|█████████▍| 350017/371472 [5:41:39<1:45:07,  3.40it/s]
 94%|█████████▍| 350018/371472 [5:41:40<1:41:14,  3.53it/s]
 94%|█████████▍| 350019/371472 [5:41:40<1:42:16,  3.50it/s]
 94%|█████████▍| 350020/371472 [5:41:40<1:42:38,  3.48it/s]
                                                           
{'loss': 2.4453, 'learning_rate': 1.5200176689866192e-07, 'epoch': 15.08}
+
 94%|█████████▍| 350020/371472 [5:41:40<1:42:38,  3.48it/s]
 94%|█████████▍| 350021/371472 [5:41:41<1:49:59,  3.25it/s]
 94%|█████████▍| 350022/371472 [5:41:41<1:50:53,  3.22it/s]
 94%|█████████▍| 350023/371472 [5:41:41<1:48:30,  3.29it/s]
 94%|█████████▍| 350024/371472 [5:41:41<1:43:20,  3.46it/s]
 94%|█████████▍| 350025/371472 [5:41:42<1:39:50,  3.58it/s]
 94%|█████████▍| 350026/371472 [5:41:42<1:42:18,  3.49it/s]
 94%|█████████▍| 350027/371472 [5:41:42<1:43:08,  3.47it/s]
 94%|█████████▍| 350028/371472 [5:41:43<1:41:36,  3.52it/s]
 94%|█████████▍| 350029/371472 [5:41:43<1:40:49,  3.54it/s]
 94%|█████████▍| 350030/371472 [5:41:43<1:45:56,  3.37it/s]
 94%|█████████▍| 350031/371472 [5:41:44<1:44:19,  3.43it/s]
 94%|█████████▍| 350032/371472 [5:41:44<1:47:24,  3.33it/s]
 94%|█████████▍| 350033/371472 [5:41:44<1:47:24,  3.33it/s]
 94%|█████████▍| 350034/371472 [5:41:44<1:50:35,  3.23it/s]
 94%|█████████▍| 350035/371472 [5:41:45<1:44:50,  3.41it/s]
 94%|█████████▍| 350036/371472 [5:41:45<1:44:07,  3.43it/s]
 94%|█████████▍| 350037/371472 [5:41:45<1:41:47,  3.51it/s]
 94%|█████████▍| 350038/371472 [5:41:46<1:41:11,  3.53it/s]
 94%|█████████▍| 350039/371472 [5:41:46<1:40:15,  3.56it/s]
 94%|█████████▍| 350040/371472 [5:41:46<1:41:25,  3.52it/s]
                                                           
{'loss': 2.5895, 'learning_rate': 1.5195328492318296e-07, 'epoch': 15.08}
+
 94%|█████████▍| 350040/371472 [5:41:46<1:41:25,  3.52it/s]
 94%|█████████▍| 350041/371472 [5:41:47<1:59:25,  2.99it/s]
 94%|█████████▍| 350042/371472 [5:41:47<1:56:43,  3.06it/s]
 94%|█████████▍| 350043/371472 [5:41:47<1:49:53,  3.25it/s]
 94%|█████████▍| 350044/371472 [5:41:47<1:44:10,  3.43it/s]
 94%|█████████▍| 350045/371472 [5:41:48<1:45:12,  3.39it/s]
 94%|█████████▍| 350046/371472 [5:41:48<1:43:00,  3.47it/s]
 94%|█████████▍| 350047/371472 [5:41:48<1:38:37,  3.62it/s]
 94%|█████████▍| 350048/371472 [5:41:48<1:34:42,  3.77it/s]
 94%|█████████▍| 350049/371472 [5:41:49<1:37:10,  3.67it/s]
 94%|█████████▍| 350050/371472 [5:41:49<1:33:57,  3.80it/s]
 94%|█████████▍| 350051/371472 [5:41:49<1:35:23,  3.74it/s]
 94%|█████████▍| 350052/371472 [5:41:50<1:34:44,  3.77it/s]
 94%|█████████▍| 350053/371472 [5:41:50<1:43:53,  3.44it/s]
 94%|█████████▍| 350054/371472 [5:41:50<1:43:24,  3.45it/s]
 94%|█████████▍| 350055/371472 [5:41:50<1:44:24,  3.42it/s]
 94%|█████████▍| 350056/371472 [5:41:51<1:41:44,  3.51it/s]
 94%|█████████▍| 350057/371472 [5:41:51<1:38:01,  3.64it/s]
 94%|█████████▍| 350058/371472 [5:41:51<1:39:00,  3.60it/s]
 94%|█████████▍| 350059/371472 [5:41:52<1:39:38,  3.58it/s]
 94%|█████████▍| 350060/371472 [5:41:52<1:38:15,  3.63it/s]
                                                           
{'loss': 2.552, 'learning_rate': 1.5190480294770412e-07, 'epoch': 15.08}
+
 94%|█████████▍| 350060/371472 [5:41:52<1:38:15,  3.63it/s]
 94%|█████████▍| 350061/371472 [5:41:52<1:45:53,  3.37it/s]
 94%|█████████▍| 350062/371472 [5:41:52<1:40:03,  3.57it/s]
 94%|█████████▍| 350063/371472 [5:41:53<1:42:26,  3.48it/s]
 94%|█████████▍| 350064/371472 [5:41:53<1:40:24,  3.55it/s]
 94%|█████████▍| 350065/371472 [5:41:53<1:38:30,  3.62it/s]
 94%|█████████▍| 350066/371472 [5:41:54<1:37:36,  3.65it/s]
 94%|█████████▍| 350067/371472 [5:41:54<1:45:23,  3.39it/s]
 94%|█████████▍| 350068/371472 [5:41:54<1:48:00,  3.30it/s]
 94%|█████████▍| 350069/371472 [5:41:54<1:44:43,  3.41it/s]
 94%|█████████▍| 350070/371472 [5:41:55<1:42:32,  3.48it/s]
 94%|█████████▍| 350071/371472 [5:41:55<1:43:19,  3.45it/s]
 94%|█████████▍| 350072/371472 [5:41:55<1:38:27,  3.62it/s]
 94%|█████████▍| 350073/371472 [5:41:56<1:40:34,  3.55it/s]
 94%|█████████▍| 350074/371472 [5:41:56<1:38:52,  3.61it/s]
 94%|█████████▍| 350075/371472 [5:41:56<1:45:07,  3.39it/s]
 94%|█████████▍| 350076/371472 [5:41:57<1:52:12,  3.18it/s]
 94%|█████████▍| 350077/371472 [5:41:57<1:45:30,  3.38it/s]
 94%|█████████▍| 350078/371472 [5:41:57<1:40:28,  3.55it/s]
 94%|█████████▍| 350079/371472 [5:41:57<1:42:47,  3.47it/s]
 94%|█████████▍| 350080/371472 [5:41:58<1:37:35,  3.65it/s]
                                                           
{'loss': 2.5067, 'learning_rate': 1.518563209722252e-07, 'epoch': 15.08}
+
 94%|█████████▍| 350080/371472 [5:41:58<1:37:35,  3.65it/s]
 94%|█████████▍| 350081/371472 [5:41:58<1:42:49,  3.47it/s]
 94%|█████████▍| 350082/371472 [5:41:58<1:42:02,  3.49it/s]
 94%|█████████▍| 350083/371472 [5:41:58<1:38:56,  3.60it/s]
 94%|█████████▍| 350084/371472 [5:41:59<1:35:43,  3.72it/s]
 94%|█████████▍| 350085/371472 [5:41:59<1:40:10,  3.56it/s]
 94%|█████████▍| 350086/371472 [5:41:59<1:40:43,  3.54it/s]
 94%|█████████▍| 350087/371472 [5:42:00<1:40:45,  3.54it/s]
 94%|█████████▍| 350088/371472 [5:42:00<1:38:28,  3.62it/s]
 94%|█████████▍| 350089/371472 [5:42:00<1:38:34,  3.62it/s]
 94%|█████████▍| 350090/371472 [5:42:00<1:34:13,  3.78it/s]
 94%|█████████▍| 350091/371472 [5:42:01<1:36:24,  3.70it/s]
 94%|█████████▍| 350092/371472 [5:42:01<1:36:59,  3.67it/s]
 94%|█████████▍| 350093/371472 [5:42:01<1:37:10,  3.67it/s]
 94%|█████████▍| 350094/371472 [5:42:01<1:34:52,  3.76it/s]
 94%|█████████▍| 350095/371472 [5:42:02<1:35:04,  3.75it/s]
 94%|█████████▍| 350096/371472 [5:42:02<1:37:03,  3.67it/s]
 94%|█████████▍| 350097/371472 [5:42:02<1:36:29,  3.69it/s]
 94%|█████████▍| 350098/371472 [5:42:03<1:35:38,  3.72it/s]
 94%|█████████▍| 350099/371472 [5:42:03<1:50:09,  3.23it/s]
 94%|█████████▍| 350100/371472 [5:42:03<1:44:11,  3.42it/s]
                                                           
{'loss': 2.6506, 'learning_rate': 1.5180783899674634e-07, 'epoch': 15.08}
+
 94%|█████████▍| 350100/371472 [5:42:03<1:44:11,  3.42it/s]
 94%|█████████▍| 350101/371472 [5:42:03<1:40:20,  3.55it/s]
 94%|█████████▍| 350102/371472 [5:42:04<1:38:37,  3.61it/s]
 94%|█████████▍| 350103/371472 [5:42:04<1:35:31,  3.73it/s]
 94%|█████████▍| 350104/371472 [5:42:04<1:43:24,  3.44it/s]
 94%|█████████▍| 350105/371472 [5:42:05<1:46:27,  3.35it/s]
 94%|█████████▍| 350106/371472 [5:42:05<1:43:34,  3.44it/s]
 94%|█████████▍| 350107/371472 [5:42:05<1:46:04,  3.36it/s]
 94%|█████████▍| 350108/371472 [5:42:05<1:42:07,  3.49it/s]
 94%|█████████▍| 350109/371472 [5:42:06<1:42:12,  3.48it/s]
 94%|█████████▍| 350110/371472 [5:42:06<1:42:02,  3.49it/s]
 94%|█████████▍| 350111/371472 [5:42:06<1:43:21,  3.44it/s]
 94%|█████████▍| 350112/371472 [5:42:07<1:39:48,  3.57it/s]
 94%|█████████▍| 350113/371472 [5:42:07<1:37:11,  3.66it/s]
 94%|█████████▍| 350114/371472 [5:42:07<1:33:05,  3.82it/s]
 94%|█████████▍| 350115/371472 [5:42:07<1:38:44,  3.61it/s]
 94%|█████████▍| 350116/371472 [5:42:08<1:43:13,  3.45it/s]
 94%|█████████▍| 350117/371472 [5:42:08<1:40:35,  3.54it/s]
 94%|█████████▍| 350118/371472 [5:42:08<1:39:14,  3.59it/s]
 94%|█████████▍| 350119/371472 [5:42:09<1:40:13,  3.55it/s]
 94%|█████████▍| 350120/371472 [5:42:09<1:35:48,  3.71it/s]
                                                           
{'loss': 2.6057, 'learning_rate': 1.5175935702126738e-07, 'epoch': 15.08}
+
 94%|█████████▍| 350120/371472 [5:42:09<1:35:48,  3.71it/s]
 94%|█████████▍| 350121/371472 [5:42:09<1:35:01,  3.75it/s]
 94%|█████████▍| 350122/371472 [5:42:09<1:34:36,  3.76it/s]
 94%|█████████▍| 350123/371472 [5:42:10<1:34:14,  3.78it/s]
 94%|█████████▍| 350124/371472 [5:42:10<1:35:22,  3.73it/s]
 94%|█████████▍| 350125/371472 [5:42:10<1:36:28,  3.69it/s]
 94%|█████████▍| 350126/371472 [5:42:10<1:42:28,  3.47it/s]
 94%|█████████▍| 350127/371472 [5:42:11<1:42:03,  3.49it/s]
 94%|█████████▍| 350128/371472 [5:42:11<1:40:22,  3.54it/s]
 94%|█████████▍| 350129/371472 [5:42:11<1:41:40,  3.50it/s]
 94%|█████████▍| 350130/371472 [5:42:12<1:54:34,  3.10it/s]
 94%|█████████▍| 350131/371472 [5:42:12<1:47:47,  3.30it/s]
 94%|█████████▍| 350132/371472 [5:42:12<1:45:23,  3.37it/s]
 94%|█████████▍| 350133/371472 [5:42:12<1:38:16,  3.62it/s]
 94%|█████████▍| 350134/371472 [5:42:13<1:36:42,  3.68it/s]
 94%|█████████▍| 350135/371472 [5:42:13<1:36:56,  3.67it/s]
 94%|█████████▍| 350136/371472 [5:42:13<1:36:55,  3.67it/s]
 94%|█████████▍| 350137/371472 [5:42:14<1:36:08,  3.70it/s]
 94%|█████████▍| 350138/371472 [5:42:14<1:38:04,  3.63it/s]
 94%|█████████▍| 350139/371472 [5:42:14<1:36:37,  3.68it/s]
 94%|█████████▍| 350140/371472 [5:42:14<1:43:00,  3.45it/s]
                                                           
{'loss': 2.4937, 'learning_rate': 1.5171087504578856e-07, 'epoch': 15.08}
+
 94%|█████████▍| 350140/371472 [5:42:14<1:43:00,  3.45it/s]
 94%|█████████▍| 350141/371472 [5:42:15<1:51:19,  3.19it/s]
 94%|█████████▍| 350142/371472 [5:42:15<1:50:40,  3.21it/s]
 94%|█████████▍| 350143/371472 [5:42:15<1:49:43,  3.24it/s]
 94%|█████████▍| 350144/371472 [5:42:16<1:46:51,  3.33it/s]
 94%|█████████▍| 350145/371472 [5:42:16<1:46:55,  3.32it/s]
 94%|█████████▍| 350146/371472 [5:42:16<1:44:53,  3.39it/s]
 94%|█████████▍| 350147/371472 [5:42:17<1:40:37,  3.53it/s]
 94%|█████████▍| 350148/371472 [5:42:17<1:46:00,  3.35it/s]
 94%|█████████▍| 350149/371472 [5:42:17<1:40:42,  3.53it/s]
 94%|█████████▍| 350150/371472 [5:42:17<1:40:35,  3.53it/s]
 94%|█████████▍| 350151/371472 [5:42:18<1:42:21,  3.47it/s]
 94%|█████████▍| 350152/371472 [5:42:18<1:41:37,  3.50it/s]
 94%|█████████▍| 350153/371472 [5:42:18<1:38:09,  3.62it/s]
 94%|█████████▍| 350154/371472 [5:42:19<1:42:36,  3.46it/s]
 94%|█████████▍| 350155/371472 [5:42:19<1:40:22,  3.54it/s]
 94%|█████████▍| 350156/371472 [5:42:19<1:39:47,  3.56it/s]
 94%|█████████▍| 350157/371472 [5:42:19<1:38:39,  3.60it/s]
 94%|█████████▍| 350158/371472 [5:42:20<1:43:20,  3.44it/s]
 94%|█████████▍| 350159/371472 [5:42:20<1:44:36,  3.40it/s]
 94%|█████████▍| 350160/371472 [5:42:20<1:44:36,  3.40it/s]
                                                           
{'loss': 2.5075, 'learning_rate': 1.5166239307030963e-07, 'epoch': 15.08}
+
 94%|█████████▍| 350160/371472 [5:42:20<1:44:36,  3.40it/s]
 94%|█████████▍| 350161/371472 [5:42:21<1:48:11,  3.28it/s]
 94%|█████████▍| 350162/371472 [5:42:21<1:46:44,  3.33it/s]
 94%|█████████▍| 350163/371472 [5:42:21<1:42:29,  3.47it/s]
 94%|█████████▍| 350164/371472 [5:42:21<1:44:24,  3.40it/s]
 94%|█████████▍| 350165/371472 [5:42:22<1:49:22,  3.25it/s]
 94%|█████████▍| 350166/371472 [5:42:22<1:44:51,  3.39it/s]
 94%|█████████▍| 350167/371472 [5:42:22<1:39:14,  3.58it/s]
 94%|█████████▍| 350168/371472 [5:42:23<1:37:40,  3.64it/s]
 94%|█████████▍| 350169/371472 [5:42:23<1:45:27,  3.37it/s]
 94%|█████████▍| 350170/371472 [5:42:23<1:41:28,  3.50it/s]
 94%|█████████▍| 350171/371472 [5:42:23<1:38:36,  3.60it/s]
 94%|█████████▍| 350172/371472 [5:42:24<1:37:40,  3.63it/s]
 94%|█████████▍| 350173/371472 [5:42:24<1:41:03,  3.51it/s]
 94%|█████████▍| 350174/371472 [5:42:24<1:46:14,  3.34it/s]
 94%|█████████▍| 350175/371472 [5:42:25<1:41:32,  3.50it/s]
 94%|█████████▍| 350176/371472 [5:42:25<1:46:59,  3.32it/s]
 94%|█████████▍| 350177/371472 [5:42:25<1:46:04,  3.35it/s]
 94%|█████��███▍| 350178/371472 [5:42:25<1:39:42,  3.56it/s]
 94%|█████████▍| 350179/371472 [5:42:26<1:41:10,  3.51it/s]
 94%|█████████▍| 350180/371472 [5:42:26<1:41:01,  3.51it/s]
                                                           
{'loss': 2.5972, 'learning_rate': 1.5161391109483076e-07, 'epoch': 15.08}
+
 94%|█████████▍| 350180/371472 [5:42:26<1:41:01,  3.51it/s]
 94%|█████████▍| 350181/371472 [5:42:26<1:44:04,  3.41it/s]
 94%|█████████▍| 350182/371472 [5:42:27<1:41:45,  3.49it/s]
 94%|█████████▍| 350183/371472 [5:42:27<1:38:22,  3.61it/s]
 94%|█████████▍| 350184/371472 [5:42:27<1:37:00,  3.66it/s]
 94%|█████████▍| 350185/371472 [5:42:27<1:38:37,  3.60it/s]
 94%|█████████▍| 350186/371472 [5:42:28<1:42:12,  3.47it/s]
 94%|█████████▍| 350187/371472 [5:42:28<1:37:12,  3.65it/s]
 94%|█████████▍| 350188/371472 [5:42:28<1:35:29,  3.71it/s]
 94%|█████████▍| 350189/371472 [5:42:29<1:37:25,  3.64it/s]
 94%|█████████▍| 350190/371472 [5:42:29<1:41:17,  3.50it/s]
 94%|█████████▍| 350191/371472 [5:42:29<1:41:44,  3.49it/s]
 94%|█████████▍| 350192/371472 [5:42:29<1:38:24,  3.60it/s]
 94%|█████████▍| 350193/371472 [5:42:30<1:41:05,  3.51it/s]
 94%|█████████▍| 350194/371472 [5:42:30<1:41:07,  3.51it/s]
 94%|█████████▍| 350195/371472 [5:42:30<1:38:24,  3.60it/s]
 94%|█████████▍| 350196/371472 [5:42:30<1:36:00,  3.69it/s]
 94%|█████████▍| 350197/371472 [5:42:31<1:40:49,  3.52it/s]
 94%|█████████▍| 350198/371472 [5:42:31<1:39:37,  3.56it/s]
 94%|█████████▍| 350199/371472 [5:42:31<1:40:41,  3.52it/s]
 94%|█████████▍| 350200/371472 [5:42:32<1:39:17,  3.57it/s]
                                                           
{'loss': 2.5704, 'learning_rate': 1.5156542911935183e-07, 'epoch': 15.08}
+
 94%|█████████▍| 350200/371472 [5:42:32<1:39:17,  3.57it/s]
 94%|█████████▍| 350201/371472 [5:42:32<1:44:01,  3.41it/s]
 94%|█████████▍| 350202/371472 [5:42:32<1:39:01,  3.58it/s]
 94%|█████████▍| 350203/371472 [5:42:32<1:35:06,  3.73it/s]
 94%|█████████▍| 350204/371472 [5:42:33<1:43:56,  3.41it/s]
 94%|█████████▍| 350205/371472 [5:42:33<1:43:47,  3.42it/s]
 94%|█████████▍| 350206/371472 [5:42:33<1:43:40,  3.42it/s]
 94%|█████████▍| 350207/371472 [5:42:34<1:43:35,  3.42it/s]
 94%|█████████▍| 350208/371472 [5:42:34<1:45:59,  3.34it/s]
 94%|█████████▍| 350209/371472 [5:42:34<1:54:35,  3.09it/s]
 94%|█████████▍| 350210/371472 [5:42:35<1:50:55,  3.19it/s]
 94%|█████████▍| 350211/371472 [5:42:35<1:52:19,  3.15it/s]
 94%|█████████▍| 350212/371472 [5:42:35<1:47:58,  3.28it/s]
 94%|█████████▍| 350213/371472 [5:42:36<1:43:38,  3.42it/s]
 94%|█████████▍| 350214/371472 [5:42:36<1:41:15,  3.50it/s]
 94%|█████████▍| 350215/371472 [5:42:36<1:41:42,  3.48it/s]
 94%|█████████▍| 350216/371472 [5:42:36<1:38:51,  3.58it/s]
 94%|█████████▍| 350217/371472 [5:42:37<1:39:31,  3.56it/s]
 94%|█████████▍| 350218/371472 [5:42:37<1:37:26,  3.64it/s]
 94%|█████████▍| 350219/371472 [5:42:37<1:40:40,  3.52it/s]
 94%|█████████▍| 350220/371472 [5:42:37<1:39:07,  3.57it/s]
                                                           
{'loss': 2.5659, 'learning_rate': 1.5151694714387298e-07, 'epoch': 15.08}
+
 94%|█████████▍| 350220/371472 [5:42:37<1:39:07,  3.57it/s]
 94%|█████████▍| 350221/371472 [5:42:38<1:44:01,  3.40it/s]
 94%|█████████▍| 350222/371472 [5:42:38<1:38:57,  3.58it/s]
 94%|█████████▍| 350223/371472 [5:42:38<1:38:29,  3.60it/s]
 94%|█████████▍| 350224/371472 [5:42:39<1:36:25,  3.67it/s]
 94%|█████████▍| 350225/371472 [5:42:39<1:40:10,  3.54it/s]
 94%|█████████▍| 350226/371472 [5:42:39<1:36:25,  3.67it/s]
 94%|█████████▍| 350227/371472 [5:42:39<1:38:57,  3.58it/s]
 94%|█████████▍| 350228/371472 [5:42:40<1:36:04,  3.69it/s]
 94%|█████████▍| 350229/371472 [5:42:40<1:34:59,  3.73it/s]
 94%|█████████▍| 350230/371472 [5:42:40<1:32:55,  3.81it/s]
 94%|█████████▍| 350231/371472 [5:42:40<1:37:13,  3.64it/s]
 94%|█████████▍| 350232/371472 [5:42:41<1:36:03,  3.69it/s]
 94%|█████████▍| 350233/371472 [5:42:41<1:36:57,  3.65it/s]
 94%|█████████▍| 350234/371472 [5:42:41<1:37:12,  3.64it/s]
 94%|█████████▍| 350235/371472 [5:42:42<1:35:41,  3.70it/s]
 94%|█████████▍| 350236/371472 [5:42:42<1:35:46,  3.70it/s]
 94%|█████████▍| 350237/371472 [5:42:42<1:38:08,  3.61it/s]
 94%|█████████▍| 350238/371472 [5:42:42<1:36:48,  3.66it/s]
 94%|█████████▍| 350239/371472 [5:42:43<1:33:32,  3.78it/s]
 94%|█████████▍| 350240/371472 [5:42:43<1:31:43,  3.86it/s]
                                                           
{'loss': 2.6758, 'learning_rate': 1.5146846516839405e-07, 'epoch': 15.09}
+
 94%|█████████▍| 350240/371472 [5:42:43<1:31:43,  3.86it/s]
 94%|█████████▍| 350241/371472 [5:42:43<1:34:02,  3.76it/s]
 94%|█████████▍| 350242/371472 [5:42:44<1:43:57,  3.40it/s]
 94%|█████████▍| 350243/371472 [5:42:44<1:42:35,  3.45it/s]
 94%|█████████▍| 350244/371472 [5:42:44<1:43:39,  3.41it/s]
 94%|█████████▍| 350245/371472 [5:42:44<1:40:14,  3.53it/s]
 94%|█████████▍| 350246/371472 [5:42:45<1:37:22,  3.63it/s]
 94%|█████████▍| 350247/371472 [5:42:45<1:37:22,  3.63it/s]
 94%|█████████▍| 350248/371472 [5:42:45<1:38:22,  3.60it/s]
 94%|█████████▍| 350249/371472 [5:42:46<1:41:59,  3.47it/s]
 94%|█████████▍| 350250/371472 [5:42:46<1:42:19,  3.46it/s]
 94%|█████████▍| 350251/371472 [5:42:46<1:41:32,  3.48it/s]
 94%|█████████▍| 350252/371472 [5:42:46<1:41:44,  3.48it/s]
 94%|█████████▍| 350253/371472 [5:42:47<1:42:21,  3.45it/s]
 94%|█████████▍| 350254/371472 [5:42:47<1:39:48,  3.54it/s]
 94%|█████████▍| 350255/371472 [5:42:47<1:37:13,  3.64it/s]
 94%|█████████▍| 350256/371472 [5:42:47<1:33:14,  3.79it/s]
 94%|█████████▍| 350257/371472 [5:42:48<1:34:27,  3.74it/s]
 94%|█████████▍| 350258/371472 [5:42:48<1:33:14,  3.79it/s]
 94%|█████████▍| 350259/371472 [5:42:48<1:33:08,  3.80it/s]
 94%|█████████▍| 350260/371472 [5:42:48<1:33:03,  3.80it/s]
                                                           
{'loss': 2.5076, 'learning_rate': 1.514199831929152e-07, 'epoch': 15.09}
+
 94%|█████████▍| 350260/371472 [5:42:48<1:33:03,  3.80it/s]
 94%|█████████▍| 350261/371472 [5:42:49<1:33:05,  3.80it/s]
 94%|█████████▍| 350262/371472 [5:42:49<1:33:34,  3.78it/s]
 94%|█████████▍| 350263/371472 [5:42:49<1:35:07,  3.72it/s]
 94%|█████████▍| 350264/371472 [5:42:50<1:34:29,  3.74it/s]
 94%|█████████▍| 350265/371472 [5:42:50<1:40:05,  3.53it/s]
 94%|█████████▍| 350266/371472 [5:42:50<1:40:08,  3.53it/s]
 94%|█████████▍| 350267/371472 [5:42:50<1:39:11,  3.56it/s]
 94%|█████████▍| 350268/371472 [5:42:51<1:35:46,  3.69it/s]
 94%|█████████▍| 350269/371472 [5:42:51<1:33:51,  3.77it/s]
 94%|█████████▍| 350270/371472 [5:42:51<1:34:38,  3.73it/s]
 94%|█████████▍| 350271/371472 [5:42:51<1:35:08,  3.71it/s]
 94%|█████████▍| 350272/371472 [5:42:52<1:35:25,  3.70it/s]
 94%|█████████▍| 350273/371472 [5:42:52<1:38:57,  3.57it/s]
 94%|█████████▍| 350274/371472 [5:42:52<1:37:19,  3.63it/s]
 94%|█████████▍| 350275/371472 [5:42:53<1:36:41,  3.65it/s]
 94%|█████████▍| 350276/371472 [5:42:53<1:34:06,  3.75it/s]
 94%|█████████▍| 350277/371472 [5:42:53<1:35:56,  3.68it/s]
 94%|█████████▍| 350278/371472 [5:42:53<1:37:58,  3.61it/s]
 94%|█████████▍| 350279/371472 [5:42:54<1:38:22,  3.59it/s]
 94%|█████████▍| 350280/371472 [5:42:54<1:36:55,  3.64it/s]
                                                           
{'loss': 2.6375, 'learning_rate': 1.5137150121743627e-07, 'epoch': 15.09}
+
 94%|█████████▍| 350280/371472 [5:42:54<1:36:55,  3.64it/s]
 94%|█████████▍| 350281/371472 [5:42:54<1:38:00,  3.60it/s]
 94%|█████████▍| 350282/371472 [5:42:55<1:36:51,  3.65it/s]
 94%|█████████▍| 350283/371472 [5:42:55<1:36:44,  3.65it/s]
 94%|█████████▍| 350284/371472 [5:42:55<1:39:24,  3.55it/s]
 94%|█████████▍| 350285/371472 [5:42:55<1:43:54,  3.40it/s]
 94%|█████████▍| 350286/371472 [5:42:56<1:40:22,  3.52it/s]
 94%|█████████▍| 350287/371472 [5:42:56<1:40:38,  3.51it/s]
 94%|█████████▍| 350288/371472 [5:42:56<1:48:22,  3.26it/s]
 94%|█████████▍| 350289/371472 [5:42:57<1:48:10,  3.26it/s]
 94%|��████████▍| 350290/371472 [5:42:57<1:44:46,  3.37it/s]
 94%|█████████▍| 350291/371472 [5:42:57<1:42:51,  3.43it/s]
 94%|█████████▍| 350292/371472 [5:42:57<1:45:12,  3.36it/s]
 94%|█████████▍| 350293/371472 [5:42:58<1:44:44,  3.37it/s]
 94%|█████████▍| 350294/371472 [5:42:58<1:39:17,  3.55it/s]
 94%|█████████▍| 350295/371472 [5:42:58<1:37:41,  3.61it/s]
 94%|█████████▍| 350296/371472 [5:42:59<1:36:40,  3.65it/s]
 94%|█████████▍| 350297/371472 [5:42:59<1:34:49,  3.72it/s]
 94%|█████████▍| 350298/371472 [5:42:59<1:36:08,  3.67it/s]
 94%|█████████▍| 350299/371472 [5:42:59<1:35:31,  3.69it/s]
 94%|█████████▍| 350300/371472 [5:43:00<1:36:37,  3.65it/s]
                                                           
{'loss': 2.6214, 'learning_rate': 1.5132301924195742e-07, 'epoch': 15.09}
+
 94%|█████████▍| 350300/371472 [5:43:00<1:36:37,  3.65it/s]
 94%|█████████▍| 350301/371472 [5:43:00<1:36:24,  3.66it/s]
 94%|█████████▍| 350302/371472 [5:43:00<1:43:49,  3.40it/s]
 94%|█████████▍| 350303/371472 [5:43:01<1:42:37,  3.44it/s]
 94%|█████████▍| 350304/371472 [5:43:01<1:37:24,  3.62it/s]
 94%|█████████▍| 350305/371472 [5:43:01<1:35:55,  3.68it/s]
 94%|█████████▍| 350306/371472 [5:43:01<1:39:07,  3.56it/s]
 94%|█████████▍| 350307/371472 [5:43:02<1:46:37,  3.31it/s]
 94%|█████████▍| 350308/371472 [5:43:02<1:40:22,  3.51it/s]
 94%|█████████▍| 350309/371472 [5:43:02<1:40:25,  3.51it/s]
 94%|█████████▍| 350310/371472 [5:43:02<1:39:01,  3.56it/s]
 94%|█████████▍| 350311/371472 [5:43:03<1:37:15,  3.63it/s]
 94%|█████████▍| 350312/371472 [5:43:03<1:37:51,  3.60it/s]
 94%|█████████▍| 350313/371472 [5:43:03<1:40:46,  3.50it/s]
 94%|█████████▍| 350314/371472 [5:43:04<1:42:55,  3.43it/s]
 94%|█████████▍| 350315/371472 [5:43:04<1:39:48,  3.53it/s]
 94%|█████████▍| 350316/371472 [5:43:04<1:36:55,  3.64it/s]
 94%|█████████▍| 350317/371472 [5:43:04<1:37:38,  3.61it/s]
 94%|█████████▍| 350318/371472 [5:43:05<1:37:40,  3.61it/s]
 94%|█████████▍| 350319/371472 [5:43:05<1:36:59,  3.64it/s]
 94%|█████████▍| 350320/371472 [5:43:05<1:36:42,  3.65it/s]
                                                           
{'loss': 2.5668, 'learning_rate': 1.5127453726647847e-07, 'epoch': 15.09}
+
 94%|█████████▍| 350320/371472 [5:43:05<1:36:42,  3.65it/s]
 94%|█████████▍| 350321/371472 [5:43:06<1:37:15,  3.62it/s]
 94%|█████████▍| 350322/371472 [5:43:06<1:49:12,  3.23it/s]
 94%|█████████▍| 350323/371472 [5:43:06<1:48:36,  3.25it/s]
 94%|█████████▍| 350324/371472 [5:43:07<1:44:40,  3.37it/s]
 94%|█████████▍| 350325/371472 [5:43:07<1:42:27,  3.44it/s]
 94%|█████████▍| 350326/371472 [5:43:07<1:38:38,  3.57it/s]
 94%|█████████▍| 350327/371472 [5:43:07<1:39:41,  3.54it/s]
 94%|█████████▍| 350328/371472 [5:43:08<1:37:54,  3.60it/s]
 94%|█████████▍| 350329/371472 [5:43:08<1:47:19,  3.28it/s]
 94%|█████████▍| 350330/371472 [5:43:08<1:43:00,  3.42it/s]
 94%|█████████▍| 350331/371472 [5:43:08<1:39:49,  3.53it/s]
 94%|█████████▍| 350332/371472 [5:43:09<1:38:37,  3.57it/s]
 94%|█████████▍| 350333/371472 [5:43:09<1:36:55,  3.63it/s]
 94%|█████████▍| 350334/371472 [5:43:09<1:46:57,  3.29it/s]
 94%|█████████▍| 350335/371472 [5:43:10<1:45:50,  3.33it/s]
 94%|█████████▍| 350336/371472 [5:43:10<1:45:00,  3.35it/s]
 94%|█████████▍| 350337/371472 [5:43:10<1:50:09,  3.20it/s]
 94%|█████████▍| 350338/371472 [5:43:11<1:51:20,  3.16it/s]
 94%|█████████▍| 350339/371472 [5:43:11<1:48:50,  3.24it/s]
 94%|█████████▍| 350340/371472 [5:43:11<1:44:49,  3.36it/s]
                                                           
{'loss': 2.5264, 'learning_rate': 1.5122605529099962e-07, 'epoch': 15.09}
+
 94%|█████████▍| 350340/371472 [5:43:11<1:44:49,  3.36it/s]
 94%|█████████▍| 350341/371472 [5:43:11<1:40:53,  3.49it/s]
 94%|█████████▍| 350342/371472 [5:43:12<1:40:36,  3.50it/s]
 94%|█████████▍| 350343/371472 [5:43:12<1:38:51,  3.56it/s]
 94%|█████████▍| 350344/371472 [5:43:12<1:36:50,  3.64it/s]
 94%|█████████▍| 350345/371472 [5:43:13<1:36:42,  3.64it/s]
 94%|█████████▍| 350346/371472 [5:43:13<1:33:30,  3.77it/s]
 94%|█████████▍| 350347/371472 [5:43:13<1:35:57,  3.67it/s]
 94%|█████████▍| 350348/371472 [5:43:13<1:36:59,  3.63it/s]
 94%|█████████▍| 350349/371472 [5:43:14<1:42:07,  3.45it/s]
 94%|█████████▍| 350350/371472 [5:43:14<1:39:54,  3.52it/s]
 94%|█████████▍| 350351/371472 [5:43:14<1:40:22,  3.51it/s]
 94%|█████████▍| 350352/371472 [5:43:15<1:42:51,  3.42it/s]
 94%|█████████▍| 350353/371472 [5:43:15<1:40:01,  3.52it/s]
 94%|█████████▍| 350354/371472 [5:43:15<1:39:58,  3.52it/s]
 94%|█████████▍| 350355/371472 [5:43:15<1:40:45,  3.49it/s]
 94%|█████████▍| 350356/371472 [5:43:16<1:38:41,  3.57it/s]
 94%|█████████▍| 350357/371472 [5:43:16<1:36:57,  3.63it/s]
 94%|█████████▍| 350358/371472 [5:43:16<1:39:37,  3.53it/s]
 94%|█████████▍| 350359/371472 [5:43:17<1:45:57,  3.32it/s]
 94%|█████████▍| 350360/371472 [5:43:17<1:40:21,  3.51it/s]
                                                           
{'loss': 2.5393, 'learning_rate': 1.511775733155207e-07, 'epoch': 15.09}
+
 94%|█████████▍| 350360/371472 [5:43:17<1:40:21,  3.51it/s]
 94%|█████████▍| 350361/371472 [5:43:17<1:37:52,  3.59it/s]
 94%|█████████▍| 350362/371472 [5:43:17<1:44:29,  3.37it/s]
 94%|█████████▍| 350363/371472 [5:43:18<1:45:23,  3.34it/s]
 94%|█████████▍| 350364/371472 [5:43:18<1:45:12,  3.34it/s]
 94%|█████████▍| 350365/371472 [5:43:18<1:40:11,  3.51it/s]
 94%|█████████▍| 350366/371472 [5:43:19<1:38:28,  3.57it/s]
 94%|█████████▍| 350367/371472 [5:43:19<1:44:10,  3.38it/s]
 94%|█████████▍| 350368/371472 [5:43:19<1:41:21,  3.47it/s]
 94%|█████████▍| 350369/371472 [5:43:19<1:41:26,  3.47it/s]
 94%|█████████▍| 350370/371472 [5:43:20<1:45:38,  3.33it/s]
 94%|█████████▍| 350371/371472 [5:43:20<1:41:21,  3.47it/s]
 94%|█████████▍| 350372/371472 [5:43:20<1:37:46,  3.60it/s]
 94%|█████████▍| 350373/371472 [5:43:21<1:33:55,  3.74it/s]
 94%|█████████▍| 350374/371472 [5:43:21<1:39:04,  3.55it/s]
 94%|█████████▍| 350375/371472 [5:43:21<1:37:14,  3.62it/s]
 94%|█████████▍| 350376/371472 [5:43:21<1:37:32,  3.60it/s]
 94%|█████████▍| 350377/371472 [5:43:22<1:40:19,  3.50it/s]
 94%|█████████▍| 350378/371472 [5:43:22<1:38:29,  3.57it/s]
 94%|█████████▍| 350379/371472 [5:43:22<1:35:40,  3.67it/s]
 94%|█████████▍| 350380/371472 [5:43:23<1:47:08,  3.28it/s]
                                                           
{'loss': 2.6075, 'learning_rate': 1.5112909134004174e-07, 'epoch': 15.09}
+
 94%|█████████▍| 350380/371472 [5:43:23<1:47:08,  3.28it/s]
 94%|█████████▍| 350381/371472 [5:43:23<1:44:35,  3.36it/s]
 94%|█████████▍| 350382/371472 [5:43:23<1:42:10,  3.44it/s]
 94%|█████████▍| 350383/371472 [5:43:23<1:41:43,  3.46it/s]
 94%|█████████▍| 350384/371472 [5:43:24<1:46:19,  3.31it/s]
 94%|█████████▍| 350385/371472 [5:43:24<1:43:41,  3.39it/s]
 94%|█████████▍| 350386/371472 [5:43:25<2:01:31,  2.89it/s]
 94%|█████████▍| 350387/371472 [5:43:25<2:04:58,  2.81it/s]
 94%|█████████▍| 350388/371472 [5:43:25<2:01:34,  2.89it/s]
 94%|█████████▍| 350389/371472 [5:43:25<1:51:44,  3.14it/s]
 94%|█████████▍| 350390/371472 [5:43:26<1:43:19,  3.40it/s]
 94%|█████████▍| 350391/371472 [5:43:26<1:39:12,  3.54it/s]
 94%|█████████▍| 350392/371472 [5:43:26<1:40:59,  3.48it/s]
 94%|█████████▍| 350393/371472 [5:43:27<1:40:39,  3.49it/s]
 94%|█████████▍| 350394/371472 [5:43:27<1:42:14,  3.44it/s]
 94%|█████████▍| 350395/371472 [5:43:27<1:39:34,  3.53it/s]
 94%|█████████▍| 350396/371472 [5:43:27<1:39:49,  3.52it/s]
 94%|█████████▍| 350397/371472 [5:43:28<1:39:59,  3.51it/s]
 94%|█████████▍| 350398/371472 [5:43:28<1:41:06,  3.47it/s]
 94%|█████████▍| 350399/371472 [5:43:28<1:39:16,  3.54it/s]
 94%|█████████▍| 350400/371472 [5:43:29<1:39:30,  3.53it/s]
                                                           
{'loss': 2.4926, 'learning_rate': 1.5108060936456291e-07, 'epoch': 15.09}
+
 94%|█████████▍| 350400/371472 [5:43:29<1:39:30,  3.53it/s]
 94%|█████████▍| 350401/371472 [5:43:29<1:39:09,  3.54it/s]
 94%|█████████▍| 350402/371472 [5:43:29<1:38:39,  3.56it/s]
 94%|█████████▍| 350403/371472 [5:43:29<1:36:31,  3.64it/s]
 94%|█████████▍| 350404/371472 [5:43:30<1:37:51,  3.59it/s]
 94%|█████████▍| 350405/371472 [5:43:30<1:37:13,  3.61it/s]
 94%|█████████▍| 350406/371472 [5:43:30<1:36:27,  3.64it/s]
 94%|█████████▍| 350407/371472 [5:43:31<1:42:52,  3.41it/s]
 94%|█████████▍| 350408/371472 [5:43:31<1:46:29,  3.30it/s]
 94%|█████████▍| 350409/371472 [5:43:31<1:43:09,  3.40it/s]
 94%|█████████▍| 350410/371472 [5:43:31<1:40:52,  3.48it/s]
 94%|█████████▍| 350411/371472 [5:43:32<1:35:46,  3.67it/s]
 94%|█████████▍| 350412/371472 [5:43:32<1:33:09,  3.77it/s]
 94%|█████████▍| 350413/371472 [5:43:32<1:32:16,  3.80it/s]
 94%|█████████▍| 350414/371472 [5:43:33<1:44:20,  3.36it/s]
 94%|█████████▍| 350415/371472 [5:43:33<1:41:04,  3.47it/s]
 94%|█████████▍| 350416/371472 [5:43:33<1:41:35,  3.45it/s]
 94%|█████████▍| 350417/371472 [5:43:33<1:37:55,  3.58it/s]
 94%|█████████▍| 350418/371472 [5:43:34<1:47:05,  3.28it/s]
 94%|█████████▍| 350419/371472 [5:43:34<1:40:29,  3.49it/s]
 94%|█████████▍| 350420/371472 [5:43:34<1:37:58,  3.58it/s]
                                                           
{'loss': 2.5907, 'learning_rate': 1.5103212738908396e-07, 'epoch': 15.09}
+
 94%|█████████▍| 350420/371472 [5:43:34<1:37:58,  3.58it/s]
 94%|█████████▍| 350421/371472 [5:43:35<1:39:58,  3.51it/s]
 94%|█████████▍| 350422/371472 [5:43:35<1:44:50,  3.35it/s]
 94%|█████████▍| 350423/371472 [5:43:35<1:40:50,  3.48it/s]
 94%|█████████▍| 350424/371472 [5:43:35<1:40:34,  3.49it/s]
 94%|█████████▍| 350425/371472 [5:43:36<1:50:53,  3.16it/s]
 94%|█████████▍| 350426/371472 [5:43:36<1:50:54,  3.16it/s]
 94%|█████████▍| 350427/371472 [5:43:36<1:57:04,  3.00it/s]
 94%|█████████▍| 350428/371472 [5:43:37<1:50:00,  3.19it/s]
 94%|█████████▍| 350429/371472 [5:43:37<1:46:14,  3.30it/s]
 94%|█████████▍| 350430/371472 [5:43:37<1:40:37,  3.49it/s]
 94%|█████████▍| 350431/371472 [5:43:38<1:37:59,  3.58it/s]
 94%|█████████▍| 350432/371472 [5:43:38<1:37:59,  3.58it/s]
 94%|█████████▍| 350433/371472 [5:43:38<1:52:47,  3.11it/s]
 94%|█████████▍| 350434/371472 [5:43:39<1:59:28,  2.93it/s]
 94%|█████████▍| 350435/371472 [5:43:39<1:54:42,  3.06it/s]
 94%|█████████▍| 350436/371472 [5:43:39<1:55:45,  3.03it/s]
 94%|█████████▍| 350437/371472 [5:43:39<1:48:18,  3.24it/s]
 94%|█████████▍| 350438/371472 [5:43:40<1:47:23,  3.26it/s]
 94%|█████████▍| 350439/371472 [5:43:40<1:46:28,  3.29it/s]
 94%|█████████▍| 350440/371472 [5:43:40<1:42:42,  3.41it/s]
                                                           
{'loss': 2.6534, 'learning_rate': 1.509836454136051e-07, 'epoch': 15.09}
+
 94%|█████████▍| 350440/371472 [5:43:40<1:42:42,  3.41it/s]
 94%|█████████▍| 350441/371472 [5:43:41<1:41:12,  3.46it/s]
 94%|█████████▍| 350442/371472 [5:43:41<1:39:22,  3.53it/s]
 94%|█████████▍| 350443/371472 [5:43:41<1:41:09,  3.46it/s]
 94%|█████████▍| 350444/371472 [5:43:41<1:40:20,  3.49it/s]
 94%|█████████▍| 350445/371472 [5:43:42<1:37:20,  3.60it/s]
 94%|█████████▍| 350446/371472 [5:43:42<1:36:00,  3.65it/s]
 94%|█████████▍| 350447/371472 [5:43:42<1:34:04,  3.73it/s]
 94%|█████████▍| 350448/371472 [5:43:43<1:37:34,  3.59it/s]
 94%|█████████▍| 350449/371472 [5:43:43<1:38:28,  3.56it/s]
 94%|█████████▍| 350450/371472 [5:43:43<1:37:18,  3.60it/s]
 94%|█████████▍| 350451/371472 [5:43:43<1:36:27,  3.63it/s]
 94%|█████████▍| 350452/371472 [5:43:44<1:36:57,  3.61it/s]
 94%|█████████▍| 350453/371472 [5:43:44<1:33:14,  3.76it/s]
 94%|█████████▍| 350454/371472 [5:43:44<1:37:47,  3.58it/s]
 94%|█████████▍| 350455/371472 [5:43:45<1:39:32,  3.52it/s]
 94%|█████████▍| 350456/371472 [5:43:45<1:38:58,  3.54it/s]
 94%|█████████▍| 350457/371472 [5:43:45<1:37:07,  3.61it/s]
 94%|█████████▍| 350458/371472 [5:43:45<1:34:12,  3.72it/s]
 94%|█████████▍| 350459/371472 [5:43:46<1:32:45,  3.78it/s]
 94%|█████████▍| 350460/371472 [5:43:46<1:36:59,  3.61it/s]
                                                           
{'loss': 2.5793, 'learning_rate': 1.5093516343812618e-07, 'epoch': 15.09}
+
 94%|█████████▍| 350460/371472 [5:43:46<1:36:59,  3.61it/s]
 94%|█████████▍| 350461/371472 [5:43:46<1:42:51,  3.40it/s]
 94%|█████████▍| 350462/371472 [5:43:46<1:39:36,  3.52it/s]
 94%|█████████▍| 350463/371472 [5:43:47<1:36:04,  3.64it/s]
 94%|█████████▍| 350464/371472 [5:43:47<1:37:37,  3.59it/s]
 94%|█████████▍| 350465/371472 [5:43:47<1:45:37,  3.31it/s]
 94%|█████████▍| 350466/371472 [5:43:48<1:40:48,  3.47it/s]
 94%|█████████▍| 350467/371472 [5:43:48<1:35:03,  3.68it/s]
 94%|█████████▍| 350468/371472 [5:43:48<1:38:34,  3.55it/s]
 94%|█████████▍| 350469/371472 [5:43:48<1:40:51,  3.47it/s]
 94%|█████████▍| 350470/371472 [5:43:49<1:42:58,  3.40it/s]
 94%|█████████▍| 350471/371472 [5:43:49<1:43:47,  3.37it/s]
 94%|█████████▍| 350472/371472 [5:43:49<1:48:52,  3.21it/s]
 94%|█████████▍| 350473/371472 [5:43:50<1:49:10,  3.21it/s]
 94%|█████████▍| 350474/371472 [5:43:50<1:52:36,  3.11it/s]
 94%|█████████▍| 350475/371472 [5:43:50<1:45:46,  3.31it/s]
 94%|█████████▍| 350476/371472 [5:43:51<1:47:31,  3.25it/s]
 94%|█████████▍| 350477/371472 [5:43:51<1:42:14,  3.42it/s]
 94%|█████████▍| 350478/371472 [5:43:51<1:42:58,  3.40it/s]
 94%|█████████▍| 350479/371472 [5:43:51<1:38:46,  3.54it/s]
 94%|█████████▍| 350480/371472 [5:43:52<1:36:53,  3.61it/s]
                                                           
{'loss': 2.5142, 'learning_rate': 1.5088668146264733e-07, 'epoch': 15.1}
+
 94%|█████████▍| 350480/371472 [5:43:52<1:36:53,  3.61it/s]
 94%|█████████▍| 350481/371472 [5:43:52<1:34:46,  3.69it/s]
 94%|█████████▍| 350482/371472 [5:43:52<1:36:27,  3.63it/s]
 94%|█████████▍| 350483/371472 [5:43:53<1:39:09,  3.53it/s]
 94%|█████████▍| 350484/371472 [5:43:53<1:35:28,  3.66it/s]
 94%|█████████▍| 350485/371472 [5:43:53<1:35:11,  3.67it/s]
 94%|█████████▍| 350486/371472 [5:43:53<1:32:05,  3.80it/s]
 94%|█████████▍| 350487/371472 [5:43:54<1:34:35,  3.70it/s]
 94%|█████████▍| 350488/371472 [5:43:54<1:34:52,  3.69it/s]
 94%|█████████▍| 350489/371472 [5:43:54<1:36:30,  3.62it/s]
 94%|█████████▍| 350490/371472 [5:43:54<1:36:54,  3.61it/s]
 94%|█████████▍| 350491/371472 [5:43:55<1:41:45,  3.44it/s]
 94%|█████████▍| 350492/371472 [5:43:55<1:42:20,  3.42it/s]
 94%|█████████▍| 350493/371472 [5:43:55<1:43:48,  3.37it/s]
 94%|█████████▍| 350494/371472 [5:43:56<1:40:13,  3.49it/s]
 94%|█████████▍| 350495/371472 [5:43:56<1:35:55,  3.64it/s]
 94%|█████████▍| 350496/371472 [5:43:56<1:36:35,  3.62it/s]
 94%|█████████▍| 350497/371472 [5:43:56<1:33:07,  3.75it/s]
 94%|█████████▍| 350498/371472 [5:43:57<1:30:59,  3.84it/s]
 94%|█████████▍| 350499/371472 [5:43:57<1:33:09,  3.75it/s]
 94%|█████████▍| 350500/371472 [5:43:57<1:43:12,  3.39it/s]
                                                           
{'loss': 2.6497, 'learning_rate': 1.508381994871684e-07, 'epoch': 15.1}
+
 94%|█████████▍| 350500/371472 [5:43:57<1:43:12,  3.39it/s]
 94%|█████████▍| 350501/371472 [5:43:58<1:38:37,  3.54it/s]
 94%|█████████▍| 350502/371472 [5:43:58<1:46:37,  3.28it/s]
 94%|█████████▍| 350503/371472 [5:43:58<1:44:37,  3.34it/s]
 94%|█████████▍| 350504/371472 [5:43:59<1:45:21,  3.32it/s]
 94%|█████████▍| 350505/371472 [5:43:59<1:38:46,  3.54it/s]
 94%|█████████▍| 350506/371472 [5:43:59<1:41:11,  3.45it/s]
 94%|█████████▍| 350507/371472 [5:43:59<1:37:04,  3.60it/s]
 94%|█████████▍| 350508/371472 [5:44:00<1:36:14,  3.63it/s]
 94%|█████████▍| 350509/371472 [5:44:00<1:39:40,  3.51it/s]
 94%|█████████▍| 350510/371472 [5:44:00<1:40:31,  3.48it/s]
 94%|█████████▍| 350511/371472 [5:44:01<1:45:26,  3.31it/s]
 94%|█████████▍| 350512/371472 [5:44:01<1:40:27,  3.48it/s]
 94%|█████████▍| 350513/371472 [5:44:01<1:42:56,  3.39it/s]
 94%|█████████▍| 350514/371472 [5:44:01<1:40:29,  3.48it/s]
 94%|█████████▍| 350515/371472 [5:44:02<1:42:08,  3.42it/s]
 94%|█████████▍| 350516/371472 [5:44:02<1:48:39,  3.21it/s]
 94%|█████████▍| 350517/371472 [5:44:02<1:53:56,  3.07it/s]
 94%|█████████▍| 350518/371472 [5:44:03<1:46:51,  3.27it/s]
 94%|█████████▍| 350519/371472 [5:44:03<1:45:04,  3.32it/s]
 94%|█████████▍| 350520/371472 [5:44:03<1:43:38,  3.37it/s]
                                                           
{'loss': 2.6393, 'learning_rate': 1.5078971751168955e-07, 'epoch': 15.1}
+
 94%|█████████▍| 350520/371472 [5:44:03<1:43:38,  3.37it/s]
 94%|█████████▍| 350521/371472 [5:44:03<1:40:33,  3.47it/s]
 94%|█████████▍| 350522/371472 [5:44:04<1:40:18,  3.48it/s]
 94%|█████████▍| 350523/371472 [5:44:04<1:39:01,  3.53it/s]
 94%|█████████▍| 350524/371472 [5:44:04<1:35:37,  3.65it/s]
 94%|█████████▍| 350525/371472 [5:44:05<1:34:16,  3.70it/s]
 94%|█████████▍| 350526/371472 [5:44:05<1:38:25,  3.55it/s]
 94%|█████████▍| 350527/371472 [5:44:05<1:36:06,  3.63it/s]
 94%|█████████▍| 350528/371472 [5:44:05<1:38:08,  3.56it/s]
 94%|█████████▍| 350529/371472 [5:44:06<1:35:25,  3.66it/s]
 94%|█████████▍| 350530/371472 [5:44:06<1:41:15,  3.45it/s]
 94%|█████████▍| 350531/371472 [5:44:06<1:39:39,  3.50it/s]
 94%|█████████▍| 350532/371472 [5:44:07<1:38:18,  3.55it/s]
 94%|█████████▍| 350533/371472 [5:44:07<1:34:40,  3.69it/s]
 94%|█████████▍| 350534/371472 [5:44:07<1:35:44,  3.64it/s]
 94%|█████████▍| 350535/371472 [5:44:07<1:34:25,  3.70it/s]
 94%|█████████▍| 350536/371472 [5:44:08<1:35:48,  3.64it/s]
 94%|█████████▍| 350537/371472 [5:44:08<1:41:08,  3.45it/s]
 94%|█████████▍| 350538/371472 [5:44:08<1:42:01,  3.42it/s]
 94%|█████████▍| 350539/371472 [5:44:09<1:44:40,  3.33it/s]
 94%|█████████▍| 350540/371472 [5:44:09<1:43:34,  3.37it/s]
                                                           
{'loss': 2.4884, 'learning_rate': 1.5074123553621063e-07, 'epoch': 15.1}
+
 94%|█████████▍| 350540/371472 [5:44:09<1:43:34,  3.37it/s]
 94%|█████████▍| 350541/371472 [5:44:09<1:41:28,  3.44it/s]
 94%|█████████▍| 350542/371472 [5:44:09<1:46:21,  3.28it/s]
 94%|█████████▍| 350543/371472 [5:44:10<1:44:37,  3.33it/s]
 94%|█████████▍| 350544/371472 [5:44:10<1:40:19,  3.48it/s]
 94%|█████████▍| 350545/371472 [5:44:10<1:36:22,  3.62it/s]
 94%|█████████▍| 350546/371472 [5:44:11<1:36:22,  3.62it/s]
 94%|█████████▍| 350547/371472 [5:44:11<1:34:04,  3.71it/s]
 94%|█████████▍| 350548/371472 [5:44:11<1:32:29,  3.77it/s]
 94%|█████████▍| 350549/371472 [5:44:11<1:33:11,  3.74it/s]
 94%|█████████▍| 350550/371472 [5:44:12<1:33:54,  3.71it/s]
 94%|█████████▍| 350551/371472 [5:44:12<1:38:23,  3.54it/s]
 94%|█████████▍| 350552/371472 [5:44:12<1:39:57,  3.49it/s]
 94%|█████████▍| 350553/371472 [5:44:12<1:37:38,  3.57it/s]
 94%|█████████▍| 350554/371472 [5:44:13<1:38:26,  3.54it/s]
 94%|█████████▍| 350555/371472 [5:44:13<1:38:56,  3.52it/s]
 94%|█████████▍| 350556/371472 [5:44:13<1:35:05,  3.67it/s]
 94%|█████████▍| 350557/371472 [5:44:14<1:33:35,  3.72it/s]
 94%|█████████▍| 350558/371472 [5:44:14<1:34:52,  3.67it/s]
 94%|█████████▍| 350559/371472 [5:44:14<1:34:36,  3.68it/s]
 94%|█████████▍| 350560/371472 [5:44:14<1:34:02,  3.71it/s]
                                                           
{'loss': 2.6883, 'learning_rate': 1.5069275356073178e-07, 'epoch': 15.1}
+
 94%|█████████▍| 350560/371472 [5:44:14<1:34:02,  3.71it/s]
 94%|█████████▍| 350561/371472 [5:44:15<1:34:55,  3.67it/s]
 94%|█████████▍| 350562/371472 [5:44:15<1:34:55,  3.67it/s]
 94%|█████████▍| 350563/371472 [5:44:15<1:36:06,  3.63it/s]
 94%|█████████▍| 350564/371472 [5:44:15<1:34:14,  3.70it/s]
 94%|█████████▍| 350565/371472 [5:44:16<1:38:30,  3.54it/s]
 94%|█████████▍| 350566/371472 [5:44:16<1:38:20,  3.54it/s]
 94%|█████████▍| 350567/371472 [5:44:16<1:38:14,  3.55it/s]
 94%|█████████▍| 350568/371472 [5:44:17<1:39:02,  3.52it/s]
 94%|█████████▍| 350569/371472 [5:44:17<1:41:40,  3.43it/s]
 94%|█████████▍| 350570/371472 [5:44:17<1:38:36,  3.53it/s]
 94%|█████████▍| 350571/371472 [5:44:17<1:39:57,  3.49it/s]
 94%|████████��▍| 350572/371472 [5:44:18<1:41:00,  3.45it/s]
 94%|█████████▍| 350573/371472 [5:44:18<1:40:56,  3.45it/s]
 94%|█████████▍| 350574/371472 [5:44:18<1:45:46,  3.29it/s]
 94%|█████████▍| 350575/371472 [5:44:19<1:42:47,  3.39it/s]
 94%|█████████▍| 350576/371472 [5:44:19<1:36:28,  3.61it/s]
 94%|█████████▍| 350577/371472 [5:44:19<1:33:50,  3.71it/s]
 94%|█████████▍| 350578/371472 [5:44:19<1:33:15,  3.73it/s]
 94%|█████████▍| 350579/371472 [5:44:20<1:35:26,  3.65it/s]
 94%|█████████▍| 350580/371472 [5:44:20<1:36:57,  3.59it/s]
                                                           
{'loss': 2.4369, 'learning_rate': 1.506442715852528e-07, 'epoch': 15.1}
+
 94%|█████████▍| 350580/371472 [5:44:20<1:36:57,  3.59it/s]
 94%|█████████▍| 350581/371472 [5:44:20<1:41:18,  3.44it/s]
 94%|█████████▍| 350582/371472 [5:44:21<1:43:08,  3.38it/s]
 94%|█████████▍| 350583/371472 [5:44:21<1:38:41,  3.53it/s]
 94%|█████████▍| 350584/371472 [5:44:21<1:42:53,  3.38it/s]
 94%|█████████▍| 350585/371472 [5:44:22<1:41:41,  3.42it/s]
 94%|█████████▍| 350586/371472 [5:44:22<1:50:20,  3.15it/s]
 94%|█████████▍| 350587/371472 [5:44:22<1:46:05,  3.28it/s]
 94%|█████████▍| 350588/371472 [5:44:22<1:40:29,  3.46it/s]
 94%|█████████▍| 350589/371472 [5:44:23<1:41:46,  3.42it/s]
 94%|█████████▍| 350590/371472 [5:44:23<1:40:55,  3.45it/s]
 94%|█████████▍| 350591/371472 [5:44:23<1:39:41,  3.49it/s]
 94%|█████████▍| 350592/371472 [5:44:24<1:38:41,  3.53it/s]
 94%|█████████▍| 350593/371472 [5:44:24<1:33:55,  3.70it/s]
 94%|█████████▍| 350594/371472 [5:44:24<1:30:18,  3.85it/s]
 94%|█████████▍| 350595/371472 [5:44:24<1:33:39,  3.71it/s]
 94%|█████████▍| 350596/371472 [5:44:25<1:34:47,  3.67it/s]
 94%|█████████▍| 350597/371472 [5:44:25<1:37:55,  3.55it/s]
 94%|█████████▍| 350598/371472 [5:44:25<1:49:55,  3.16it/s]
 94%|█████████▍| 350599/371472 [5:44:26<1:52:55,  3.08it/s]
 94%|█████████▍| 350600/371472 [5:44:26<1:55:47,  3.00it/s]
                                                           
{'loss': 2.4566, 'learning_rate': 1.5059578960977397e-07, 'epoch': 15.1}
+
 94%|█████████▍| 350600/371472 [5:44:26<1:55:47,  3.00it/s]
 94%|█████████▍| 350601/371472 [5:44:26<1:51:12,  3.13it/s]
 94%|█████████▍| 350602/371472 [5:44:27<1:47:39,  3.23it/s]
 94%|█████████▍| 350603/371472 [5:44:27<1:50:17,  3.15it/s]
 94%|█████████▍| 350604/371472 [5:44:27<1:46:19,  3.27it/s]
 94%|█████████▍| 350605/371472 [5:44:28<1:48:19,  3.21it/s]
 94%|█████████▍| 350606/371472 [5:44:28<1:42:38,  3.39it/s]
 94%|█████████▍| 350607/371472 [5:44:28<1:43:48,  3.35it/s]
 94%|█████████▍| 350608/371472 [5:44:28<1:38:38,  3.53it/s]
 94%|█████████▍| 350609/371472 [5:44:29<1:39:30,  3.49it/s]
 94%|█████████▍| 350610/371472 [5:44:29<1:43:59,  3.34it/s]
 94%|█████████▍| 350611/371472 [5:44:29<1:44:11,  3.34it/s]
 94%|█████████▍| 350612/371472 [5:44:29<1:40:03,  3.47it/s]
 94%|█████████▍| 350613/371472 [5:44:30<1:38:07,  3.54it/s]
 94%|█████████▍| 350614/371472 [5:44:30<1:40:31,  3.46it/s]
 94%|█████████▍| 350615/371472 [5:44:30<1:36:48,  3.59it/s]
 94%|█████████▍| 350616/371472 [5:44:31<1:38:15,  3.54it/s]
 94%|█████████▍| 350617/371472 [5:44:31<1:43:21,  3.36it/s]
 94%|█████████▍| 350618/371472 [5:44:31<1:39:52,  3.48it/s]
 94%|█████████▍| 350619/371472 [5:44:31<1:38:16,  3.54it/s]
 94%|█████████▍| 350620/371472 [5:44:32<1:50:09,  3.15it/s]
                                                           
{'loss': 2.6296, 'learning_rate': 1.5054730763429504e-07, 'epoch': 15.1}
+
 94%|█████████▍| 350620/371472 [5:44:32<1:50:09,  3.15it/s]
 94%|█████████▍| 350621/371472 [5:44:32<1:43:25,  3.36it/s]
 94%|█████████▍| 350622/371472 [5:44:32<1:42:11,  3.40it/s]
 94%|█████████▍| 350623/371472 [5:44:33<1:43:27,  3.36it/s]
 94%|█████████▍| 350624/371472 [5:44:33<1:40:07,  3.47it/s]
 94%|█████████▍| 350625/371472 [5:44:33<1:43:59,  3.34it/s]
 94%|█████████▍| 350626/371472 [5:44:34<1:39:11,  3.50it/s]
 94%|█████████▍| 350627/371472 [5:44:34<1:38:39,  3.52it/s]
 94%|███████���█▍| 350628/371472 [5:44:34<1:35:18,  3.64it/s]
 94%|█████████▍| 350629/371472 [5:44:34<1:40:58,  3.44it/s]
 94%|█████████▍| 350630/371472 [5:44:35<1:37:27,  3.56it/s]
 94%|█████████▍| 350631/371472 [5:44:35<1:33:46,  3.70it/s]
 94%|█████████▍| 350632/371472 [5:44:35<1:37:36,  3.56it/s]
 94%|█████████▍| 350633/371472 [5:44:36<1:47:38,  3.23it/s]
 94%|█████████▍| 350634/371472 [5:44:36<1:40:32,  3.45it/s]
 94%|█████████▍| 350635/371472 [5:44:36<1:39:49,  3.48it/s]
 94%|█████████▍| 350636/371472 [5:44:36<1:37:31,  3.56it/s]
 94%|█████████▍| 350637/371472 [5:44:37<1:33:44,  3.70it/s]
 94%|█████████▍| 350638/371472 [5:44:37<1:34:20,  3.68it/s]
 94%|█████████▍| 350639/371472 [5:44:37<1:31:58,  3.78it/s]
 94%|█████████▍| 350640/371472 [5:44:37<1:36:29,  3.60it/s]
                                                           
{'loss': 2.5365, 'learning_rate': 1.504988256588162e-07, 'epoch': 15.1}
+
 94%|█████████▍| 350640/371472 [5:44:37<1:36:29,  3.60it/s]
 94%|█████████▍| 350641/371472 [5:44:38<1:36:16,  3.61it/s]
 94%|█████████▍| 350642/371472 [5:44:38<1:36:12,  3.61it/s]
 94%|█████████▍| 350643/371472 [5:44:38<1:35:06,  3.65it/s]
 94%|█████████▍| 350644/371472 [5:44:39<1:32:30,  3.75it/s]
 94%|█████████▍| 350645/371472 [5:44:39<1:37:10,  3.57it/s]
 94%|█████████▍| 350646/371472 [5:44:39<1:36:06,  3.61it/s]
 94%|█████████▍| 350647/371472 [5:44:39<1:38:02,  3.54it/s]
 94%|█████████▍| 350648/371472 [5:44:40<1:32:53,  3.74it/s]
 94%|█████████▍| 350649/371472 [5:44:40<1:32:30,  3.75it/s]
 94%|█████████▍| 350650/371472 [5:44:40<1:34:25,  3.68it/s]
 94%|█████████▍| 350651/371472 [5:44:40<1:34:29,  3.67it/s]
 94%|█████████▍| 350652/371472 [5:44:41<1:34:46,  3.66it/s]
 94%|█████████▍| 350653/371472 [5:44:41<1:37:13,  3.57it/s]
 94%|█████████▍| 350654/371472 [5:44:41<1:38:46,  3.51it/s]
 94%|█████████▍| 350655/371472 [5:44:42<1:46:22,  3.26it/s]
 94%|█████████▍| 350656/371472 [5:44:42<1:48:40,  3.19it/s]
 94%|█████████▍| 350657/371472 [5:44:42<1:48:20,  3.20it/s]
 94%|█████████▍| 350658/371472 [5:44:43<1:42:09,  3.40it/s]
 94%|█████████▍| 350659/371472 [5:44:43<1:36:58,  3.58it/s]
 94%|█████████▍| 350660/371472 [5:44:43<1:33:48,  3.70it/s]
                                                           
{'loss': 2.63, 'learning_rate': 1.5045034368333727e-07, 'epoch': 15.1}
+
 94%|█████████▍| 350660/371472 [5:44:43<1:33:48,  3.70it/s]
 94%|█████████▍| 350661/371472 [5:44:43<1:37:37,  3.55it/s]
 94%|█████████▍| 350662/371472 [5:44:44<1:35:12,  3.64it/s]
 94%|█████████▍| 350663/371472 [5:44:44<1:34:32,  3.67it/s]
 94%|█████████▍| 350664/371472 [5:44:44<1:32:11,  3.76it/s]
 94%|█████████▍| 350665/371472 [5:44:44<1:32:27,  3.75it/s]
 94%|█████████▍| 350666/371472 [5:44:45<1:31:53,  3.77it/s]
 94%|█████████▍| 350667/371472 [5:44:45<1:32:01,  3.77it/s]
 94%|█████████▍| 350668/371472 [5:44:45<1:32:26,  3.75it/s]
 94%|█████████▍| 350669/371472 [5:44:46<1:42:16,  3.39it/s]
 94%|█████████▍| 350670/371472 [5:44:46<1:38:24,  3.52it/s]
 94%|█████████▍| 350671/371472 [5:44:46<1:41:12,  3.43it/s]
 94%|█████████▍| 350672/371472 [5:44:46<1:38:46,  3.51it/s]
 94%|█████████▍| 350673/371472 [5:44:47<1:37:38,  3.55it/s]
 94%|█████████▍| 350674/371472 [5:44:47<1:34:42,  3.66it/s]
 94%|█████████▍| 350675/371472 [5:44:47<1:37:16,  3.56it/s]
 94%|█████████▍| 350676/371472 [5:44:48<1:37:00,  3.57it/s]
 94%|█████████▍| 350677/371472 [5:44:48<1:34:05,  3.68it/s]
 94%|█████████▍| 350678/371472 [5:44:48<1:36:43,  3.58it/s]
 94%|█████████▍| 350679/371472 [5:44:48<1:35:50,  3.62it/s]
 94%|█████████▍| 350680/371472 [5:44:49<1:34:29,  3.67it/s]
                                                           
{'loss': 2.5121, 'learning_rate': 1.5040186170785842e-07, 'epoch': 15.1}
+
 94%|█████████▍| 350680/371472 [5:44:49<1:34:29,  3.67it/s]
 94%|█████████▍| 350681/371472 [5:44:49<1:36:40,  3.58it/s]
 94%|█████████▍| 350682/371472 [5:44:49<1:36:31,  3.59it/s]
 94%|█████████▍| 350683/371472 [5:44:49<1:35:10,  3.64it/s]
 94%|██████��██▍| 350684/371472 [5:44:50<1:40:09,  3.46it/s]
 94%|█████████▍| 350685/371472 [5:44:50<1:38:49,  3.51it/s]
 94%|█████████▍| 350686/371472 [5:44:50<1:40:39,  3.44it/s]
 94%|█████████▍| 350687/371472 [5:44:51<1:35:08,  3.64it/s]
 94%|█████████▍| 350688/371472 [5:44:51<1:36:38,  3.58it/s]
 94%|█████████▍| 350689/371472 [5:44:51<1:33:59,  3.68it/s]
 94%|█████████▍| 350690/371472 [5:44:51<1:35:38,  3.62it/s]
 94%|█████████▍| 350691/371472 [5:44:52<1:35:41,  3.62it/s]
 94%|█████████▍| 350692/371472 [5:44:52<1:35:49,  3.61it/s]
 94%|█████████▍| 350693/371472 [5:44:52<1:35:39,  3.62it/s]
 94%|█████████▍| 350694/371472 [5:44:53<1:32:50,  3.73it/s]
 94%|█████████▍| 350695/371472 [5:44:53<1:34:06,  3.68it/s]
 94%|█████████▍| 350696/371472 [5:44:53<1:45:14,  3.29it/s]
 94%|█████████▍| 350697/371472 [5:44:53<1:39:04,  3.50it/s]
 94%|█████████▍| 350698/371472 [5:44:54<1:46:16,  3.26it/s]
 94%|█████████▍| 350699/371472 [5:44:54<1:41:57,  3.40it/s]
 94%|█████████▍| 350700/371472 [5:44:54<1:40:56,  3.43it/s]
                                                           
{'loss': 2.5946, 'learning_rate': 1.5035337973237946e-07, 'epoch': 15.11}
+
 94%|█████████▍| 350700/371472 [5:44:54<1:40:56,  3.43it/s]
 94%|█████████▍| 350701/371472 [5:44:55<1:38:31,  3.51it/s]
 94%|█████████▍| 350702/371472 [5:44:55<1:35:30,  3.62it/s]
 94%|█████████▍| 350703/371472 [5:44:55<1:32:34,  3.74it/s]
 94%|█████████▍| 350704/371472 [5:44:55<1:32:37,  3.74it/s]
 94%|█████████▍| 350705/371472 [5:44:56<1:31:55,  3.77it/s]
 94%|█████████▍| 350706/371472 [5:44:56<1:33:28,  3.70it/s]
 94%|█████████▍| 350707/371472 [5:44:56<1:34:09,  3.68it/s]
 94%|█████████▍| 350708/371472 [5:44:56<1:36:09,  3.60it/s]
 94%|█████████▍| 350709/371472 [5:44:57<1:33:15,  3.71it/s]
 94%|█████████▍| 350710/371472 [5:44:57<1:32:35,  3.74it/s]
 94%|█████████▍| 350711/371472 [5:44:57<1:38:36,  3.51it/s]
 94%|█████████▍| 350712/371472 [5:44:58<1:37:31,  3.55it/s]
 94%|█████████▍| 350713/371472 [5:44:58<1:46:57,  3.23it/s]
 94%|█████████▍| 350714/371472 [5:44:58<1:42:36,  3.37it/s]
 94%|█████████▍| 350715/371472 [5:44:59<1:41:04,  3.42it/s]
 94%|█████████▍| 350716/371472 [5:44:59<1:38:07,  3.53it/s]
 94%|█████████▍| 350717/371472 [5:44:59<1:40:31,  3.44it/s]
 94%|█████████▍| 350718/371472 [5:44:59<1:40:35,  3.44it/s]
 94%|█████████▍| 350719/371472 [5:45:00<1:37:59,  3.53it/s]
 94%|█████████▍| 350720/371472 [5:45:00<1:43:17,  3.35it/s]
                                                           
{'loss': 2.5452, 'learning_rate': 1.5030489775690064e-07, 'epoch': 15.11}
+
 94%|█████████▍| 350720/371472 [5:45:00<1:43:17,  3.35it/s]
 94%|█████████▍| 350721/371472 [5:45:00<1:41:47,  3.40it/s]
 94%|█████████▍| 350722/371472 [5:45:01<1:39:51,  3.46it/s]
 94%|█████████▍| 350723/371472 [5:45:01<1:37:43,  3.54it/s]
 94%|█████████▍| 350724/371472 [5:45:01<1:35:09,  3.63it/s]
 94%|█████████▍| 350725/371472 [5:45:01<1:41:52,  3.39it/s]
 94%|█████████▍| 350726/371472 [5:45:02<1:37:55,  3.53it/s]
 94%|█████████▍| 350727/371472 [5:45:02<1:36:32,  3.58it/s]
 94%|█████████▍| 350728/371472 [5:45:02<1:38:48,  3.50it/s]
 94%|█████████▍| 350729/371472 [5:45:03<1:51:10,  3.11it/s]
 94%|█████████▍| 350730/371472 [5:45:03<1:45:54,  3.26it/s]
 94%|█████████▍| 350731/371472 [5:45:03<1:40:36,  3.44it/s]
 94%|█████████▍| 350732/371472 [5:45:03<1:38:09,  3.52it/s]
 94%|█████████▍| 350733/371472 [5:45:04<1:35:14,  3.63it/s]
 94%|█████████▍| 350734/371472 [5:45:04<1:35:26,  3.62it/s]
 94%|█████████▍| 350735/371472 [5:45:04<1:31:55,  3.76it/s]
 94%|█████████▍| 350736/371472 [5:45:05<1:35:39,  3.61it/s]
 94%|█████████▍| 350737/371472 [5:45:05<1:32:19,  3.74it/s]
 94%|█████████▍| 350738/371472 [5:45:05<1:30:16,  3.83it/s]
 94%|█████████▍| 350739/371472 [5:45:05<1:30:46,  3.81it/s]
 94%|█████████▍| 350740/371472 [5:45:06<1:34:32,  3.65it/s]
                                                           
{'loss': 2.5496, 'learning_rate': 1.5025641578142168e-07, 'epoch': 15.11}
+
 94%|█████████▍| 350740/371472 [5:45:06<1:34:32,  3.65it/s]
 94%|█████████▍| 350741/371472 [5:45:06<1:39:18,  3.48it/s]
 94%|█████████▍| 350742/371472 [5:45:06<1:39:27,  3.47it/s]
 94%|█████████▍| 350743/371472 [5:45:06<1:38:36,  3.50it/s]
 94%|█████████▍| 350744/371472 [5:45:07<1:35:04,  3.63it/s]
 94%|█████████▍| 350745/371472 [5:45:07<1:35:22,  3.62it/s]
 94%|█████████▍| 350746/371472 [5:45:07<1:34:01,  3.67it/s]
 94%|█████████▍| 350747/371472 [5:45:08<1:34:05,  3.67it/s]
 94%|█████████▍| 350748/371472 [5:45:08<1:40:36,  3.43it/s]
 94%|█████████▍| 350749/371472 [5:45:08<1:43:32,  3.34it/s]
 94%|█████████▍| 350750/371472 [5:45:08<1:39:43,  3.46it/s]
 94%|█████████▍| 350751/371472 [5:45:09<1:43:00,  3.35it/s]
 94%|█████████▍| 350752/371472 [5:45:09<1:41:24,  3.41it/s]
 94%|█████████▍| 350753/371472 [5:45:09<1:38:44,  3.50it/s]
 94%|█████████▍| 350754/371472 [5:45:10<1:40:58,  3.42it/s]
 94%|█████████▍| 350755/371472 [5:45:10<1:50:33,  3.12it/s]
 94%|█████████▍| 350756/371472 [5:45:10<1:46:39,  3.24it/s]
 94%|█████████▍| 350757/371472 [5:45:11<1:50:33,  3.12it/s]
 94%|█████████▍| 350758/371472 [5:45:11<1:49:06,  3.16it/s]
 94%|█████████▍| 350759/371472 [5:45:11<1:42:15,  3.38it/s]
 94%|█████████▍| 350760/371472 [5:45:11<1:37:28,  3.54it/s]
                                                           
{'loss': 2.6479, 'learning_rate': 1.5020793380594286e-07, 'epoch': 15.11}
+
 94%|█████████▍| 350760/371472 [5:45:11<1:37:28,  3.54it/s]
 94%|█████████▍| 350761/371472 [5:45:12<1:38:17,  3.51it/s]
 94%|█████████▍| 350762/371472 [5:45:12<1:34:47,  3.64it/s]
 94%|█████████▍| 350763/371472 [5:45:12<1:34:36,  3.65it/s]
 94%|█████████▍| 350764/371472 [5:45:13<1:38:36,  3.50it/s]
 94%|█████████▍| 350765/371472 [5:45:13<1:37:11,  3.55it/s]
 94%|█████████▍| 350766/371472 [5:45:13<1:33:16,  3.70it/s]
 94%|█████████▍| 350767/371472 [5:45:13<1:38:10,  3.52it/s]
 94%|█████████▍| 350768/371472 [5:45:14<1:37:58,  3.52it/s]
 94%|█████████▍| 350769/371472 [5:45:14<1:36:52,  3.56it/s]
 94%|█████████▍| 350770/371472 [5:45:14<1:37:12,  3.55it/s]
 94%|█████████▍| 350771/371472 [5:45:15<1:38:11,  3.51it/s]
 94%|█████████▍| 350772/371472 [5:45:15<1:36:55,  3.56it/s]
 94%|█████████▍| 350773/371472 [5:45:15<1:37:13,  3.55it/s]
 94%|█████████▍| 350774/371472 [5:45:15<1:36:45,  3.56it/s]
 94%|█████████▍| 350775/371472 [5:45:16<1:36:25,  3.58it/s]
 94%|█████████▍| 350776/371472 [5:45:16<1:34:57,  3.63it/s]
 94%|█████████▍| 350777/371472 [5:45:16<1:35:54,  3.60it/s]
 94%|█████████▍| 350778/371472 [5:45:16<1:33:34,  3.69it/s]
 94%|█████████▍| 350779/371472 [5:45:17<1:35:35,  3.61it/s]
 94%|█████████▍| 350780/371472 [5:45:17<1:49:28,  3.15it/s]
                                                           
{'loss': 2.5785, 'learning_rate': 1.501594518304639e-07, 'epoch': 15.11}
+
 94%|█████████▍| 350780/371472 [5:45:17<1:49:28,  3.15it/s]
 94%|█████████▍| 350781/371472 [5:45:17<1:47:31,  3.21it/s]
 94%|█████████▍| 350782/371472 [5:45:18<1:53:19,  3.04it/s]
 94%|█████████▍| 350783/371472 [5:45:18<1:45:48,  3.26it/s]
 94%|█████████▍| 350784/371472 [5:45:18<1:42:08,  3.38it/s]
 94%|█████████▍| 350785/371472 [5:45:19<1:44:33,  3.30it/s]
 94%|█████████▍| 350786/371472 [5:45:19<1:44:00,  3.31it/s]
 94%|█████████▍| 350787/371472 [5:45:19<1:47:30,  3.21it/s]
 94%|█████████▍| 350788/371472 [5:45:20<1:47:25,  3.21it/s]
 94%|█████████▍| 350789/371472 [5:45:20<1:43:04,  3.34it/s]
 94%|█████████▍| 350790/371472 [5:45:20<1:39:30,  3.46it/s]
 94%|█████████▍| 350791/371472 [5:45:20<1:38:16,  3.51it/s]
 94%|█████████▍| 350792/371472 [5:45:21<1:37:44,  3.53it/s]
 94%|█████████▍| 350793/371472 [5:45:21<1:34:59,  3.63it/s]
 94%|█████████▍| 350794/371472 [5:45:21<1:32:24,  3.73it/s]
 94%|█████████▍| 350795/371472 [5:45:21<1:30:37,  3.80it/s]
 94%|█████████▍| 350796/371472 [5:45:22<1:32:44,  3.72it/s]
 94%|█████████▍| 350797/371472 [5:45:22<1:31:05,  3.78it/s]
 94%|█████████▍| 350798/371472 [5:45:22<1:33:42,  3.68it/s]
 94%|█████████▍| 350799/371472 [5:45:23<1:33:20,  3.69it/s]
 94%|█████████▍| 350800/371472 [5:45:23<1:33:38,  3.68it/s]
                                                           
{'loss': 2.7014, 'learning_rate': 1.5011096985498506e-07, 'epoch': 15.11}
+
 94%|█████████▍| 350800/371472 [5:45:23<1:33:38,  3.68it/s]
 94%|█████████▍| 350801/371472 [5:45:23<1:36:35,  3.57it/s]
 94%|█████████▍| 350802/371472 [5:45:23<1:42:43,  3.35it/s]
 94%|█████████▍| 350803/371472 [5:45:24<1:38:28,  3.50it/s]
 94%|█████████▍| 350804/371472 [5:45:24<1:40:51,  3.42it/s]
 94%|█████████▍| 350805/371472 [5:45:24<1:45:42,  3.26it/s]
 94%|█████████▍| 350806/371472 [5:45:25<1:42:41,  3.35it/s]
 94%|█████████▍| 350807/371472 [5:45:25<1:45:17,  3.27it/s]
 94%|█████████▍| 350808/371472 [5:45:25<1:44:20,  3.30it/s]
 94%|█████████▍| 350809/371472 [5:45:26<1:39:23,  3.46it/s]
 94%|█████████▍| 350810/371472 [5:45:26<1:47:08,  3.21it/s]
 94%|█████████▍| 350811/371472 [5:45:26<1:43:41,  3.32it/s]
 94%|█████████▍| 350812/371472 [5:45:26<1:37:45,  3.52it/s]
 94%|█████████▍| 350813/371472 [5:45:27<1:35:39,  3.60it/s]
 94%|█████████▍| 350814/371472 [5:45:27<1:38:11,  3.51it/s]
 94%|█████████▍| 350815/371472 [5:45:27<1:41:51,  3.38it/s]
 94%|█████████▍| 350816/371472 [5:45:28<1:41:51,  3.38it/s]
 94%|█████████▍| 350817/371472 [5:45:28<1:44:28,  3.30it/s]
 94%|█████████▍| 350818/371472 [5:45:28<1:40:50,  3.41it/s]
 94%|█████████▍| 350819/371472 [5:45:28<1:37:10,  3.54it/s]
 94%|█████████▍| 350820/371472 [5:45:29<1:35:59,  3.59it/s]
                                                           
{'loss': 2.3552, 'learning_rate': 1.500624878795061e-07, 'epoch': 15.11}
+
 94%|█████████▍| 350820/371472 [5:45:29<1:35:59,  3.59it/s]
 94%|█████████▍| 350821/371472 [5:45:29<1:36:36,  3.56it/s]
 94%|█████████▍| 350822/371472 [5:45:29<1:32:28,  3.72it/s]
 94%|█████████▍| 350823/371472 [5:45:30<1:33:43,  3.67it/s]
 94%|█████████▍| 350824/371472 [5:45:30<1:35:13,  3.61it/s]
 94%|█████████▍| 350825/371472 [5:45:30<1:36:35,  3.56it/s]
 94%|█████████▍| 350826/371472 [5:45:30<1:35:26,  3.61it/s]
 94%|█████████▍| 350827/371472 [5:45:31<1:32:02,  3.74it/s]
 94%|█████████▍| 350828/371472 [5:45:31<1:31:45,  3.75it/s]
 94%|█████████▍| 350829/371472 [5:45:31<1:31:51,  3.75it/s]
 94%|█████████▍| 350830/371472 [5:45:31<1:28:38,  3.88it/s]
 94%|█████████▍| 350831/371472 [5:45:32<1:35:03,  3.62it/s]
 94%|█████████▍| 350832/371472 [5:45:32<1:40:23,  3.43it/s]
 94%|█████████▍| 350833/371472 [5:45:32<1:37:13,  3.54it/s]
 94%|█████████▍| 350834/371472 [5:45:33<1:37:38,  3.52it/s]
 94%|█████████▍| 350835/371472 [5:45:33<1:38:17,  3.50it/s]
 94%|█████████▍| 350836/371472 [5:45:33<1:36:35,  3.56it/s]
 94%|█████████▍| 350837/371472 [5:45:33<1:46:09,  3.24it/s]
 94%|█████████▍| 350838/371472 [5:45:34<1:40:55,  3.41it/s]
 94%|█████████▍| 350839/371472 [5:45:34<1:44:23,  3.29it/s]
 94%|█████████▍| 350840/371472 [5:45:34<1:45:18,  3.27it/s]
                                                           
{'loss': 2.5884, 'learning_rate': 1.5001400590402728e-07, 'epoch': 15.11}
+
 94%|█████████▍| 350840/371472 [5:45:34<1:45:18,  3.27it/s]
 94%|█████████▍| 350841/371472 [5:45:35<1:40:09,  3.43it/s]
 94%|█████████▍| 350842/371472 [5:45:35<1:44:06,  3.30it/s]
 94%|█████████▍| 350843/371472 [5:45:35<1:43:36,  3.32it/s]
 94%|█████████▍| 350844/371472 [5:45:36<1:41:09,  3.40it/s]
 94%|█████████▍| 350845/371472 [5:45:36<1:39:08,  3.47it/s]
 94%|█████████▍| 350846/371472 [5:45:36<1:41:09,  3.40it/s]
 94%|█████████▍| 350847/371472 [5:45:36<1:38:56,  3.47it/s]
 94%|█████████▍| 350848/371472 [5:45:37<1:40:27,  3.42it/s]
 94%|█████████▍| 350849/371472 [5:45:37<1:40:46,  3.41it/s]
 94%|█████████▍| 350850/371472 [5:45:37<1:40:19,  3.43it/s]
 94%|█████████▍| 350851/371472 [5:45:38<1:37:41,  3.52it/s]
 94%|█████████▍| 350852/371472 [5:45:38<1:40:39,  3.41it/s]
 94%|█████████▍| 350853/371472 [5:45:38<1:46:08,  3.24it/s]
 94%|█████████▍| 350854/371472 [5:45:38<1:42:08,  3.36it/s]
 94%|█████████▍| 350855/371472 [5:45:39<1:42:09,  3.36it/s]
 94%|█████████▍| 350856/371472 [5:45:39<1:39:18,  3.46it/s]
 94%|█████████▍| 350857/371472 [5:45:39<1:39:22,  3.46it/s]
 94%|█████████▍| 350858/371472 [5:45:40<1:37:54,  3.51it/s]
 94%|█████████▍| 350859/371472 [5:45:40<1:35:38,  3.59it/s]
 94%|█████████▍| 350860/371472 [5:45:40<1:33:44,  3.66it/s]
                                                           
{'loss': 2.4191, 'learning_rate': 1.4996552392854833e-07, 'epoch': 15.11}
+
 94%|█████████▍| 350860/371472 [5:45:40<1:33:44,  3.66it/s]
 94%|█████████▍| 350861/371472 [5:45:40<1:34:26,  3.64it/s]
 94%|█████████▍| 350862/371472 [5:45:41<1:32:08,  3.73it/s]
 94%|█████████▍| 350863/371472 [5:45:41<1:33:27,  3.68it/s]
 94%|█████████▍| 350864/371472 [5:45:41<1:31:47,  3.74it/s]
 94%|█████████▍| 350865/371472 [5:45:42<1:37:25,  3.53it/s]
 94%|█████████▍| 350866/371472 [5:45:42<1:35:08,  3.61it/s]
 94%|█████████▍| 350867/371472 [5:45:42<1:35:48,  3.58it/s]
 94%|█████████▍| 350868/371472 [5:45:42<1:33:57,  3.65it/s]
 94%|█████████▍| 350869/371472 [5:45:43<1:34:20,  3.64it/s]
 94%|█████████▍| 350870/371472 [5:45:43<1:33:12,  3.68it/s]
 94%|█████████▍| 350871/371472 [5:45:43<1:35:03,  3.61it/s]
 94%|█████████▍| 350872/371472 [5:45:43<1:35:10,  3.61it/s]
 94%|█████████▍| 350873/371472 [5:45:44<1:33:57,  3.65it/s]
 94%|█████████▍| 350874/371472 [5:45:44<1:34:22,  3.64it/s]
 94%|█████████▍| 350875/371472 [5:45:44<1:34:38,  3.63it/s]
 94%|█████████▍| 350876/371472 [5:45:45<1:34:30,  3.63it/s]
 94%|█████████▍| 350877/371472 [5:45:45<1:33:21,  3.68it/s]
 94%|█████████▍| 350878/371472 [5:45:45<1:35:37,  3.59it/s]
 94%|█████████▍| 350879/371472 [5:45:45<1:33:38,  3.66it/s]
 94%|█████████▍| 350880/371472 [5:45:46<1:35:19,  3.60it/s]
                                                           
{'loss': 2.638, 'learning_rate': 1.4991704195306948e-07, 'epoch': 15.11}
+
 94%|█████████▍| 350880/371472 [5:45:46<1:35:19,  3.60it/s]
 94%|█████████▍| 350881/371472 [5:45:46<1:31:57,  3.73it/s]
 94%|█████████▍| 350882/371472 [5:45:46<1:33:58,  3.65it/s]
 94%|█████████▍| 350883/371472 [5:45:46<1:31:27,  3.75it/s]
 94%|█████████▍| 350884/371472 [5:45:47<1:37:44,  3.51it/s]
 94%|█████████▍| 350885/371472 [5:45:47<1:34:38,  3.63it/s]
 94%|█████████▍| 350886/371472 [5:45:47<1:35:33,  3.59it/s]
 94%|█████████▍| 350887/371472 [5:45:48<1:37:57,  3.50it/s]
 94%|█████████▍| 350888/371472 [5:45:48<1:38:50,  3.47it/s]
 94%|█████████▍| 350889/371472 [5:45:48<1:42:29,  3.35it/s]
 94%|█████████▍| 350890/371472 [5:45:49<1:40:41,  3.41it/s]
 94%|█████████▍| 350891/371472 [5:45:49<1:39:02,  3.46it/s]
 94%|█████████▍| 350892/371472 [5:45:49<1:36:54,  3.54it/s]
 94%|█████████▍| 350893/371472 [5:45:49<1:49:41,  3.13it/s]
 94%|█████████▍| 350894/371472 [5:45:50<1:47:58,  3.18it/s]
 94%|█████████▍| 350895/371472 [5:45:50<1:42:21,  3.35it/s]
 94%|█████████▍| 350896/371472 [5:45:50<1:41:34,  3.38it/s]
 94%|█████████▍| 350897/371472 [5:45:51<1:37:34,  3.51it/s]
 94%|█████████▍| 350898/371472 [5:45:51<1:37:30,  3.52it/s]
 94%|█████████▍| 350899/371472 [5:45:51<1:36:56,  3.54it/s]
 94%|█████████▍| 350900/371472 [5:45:51<1:38:35,  3.48it/s]
                                                           
{'loss': 2.6478, 'learning_rate': 1.4986855997759055e-07, 'epoch': 15.11}
+
 94%|█████████▍| 350900/371472 [5:45:51<1:38:35,  3.48it/s]
 94%|█████████▍| 350901/371472 [5:45:52<1:34:29,  3.63it/s]
 94%|█████████▍| 350902/371472 [5:45:52<1:41:39,  3.37it/s]
 94%|█████████▍| 350903/371472 [5:45:52<1:46:06,  3.23it/s]
 94%|█████████▍| 350904/371472 [5:45:53<1:48:26,  3.16it/s]
 94%|█████████▍| 350905/371472 [5:45:53<1:41:13,  3.39it/s]
 94%|█████████▍| 350906/371472 [5:45:53<1:39:38,  3.44it/s]
 94%|█████████▍| 350907/371472 [5:45:54<1:39:43,  3.44it/s]
 94%|█████████▍| 350908/371472 [5:45:54<1:39:22,  3.45it/s]
 94%|█████████▍| 350909/371472 [5:45:54<1:37:26,  3.52it/s]
 94%|█████████▍| 350910/371472 [5:45:54<1:40:44,  3.40it/s]
 94%|█████████▍| 350911/371472 [5:45:55<1:40:53,  3.40it/s]
 94%|█████████▍| 350912/371472 [5:45:55<1:40:55,  3.40it/s]
 94%|█████████▍| 350913/371472 [5:45:55<1:46:16,  3.22it/s]
 94%|█████████▍| 350914/371472 [5:45:56<1:52:04,  3.06it/s]
 94%|█████████▍| 350915/371472 [5:45:56<1:51:04,  3.08it/s]
 94%|█████████▍| 350916/371472 [5:45:56<1:46:50,  3.21it/s]
 94%|█████████▍| 350917/371472 [5:45:57<1:40:54,  3.39it/s]
 94%|█████████▍| 350918/371472 [5:45:57<1:39:10,  3.45it/s]
 94%|█████████▍| 350919/371472 [5:45:57<1:42:58,  3.33it/s]
 94%|█████████▍| 350920/371472 [5:45:57<1:40:13,  3.42it/s]
                                                           
{'loss': 2.5153, 'learning_rate': 1.4982007800211165e-07, 'epoch': 15.11}
+
 94%|█████████▍| 350920/371472 [5:45:57<1:40:13,  3.42it/s]
 94%|█████████▍| 350921/371472 [5:45:58<1:38:32,  3.48it/s]
 94%|█████████▍| 350922/371472 [5:45:58<1:37:26,  3.51it/s]
 94%|█████████▍| 350923/371472 [5:45:58<1:41:07,  3.39it/s]
 94%|█████████▍| 350924/371472 [5:45:59<1:39:28,  3.44it/s]
 94%|█████████▍| 350925/371472 [5:45:59<1:41:39,  3.37it/s]
 94%|█████████▍| 350926/371472 [5:45:59<1:40:03,  3.42it/s]
 94%|█████████▍| 350927/371472 [5:45:59<1:41:57,  3.36it/s]
 94%|█████████▍| 350928/371472 [5:46:00<1:44:47,  3.27it/s]
 94%|█████████▍| 350929/371472 [5:46:00<1:41:14,  3.38it/s]
 94%|█████████▍| 350930/371472 [5:46:00<1:44:50,  3.27it/s]
 94%|█████████▍| 350931/371472 [5:46:01<1:43:28,  3.31it/s]
 94%|█████████▍| 350932/371472 [5:46:01<1:38:30,  3.48it/s]
 94%|█████████▍| 350933/371472 [5:46:01<1:37:04,  3.53it/s]
 94%|█████████▍| 350934/371472 [5:46:02<1:40:05,  3.42it/s]
 94%|█████████▍| 350935/371472 [5:46:02<1:35:55,  3.57it/s]
 94%|█████████▍| 350936/371472 [5:46:02<1:30:51,  3.77it/s]
 94%|█████████▍| 350937/371472 [5:46:02<1:38:41,  3.47it/s]
 94%|█████████▍| 350938/371472 [5:46:03<1:46:51,  3.20it/s]
 94%|█████████▍| 350939/371472 [5:46:03<1:40:59,  3.39it/s]
 94%|█████████▍| 350940/371472 [5:46:03<1:42:14,  3.35it/s]
                                                           
{'loss': 2.6055, 'learning_rate': 1.4977159602663277e-07, 'epoch': 15.12}
+
 94%|█████████▍| 350940/371472 [5:46:03<1:42:14,  3.35it/s]
 94%|█████████▍| 350941/371472 [5:46:04<1:41:40,  3.37it/s]
 94%|█████████▍| 350942/371472 [5:46:04<1:36:31,  3.54it/s]
 94%|█████████▍| 350943/371472 [5:46:04<1:44:56,  3.26it/s]
 94%|█████████▍| 350944/371472 [5:46:04<1:42:19,  3.34it/s]
 94%|█████████▍| 350945/371472 [5:46:05<1:39:42,  3.43it/s]
 94%|█████████▍| 350946/371472 [5:46:05<1:42:10,  3.35it/s]
 94%|█████████▍| 350947/371472 [5:46:05<1:45:20,  3.25it/s]
 94%|█████████▍| 350948/371472 [5:46:06<1:40:58,  3.39it/s]
 94%|█████████▍| 350949/371472 [5:46:06<1:40:32,  3.40it/s]
 94%|█████████▍| 350950/371472 [5:46:06<1:38:52,  3.46it/s]
 94%|█████████▍| 350951/371472 [5:46:06<1:33:04,  3.67it/s]
 94%|█████████▍| 350952/371472 [5:46:07<1:33:45,  3.65it/s]
 94%|█████████▍| 350953/371472 [5:46:07<1:32:04,  3.71it/s]
 94%|█████████▍| 350954/371472 [5:46:07<1:32:20,  3.70it/s]
 94%|█████████▍| 350955/371472 [5:46:08<1:35:51,  3.57it/s]
 94%|█████████▍| 350956/371472 [5:46:08<1:40:08,  3.41it/s]
 94%|█████████▍| 350957/371472 [5:46:08<1:39:27,  3.44it/s]
 94%|█████████▍| 350958/371472 [5:46:08<1:38:36,  3.47it/s]
 94%|█████████▍| 350959/371472 [5:46:09<1:38:07,  3.48it/s]
 94%|█████████▍| 350960/371472 [5:46:09<1:35:01,  3.60it/s]
                                                           
{'loss': 2.5537, 'learning_rate': 1.4972311405115382e-07, 'epoch': 15.12}
+
 94%|█████████▍| 350960/371472 [5:46:09<1:35:01,  3.60it/s]
 94%|█████████▍| 350961/371472 [5:46:09<1:32:54,  3.68it/s]
 94%|█████████▍| 350962/371472 [5:46:10<1:34:16,  3.63it/s]
 94%|█████████▍| 350963/371472 [5:46:10<1:33:28,  3.66it/s]
 94%|█████████▍| 350964/371472 [5:46:10<1:35:30,  3.58it/s]
 94%|█████████▍| 350965/371472 [5:46:10<1:35:15,  3.59it/s]
 94%|█████████▍| 350966/371472 [5:46:11<1:33:51,  3.64it/s]
 94%|█████████▍| 350967/371472 [5:46:11<1:34:37,  3.61it/s]
 94%|█████████▍| 350968/371472 [5:46:11<1:36:34,  3.54it/s]
 94%|█████████▍| 350969/371472 [5:46:12<1:38:16,  3.48it/s]
 94%|█████████▍| 350970/371472 [5:46:12<1:34:33,  3.61it/s]
 94%|█████████▍| 350971/371472 [5:46:12<1:31:14,  3.74it/s]
 94%|█████████▍| 350972/371472 [5:46:12<1:33:16,  3.66it/s]
 94%|█████████▍| 350973/371472 [5:46:13<1:34:53,  3.60it/s]
 94%|█████████▍| 350974/371472 [5:46:13<1:32:17,  3.70it/s]
 94%|█████████▍| 350975/371472 [5:46:13<1:30:40,  3.77it/s]
 94%|█████████▍| 350976/371472 [5:46:13<1:31:38,  3.73it/s]
 94%|█████████▍| 350977/371472 [5:46:14<1:34:32,  3.61it/s]
 94%|█████████▍| 350978/371472 [5:46:14<1:32:34,  3.69it/s]
 94%|█████████▍| 350979/371472 [5:46:14<1:33:28,  3.65it/s]
 94%|█████████▍| 350980/371472 [5:46:14<1:30:33,  3.77it/s]
                                                           
{'loss': 2.4827, 'learning_rate': 1.49674632075675e-07, 'epoch': 15.12}
+
 94%|█████████▍| 350980/371472 [5:46:14<1:30:33,  3.77it/s]
 94%|█████████▍| 350981/371472 [5:46:15<1:28:38,  3.85it/s]
 94%|█████████▍| 350982/371472 [5:46:15<1:32:53,  3.68it/s]
 94%|█████████▍| 350983/371472 [5:46:15<1:39:22,  3.44it/s]
 94%|█████████▍| 350984/371472 [5:46:16<1:37:25,  3.51it/s]
 94%|█████████▍| 350985/371472 [5:46:16<1:35:33,  3.57it/s]
 94%|█████████▍| 350986/371472 [5:46:16<1:36:05,  3.55it/s]
 94%|█████████▍| 350987/371472 [5:46:16<1:34:44,  3.60it/s]
 94%|█████████▍| 350988/371472 [5:46:17<1:35:07,  3.59it/s]
 94%|█████████▍| 350989/371472 [5:46:17<1:49:11,  3.13it/s]
 94%|█████████▍| 350990/371472 [5:46:17<1:42:41,  3.32it/s]
 94%|█████████▍| 350991/371472 [5:46:18<1:40:00,  3.41it/s]
 94%|█████████▍| 350992/371472 [5:46:18<1:36:44,  3.53it/s]
 94%|█████████▍| 350993/371472 [5:46:18<1:39:38,  3.43it/s]
 94%|█████████▍| 350994/371472 [5:46:19<1:38:54,  3.45it/s]
 94%|█████████▍| 350995/371472 [5:46:19<1:37:13,  3.51it/s]
 94%|█████████▍| 350996/371472 [5:46:19<1:34:08,  3.63it/s]
 94%|█████████▍| 350997/371472 [5:46:19<1:35:50,  3.56it/s]
 94%|█████████▍| 350998/371472 [5:46:20<1:44:04,  3.28it/s]
 94%|█████████▍| 350999/371472 [5:46:20<1:42:37,  3.33it/s]
 94%|█████████▍| 351000/371472 [5:46:20<2:01:07,  2.82it/s]
                                                           
{'loss': 2.7124, 'learning_rate': 1.4962615010019604e-07, 'epoch': 15.12}
+
 94%|█████████▍| 351000/371472 [5:46:20<2:01:07,  2.82it/s]
 94%|█████████▍| 351001/371472 [5:46:21<1:59:07,  2.86it/s]
 94%|█████████▍| 351002/371472 [5:46:21<2:09:36,  2.63it/s]
 94%|█████████▍| 351003/371472 [5:46:22<1:55:19,  2.96it/s]
 94%|█████████▍| 351004/371472 [5:46:22<1:53:19,  3.01it/s]
 94%|█████████▍| 351005/371472 [5:46:22<1:44:02,  3.28it/s]
 94%|█████████▍| 351006/371472 [5:46:22<1:39:03,  3.44it/s]
 94%|█████████▍| 351007/371472 [5:46:23<1:37:19,  3.50it/s]
 94%|█████████▍| 351008/371472 [5:46:23<1:37:17,  3.51it/s]
 94%|█████████▍| 351009/371472 [5:46:23<1:37:29,  3.50it/s]
 94%|█████████▍| 351010/371472 [5:46:23<1:38:00,  3.48it/s]
 94%|█████████▍| 351011/371472 [5:46:24<1:45:59,  3.22it/s]
 94%|█████████▍| 351012/371472 [5:46:24<1:44:51,  3.25it/s]
 94%|█████████▍| 351013/371472 [5:46:24<1:39:51,  3.41it/s]
 94%|█████████▍| 351014/371472 [5:46:25<1:35:55,  3.55it/s]
 94%|█████████▍| 351015/371472 [5:46:25<1:54:09,  2.99it/s]
 94%|█████████▍| 351016/371472 [5:46:25<1:49:40,  3.11it/s]
 94%|█████████▍| 351017/371472 [5:46:26<1:44:45,  3.25it/s]
 94%|█████████▍| 351018/371472 [5:46:26<1:52:33,  3.03it/s]
 94%|█████████▍| 351019/371472 [5:46:26<1:47:23,  3.17it/s]
 94%|█████████▍| 351020/371472 [5:46:27<1:45:32,  3.23it/s]
                                                           
{'loss': 2.4571, 'learning_rate': 1.495776681247172e-07, 'epoch': 15.12}
+
 94%|█████████▍| 351020/371472 [5:46:27<1:45:32,  3.23it/s]
 94%|█████████▍| 351021/371472 [5:46:27<1:46:50,  3.19it/s]
 94%|█████████▍| 351022/371472 [5:46:27<1:44:49,  3.25it/s]
 94%|█████████▍| 351023/371472 [5:46:28<1:40:36,  3.39it/s]
 94%|█████████▍| 351024/371472 [5:46:28<1:37:12,  3.51it/s]
 94%|█████████▍| 351025/371472 [5:46:28<1:41:27,  3.36it/s]
 94%|█████████▍| 351026/371472 [5:46:28<1:37:49,  3.48it/s]
 94%|█████████▍| 351027/371472 [5:46:29<1:38:43,  3.45it/s]
 94%|█████████▍| 351028/371472 [5:46:29<1:40:52,  3.38it/s]
 94%|█████████▍| 351029/371472 [5:46:29<1:42:14,  3.33it/s]
 94%|█████████▍| 351030/371472 [5:46:30<1:42:01,  3.34it/s]
 94%|█████████▍| 351031/371472 [5:46:30<1:43:16,  3.30it/s]
 94%|█████████▍| 351032/371472 [5:46:30<1:41:53,  3.34it/s]
 94%|█████████▍| 351033/371472 [5:46:30<1:38:08,  3.47it/s]
 94%|█████████▍| 351034/371472 [5:46:31<1:33:25,  3.65it/s]
 94%|█████████▍| 351035/371472 [5:46:31<1:31:16,  3.73it/s]
 94%|█████████▍| 351036/371472 [5:46:31<1:35:20,  3.57it/s]
 94%|█████████▍| 351037/371472 [5:46:32<1:34:12,  3.62it/s]
 94%|█████████▍| 351038/371472 [5:46:32<1:40:43,  3.38it/s]
 94%|█████████▍| 351039/371472 [5:46:32<1:38:24,  3.46it/s]
 94%|█████████▍| 351040/371472 [5:46:32<1:38:23,  3.46it/s]
                                                           
{'loss': 2.5997, 'learning_rate': 1.4952918614923826e-07, 'epoch': 15.12}
+
 94%|█████████▍| 351040/371472 [5:46:32<1:38:23,  3.46it/s]
 94%|█████████▍| 351041/371472 [5:46:33<1:37:11,  3.50it/s]
 95%|█████████▍| 351042/371472 [5:46:33<1:34:32,  3.60it/s]
 95%|█████████▍| 351043/371472 [5:46:33<1:34:12,  3.61it/s]
 95%|█████████▍| 351044/371472 [5:46:34<1:40:04,  3.40it/s]
 95%|█████████▍| 351045/371472 [5:46:34<1:36:07,  3.54it/s]
 95%|█████████▍| 351046/371472 [5:46:34<1:35:58,  3.55it/s]
 95%|█████████▍| 351047/371472 [5:46:34<1:35:27,  3.57it/s]
 95%|█████████▍| 351048/371472 [5:46:35<1:30:58,  3.74it/s]
 95%|█████████▍| 351049/371472 [5:46:35<1:32:57,  3.66it/s]
 95%|█████████▍| 351050/371472 [5:46:35<1:34:43,  3.59it/s]
 95%|█████████▍| 351051/371472 [5:46:35<1:34:41,  3.59it/s]
 95%|█████████▍| 351052/371472 [5:46:36<1:33:57,  3.62it/s]
 95%|█████████▍| 351053/371472 [5:46:36<1:34:36,  3.60it/s]
 95%|█████████▍| 351054/371472 [5:46:36<1:36:04,  3.54it/s]
 95%|█████████▍| 351055/371472 [5:46:37<1:37:07,  3.50it/s]
 95%|█████████▍| 351056/371472 [5:46:37<1:33:36,  3.63it/s]
 95%|█████████▍| 351057/371472 [5:46:37<1:32:23,  3.68it/s]
 95%|█████████▍| 351058/371472 [5:46:37<1:34:18,  3.61it/s]
 95%|█████████▍| 351059/371472 [5:46:38<1:38:57,  3.44it/s]
 95%|█████████▍| 351060/371472 [5:46:38<1:39:00,  3.44it/s]
                                                           
{'loss': 2.5916, 'learning_rate': 1.494807041737594e-07, 'epoch': 15.12}
+
 95%|█████████▍| 351060/371472 [5:46:38<1:39:00,  3.44it/s]
 95%|█████████▍| 351061/371472 [5:46:38<1:40:03,  3.40it/s]
 95%|█████████▍| 351062/371472 [5:46:39<1:36:16,  3.53it/s]
 95%|█████████▍| 351063/371472 [5:46:39<1:33:30,  3.64it/s]
 95%|█████████▍| 351064/371472 [5:46:39<1:35:18,  3.57it/s]
 95%|█████████▍| 351065/371472 [5:46:39<1:36:38,  3.52it/s]
 95%|█████████▍| 351066/371472 [5:46:40<1:36:39,  3.52it/s]
 95%|█████████▍| 351067/371472 [5:46:40<1:37:18,  3.50it/s]
 95%|█████████▍| 351068/371472 [5:46:40<1:37:31,  3.49it/s]
 95%|█████████▍| 351069/371472 [5:46:41<1:42:28,  3.32it/s]
 95%|█████████▍| 351070/371472 [5:46:41<1:39:38,  3.41it/s]
 95%|█████████▍| 351071/371472 [5:46:41<1:36:54,  3.51it/s]
 95%|█████████▍| 351072/371472 [5:46:41<1:37:24,  3.49it/s]
 95%|█████████▍| 351073/371472 [5:46:42<1:35:02,  3.58it/s]
 95%|█████████▍| 351074/371472 [5:46:42<1:39:34,  3.41it/s]
 95%|█████████▍| 351075/371472 [5:46:42<1:34:44,  3.59it/s]
 95%|█████████▍| 351076/371472 [5:46:43<1:34:11,  3.61it/s]
 95%|█████████▍| 351077/371472 [5:46:43<1:33:51,  3.62it/s]
 95%|█████████▍| 351078/371472 [5:46:43<1:33:25,  3.64it/s]
 95%|█████████▍| 351079/371472 [5:46:43<1:30:36,  3.75it/s]
 95%|█████████▍| 351080/371472 [5:46:44<1:30:54,  3.74it/s]
                                                           
{'loss': 2.6563, 'learning_rate': 1.4943222219828046e-07, 'epoch': 15.12}
+
 95%|█████████▍| 351080/371472 [5:46:44<1:30:54,  3.74it/s]
 95%|█████████▍| 351081/371472 [5:46:44<1:33:34,  3.63it/s]
 95%|█████████▍| 351082/371472 [5:46:44<1:52:55,  3.01it/s]
 95%|█████████▍| 351083/371472 [5:46:45<1:47:21,  3.17it/s]
 95%|█████████▍| 351084/371472 [5:46:45<1:43:08,  3.29it/s]
 95%|█████████▍| 351085/371472 [5:46:45<1:36:56,  3.51it/s]
 95%|█████████▍| 351086/371472 [5:46:45<1:33:12,  3.65it/s]
 95%|█████████▍| 351087/371472 [5:46:46<1:30:31,  3.75it/s]
 95%|█████████▍| 351088/371472 [5:46:46<1:33:11,  3.65it/s]
 95%|█████████▍| 351089/371472 [5:46:46<1:33:12,  3.64it/s]
 95%|█████████▍| 351090/371472 [5:46:47<1:40:21,  3.38it/s]
 95%|█████████▍| 351091/371472 [5:46:47<1:38:17,  3.46it/s]
 95%|█████████▍| 351092/371472 [5:46:47<1:36:24,  3.52it/s]
 95%|█████████▍| 351093/371472 [5:46:47<1:35:25,  3.56it/s]
 95%|█████████▍| 351094/371472 [5:46:48<1:33:05,  3.65it/s]
 95%|█████████▍| 351095/371472 [5:46:48<1:31:17,  3.72it/s]
 95%|█████████▍| 351096/371472 [5:46:48<1:31:57,  3.69it/s]
 95%|█████████▍| 351097/371472 [5:46:49<1:36:26,  3.52it/s]
 95%|█████████▍| 351098/371472 [5:46:49<1:37:23,  3.49it/s]
 95%|█████████▍| 351099/371472 [5:46:49<1:38:52,  3.43it/s]
 95%|█████████▍| 351100/371472 [5:46:49<1:36:17,  3.53it/s]
                                                           
{'loss': 2.695, 'learning_rate': 1.4938374022280163e-07, 'epoch': 15.12}
+
 95%|█████████▍| 351100/371472 [5:46:49<1:36:17,  3.53it/s]
 95%|█████████▍| 351101/371472 [5:46:50<1:34:16,  3.60it/s]
 95%|█████████▍| 351102/371472 [5:46:50<1:36:12,  3.53it/s]
 95%|█████████▍| 351103/371472 [5:46:50<1:36:46,  3.51it/s]
 95%|█████████▍| 351104/371472 [5:46:51<1:44:16,  3.26it/s]
 95%|█████████▍| 351105/371472 [5:46:51<1:36:21,  3.52it/s]
 95%|█████████▍| 351106/371472 [5:46:51<1:44:34,  3.25it/s]
 95%|█████████▍| 351107/371472 [5:46:51<1:39:04,  3.43it/s]
 95%|█████████▍| 351108/371472 [5:46:52<1:35:04,  3.57it/s]
 95%|█████████▍| 351109/371472 [5:46:52<1:33:31,  3.63it/s]
 95%|█████████▍| 351110/371472 [5:46:52<1:34:58,  3.57it/s]
 95%|█████████▍| 351111/371472 [5:46:53<1:35:26,  3.56it/s]
 95%|█████████▍| 351112/371472 [5:46:53<1:41:11,  3.35it/s]
 95%|█████████▍| 351113/371472 [5:46:53<1:40:48,  3.37it/s]
 95%|█████████▍| 351114/371472 [5:46:53<1:37:03,  3.50it/s]
 95%|█████████▍| 351115/371472 [5:46:54<1:36:31,  3.52it/s]
 95%|█████████▍| 351116/371472 [5:46:54<1:43:22,  3.28it/s]
 95%|█████████▍| 351117/371472 [5:46:54<1:41:28,  3.34it/s]
 95%|█████████▍| 351118/371472 [5:46:55<1:39:23,  3.41it/s]
 95%|█████████▍| 351119/371472 [5:46:55<1:35:16,  3.56it/s]
 95%|█████████▍| 351120/371472 [5:46:55<1:41:24,  3.34it/s]
                                                           
{'loss': 2.5684, 'learning_rate': 1.4933525824732268e-07, 'epoch': 15.12}
+
 95%|█████████▍| 351120/371472 [5:46:55<1:41:24,  3.34it/s]
 95%|█████████▍| 351121/371472 [5:46:55<1:37:33,  3.48it/s]
 95%|█████████▍| 351122/371472 [5:46:56<1:35:17,  3.56it/s]
 95%|█████████▍| 351123/371472 [5:46:56<1:31:51,  3.69it/s]
 95%|█████████▍| 351124/371472 [5:46:56<1:33:11,  3.64it/s]
 95%|█████████▍| 351125/371472 [5:46:57<1:42:06,  3.32it/s]
 95%|█████████▍| 351126/371472 [5:46:57<1:44:42,  3.24it/s]
 95%|█████████▍| 351127/371472 [5:46:57<1:38:25,  3.45it/s]
 95%|█████████▍| 351128/371472 [5:46:57<1:35:00,  3.57it/s]
 95%|█████████▍| 351129/371472 [5:46:58<1:35:36,  3.55it/s]
 95%|█████████▍| 351130/371472 [5:46:58<1:34:45,  3.58it/s]
 95%|█████████▍| 351131/371472 [5:46:58<1:38:34,  3.44it/s]
 95%|█████████▍| 351132/371472 [5:46:59<1:40:55,  3.36it/s]
 95%|█████████▍| 351133/371472 [5:46:59<1:36:41,  3.51it/s]
 95%|█████████▍| 351134/371472 [5:46:59<1:31:48,  3.69it/s]
 95%|█████████▍| 351135/371472 [5:46:59<1:34:06,  3.60it/s]
 95%|█████████▍| 351136/371472 [5:47:00<1:32:18,  3.67it/s]
 95%|█████████▍| 351137/371472 [5:47:00<1:41:01,  3.35it/s]
 95%|█████████▍| 351138/371472 [5:47:00<1:38:10,  3.45it/s]
 95%|█████████▍| 351139/371472 [5:47:01<1:51:27,  3.04it/s]
 95%|█████████▍| 351140/371472 [5:47:01<1:46:42,  3.18it/s]
                                                           
{'loss': 2.7178, 'learning_rate': 1.4928677627184383e-07, 'epoch': 15.12}
+
 95%|█████████▍| 351140/371472 [5:47:01<1:46:42,  3.18it/s]
 95%|█████████▍| 351141/371472 [5:47:01<1:43:21,  3.28it/s]
 95%|█████████▍| 351142/371472 [5:47:02<1:46:33,  3.18it/s]
 95%|█████████▍| 351143/371472 [5:47:02<1:40:52,  3.36it/s]
 95%|█████████▍| 351144/371472 [5:47:02<1:43:04,  3.29it/s]
 95%|█████████▍| 351145/371472 [5:47:03<1:52:15,  3.02it/s]
 95%|█████████▍| 351146/371472 [5:47:03<1:46:13,  3.19it/s]
 95%|█████████▍| 351147/371472 [5:47:03<1:42:11,  3.32it/s]
 95%|█████████▍| 351148/371472 [5:47:03<1:38:10,  3.45it/s]
 95%|█████████▍| 351149/371472 [5:47:04<1:48:12,  3.13it/s]
 95%|█████████▍| 351150/371472 [5:47:04<1:49:08,  3.10it/s]
 95%|█████████▍| 351151/371472 [5:47:04<1:42:37,  3.30it/s]
 95%|█████████▍| 351152/371472 [5:47:05<1:44:40,  3.24it/s]
 95%|█████████▍| 351153/371472 [5:47:05<1:37:54,  3.46it/s]
 95%|█████████▍| 351154/371472 [5:47:05<1:42:07,  3.32it/s]
 95%|█████████▍| 351155/371472 [5:47:06<1:44:19,  3.25it/s]
 95%|█████████▍| 351156/371472 [5:47:06<1:41:30,  3.34it/s]
 95%|█████████▍| 351157/371472 [5:47:06<1:39:10,  3.41it/s]
 95%|█████████▍| 351158/371472 [5:47:06<1:36:08,  3.52it/s]
 95%|█████████▍| 351159/371472 [5:47:07<1:32:53,  3.64it/s]
 95%|█████████▍| 351160/371472 [5:47:07<1:34:10,  3.59it/s]
                                                           
{'loss': 2.5704, 'learning_rate': 1.492382942963649e-07, 'epoch': 15.13}
+
 95%|█████████▍| 351160/371472 [5:47:07<1:34:10,  3.59it/s]
 95%|█████████▍| 351161/371472 [5:47:07<1:35:36,  3.54it/s]
 95%|█████████▍| 351162/371472 [5:47:08<1:32:58,  3.64it/s]
 95%|█████████▍| 351163/371472 [5:47:08<1:29:59,  3.76it/s]
 95%|█████████▍| 351164/371472 [5:47:08<1:33:50,  3.61it/s]
 95%|█████████▍| 351165/371472 [5:47:08<1:33:05,  3.64it/s]
 95%|█████████▍| 351166/371472 [5:47:09<1:37:26,  3.47it/s]
 95%|█████████▍| 351167/371472 [5:47:09<1:44:37,  3.23it/s]
 95%|█████████▍| 351168/371472 [5:47:09<1:41:31,  3.33it/s]
 95%|█████████▍| 351169/371472 [5:47:10<1:41:27,  3.34it/s]
 95%|█████████▍| 351170/371472 [5:47:10<1:51:08,  3.04it/s]
 95%|█████████▍| 351171/371472 [5:47:10<1:43:21,  3.27it/s]
 95%|█████████▍| 351172/371472 [5:47:11<1:44:25,  3.24it/s]
 95%|█████████▍| 351173/371472 [5:47:11<1:40:35,  3.36it/s]
 95%|█████████▍| 351174/371472 [5:47:11<1:36:58,  3.49it/s]
 95%|█████████▍| 351175/371472 [5:47:11<1:35:25,  3.55it/s]
 95%|█████████▍| 351176/371472 [5:47:12<1:33:43,  3.61it/s]
 95%|█████████▍| 351177/371472 [5:47:12<1:36:05,  3.52it/s]
 95%|█████████▍| 351178/371472 [5:47:12<1:33:32,  3.62it/s]
 95%|█████████▍| 351179/371472 [5:47:12<1:35:23,  3.55it/s]
 95%|█████████▍| 351180/371472 [5:47:13<1:32:40,  3.65it/s]
                                                           
{'loss': 2.5567, 'learning_rate': 1.4918981232088603e-07, 'epoch': 15.13}
+
 95%|█████████▍| 351180/371472 [5:47:13<1:32:40,  3.65it/s]
 95%|█████████▍| 351181/371472 [5:47:13<1:36:36,  3.50it/s]
 95%|█████████▍| 351182/371472 [5:47:13<1:34:50,  3.57it/s]
 95%|█████████▍| 351183/371472 [5:47:14<1:31:30,  3.70it/s]
 95%|█████████▍| 351184/371472 [5:47:14<1:34:15,  3.59it/s]
 95%|█████████▍| 351185/371472 [5:47:14<1:32:24,  3.66it/s]
 95%|█████████▍| 351186/371472 [5:47:14<1:34:14,  3.59it/s]
 95%|█████████▍| 351187/371472 [5:47:15<1:33:06,  3.63it/s]
 95%|█████████▍| 351188/371472 [5:47:15<1:31:18,  3.70it/s]
 95%|█████████▍| 351189/371472 [5:47:15<1:34:48,  3.57it/s]
 95%|█████████▍| 351190/371472 [5:47:16<1:34:22,  3.58it/s]
 95%|█████████▍| 351191/371472 [5:47:16<1:34:03,  3.59it/s]
 95%|█████████▍| 351192/371472 [5:47:16<1:30:45,  3.72it/s]
 95%|█████████▍| 351193/371472 [5:47:16<1:37:17,  3.47it/s]
 95%|█████████▍| 351194/371472 [5:47:17<1:34:02,  3.59it/s]
 95%|█████████▍| 351195/371472 [5:47:17<1:36:46,  3.49it/s]
 95%|█████████▍| 351196/371472 [5:47:17<1:37:54,  3.45it/s]
 95%|█████████▍| 351197/371472 [5:47:17<1:33:28,  3.61it/s]
 95%|█████████▍| 351198/371472 [5:47:18<1:36:56,  3.49it/s]
 95%|█████████▍| 351199/371472 [5:47:18<1:39:02,  3.41it/s]
 95%|█████████▍| 351200/371472 [5:47:18<1:36:19,  3.51it/s]
                                                           
{'loss': 2.7135, 'learning_rate': 1.491413303454071e-07, 'epoch': 15.13}
+
 95%|█████████▍| 351200/371472 [5:47:18<1:36:19,  3.51it/s]
 95%|█████████▍| 351201/371472 [5:47:19<1:36:03,  3.52it/s]
 95%|█████████▍| 351202/371472 [5:47:19<1:35:22,  3.54it/s]
 95%|█████████▍| 351203/371472 [5:47:19<1:36:01,  3.52it/s]
 95%|█████████▍| 351204/371472 [5:47:20<1:39:46,  3.39it/s]
 95%|█████████▍| 351205/371472 [5:47:20<1:40:17,  3.37it/s]
 95%|█████████▍| 351206/371472 [5:47:20<1:36:59,  3.48it/s]
 95%|█████████▍| 351207/371472 [5:47:20<1:35:41,  3.53it/s]
 95%|█████████▍| 351208/371472 [5:47:21<1:39:40,  3.39it/s]
 95%|█████████▍| 351209/371472 [5:47:21<1:37:30,  3.46it/s]
 95%|█████████▍| 351210/371472 [5:47:21<1:38:22,  3.43it/s]
 95%|█████████▍| 351211/371472 [5:47:22<1:36:47,  3.49it/s]
 95%|█████████▍| 351212/371472 [5:47:22<1:47:18,  3.15it/s]
 95%|█████████▍| 351213/371472 [5:47:22<1:42:33,  3.29it/s]
 95%|█████████▍| 351214/371472 [5:47:22<1:38:07,  3.44it/s]
 95%|█████████▍| 351215/371472 [5:47:23<1:39:00,  3.41it/s]
 95%|█████████▍| 351216/371472 [5:47:23<1:37:13,  3.47it/s]
 95%|█████████▍| 351217/371472 [5:47:23<1:34:13,  3.58it/s]
 95%|█████████▍| 351218/371472 [5:47:24<1:34:50,  3.56it/s]
 95%|█████████▍| 351219/371472 [5:47:24<1:35:51,  3.52it/s]
 95%|█████████▍| 351220/371472 [5:47:24<1:33:53,  3.59it/s]
                                                           
{'loss': 2.5743, 'learning_rate': 1.4909284836992827e-07, 'epoch': 15.13}
+
 95%|█████████▍| 351220/371472 [5:47:24<1:33:53,  3.59it/s]
 95%|█████████▍| 351221/371472 [5:47:24<1:32:45,  3.64it/s]
 95%|█████████▍| 351222/371472 [5:47:25<1:33:10,  3.62it/s]
 95%|█████████▍| 351223/371472 [5:47:25<1:30:04,  3.75it/s]
 95%|█████████▍| 351224/371472 [5:47:25<1:31:58,  3.67it/s]
 95%|█████████▍| 351225/371472 [5:47:25<1:31:06,  3.70it/s]
 95%|█████████▍| 351226/371472 [5:47:26<1:32:19,  3.65it/s]
 95%|█████████▍| 351227/371472 [5:47:26<1:32:35,  3.64it/s]
 95%|█████████▍| 351228/371472 [5:47:26<1:30:45,  3.72it/s]
 95%|█████████▍| 351229/371472 [5:47:27<1:28:53,  3.80it/s]
 95%|█████████▍| 351230/371472 [5:47:27<1:30:31,  3.73it/s]
 95%|█████████▍| 351231/371472 [5:47:27<1:28:57,  3.79it/s]
 95%|█████████▍| 351232/371472 [5:47:27<1:43:50,  3.25it/s]
 95%|█████████▍| 351233/371472 [5:47:28<1:50:00,  3.07it/s]
 95%|█████████▍| 351234/371472 [5:47:28<1:43:15,  3.27it/s]
 95%|█████████▍| 351235/371472 [5:47:28<1:38:57,  3.41it/s]
 95%|█████████▍| 351236/371472 [5:47:29<1:39:08,  3.40it/s]
 95%|█████████▍| 351237/371472 [5:47:29<1:40:59,  3.34it/s]
 95%|█████████▍| 351238/371472 [5:47:29<1:36:32,  3.49it/s]
 95%|█████████▍| 351239/371472 [5:47:30<1:35:08,  3.54it/s]
 95%|█████████▍| 351240/371472 [5:47:30<1:33:36,  3.60it/s]
                                                           
{'loss': 2.7896, 'learning_rate': 1.4904436639444932e-07, 'epoch': 15.13}
+
 95%|█████████▍| 351240/371472 [5:47:30<1:33:36,  3.60it/s]
 95%|█████████▍| 351241/371472 [5:47:30<1:37:25,  3.46it/s]
 95%|█████████▍| 351242/371472 [5:47:30<1:36:10,  3.51it/s]
 95%|█████████▍| 351243/371472 [5:47:31<1:37:28,  3.46it/s]
 95%|█████████▍| 351244/371472 [5:47:31<1:40:59,  3.34it/s]
 95%|█████████▍| 351245/371472 [5:47:31<1:39:21,  3.39it/s]
 95%|█████████▍| 351246/371472 [5:47:32<1:39:32,  3.39it/s]
 95%|█████████▍| 351247/371472 [5:47:32<1:39:35,  3.38it/s]
 95%|█████████▍| 351248/371472 [5:47:32<1:39:24,  3.39it/s]
 95%|█████████▍| 351249/371472 [5:47:32<1:36:01,  3.51it/s]
 95%|█████████▍| 351250/371472 [5:47:33<1:35:54,  3.51it/s]
 95%|█████████▍| 351251/371472 [5:47:33<1:38:08,  3.43it/s]
 95%|█████████▍| 351252/371472 [5:47:33<1:32:21,  3.65it/s]
 95%|█████████▍| 351253/371472 [5:47:34<1:34:16,  3.57it/s]
 95%|█████████▍| 351254/371472 [5:47:34<1:35:11,  3.54it/s]
 95%|█████████▍| 351255/371472 [5:47:34<1:31:40,  3.68it/s]
 95%|█████████▍| 351256/371472 [5:47:34<1:34:20,  3.57it/s]
 95%|█████████▍| 351257/371472 [5:47:35<1:38:11,  3.43it/s]
 95%|█████████▍| 351258/371472 [5:47:35<1:49:32,  3.08it/s]
 95%|█████████▍| 351259/371472 [5:47:35<1:43:24,  3.26it/s]
 95%|█████████▍| 351260/371472 [5:47:36<1:38:28,  3.42it/s]
                                                           
{'loss': 2.4229, 'learning_rate': 1.4899588441897047e-07, 'epoch': 15.13}
+
 95%|█████████▍| 351260/371472 [5:47:36<1:38:28,  3.42it/s]
 95%|█████████▍| 351261/371472 [5:47:36<1:37:42,  3.45it/s]
 95%|█████████▍| 351262/371472 [5:47:36<1:39:12,  3.40it/s]
 95%|█████████▍| 351263/371472 [5:47:37<1:39:20,  3.39it/s]
 95%|█████████▍| 351264/371472 [5:47:37<1:33:48,  3.59it/s]
 95%|█████████▍| 351265/371472 [5:47:37<1:34:40,  3.56it/s]
 95%|█████████▍| 351266/371472 [5:47:37<1:34:57,  3.55it/s]
 95%|█████████▍| 351267/371472 [5:47:38<1:36:43,  3.48it/s]
 95%|█████████▍| 351268/371472 [5:47:38<1:34:27,  3.57it/s]
 95%|█████████▍| 351269/371472 [5:47:38<1:32:29,  3.64it/s]
 95%|█████████▍| 351270/371472 [5:47:39<1:40:01,  3.37it/s]
 95%|█████████▍| 351271/371472 [5:47:39<1:37:52,  3.44it/s]
 95%|█████████▍| 351272/371472 [5:47:39<1:39:04,  3.40it/s]
 95%|█████████▍| 351273/371472 [5:47:39<1:50:31,  3.05it/s]
 95%|█████████▍| 351274/371472 [5:47:40<1:42:34,  3.28it/s]
 95%|█████████▍| 351275/371472 [5:47:40<1:42:15,  3.29it/s]
 95%|█████████▍| 351276/371472 [5:47:40<1:42:09,  3.30it/s]
 95%|█████████▍| 351277/371472 [5:47:41<1:39:25,  3.39it/s]
 95%|█████████▍| 351278/371472 [5:47:41<1:35:40,  3.52it/s]
 95%|█████████▍| 351279/371472 [5:47:41<1:35:28,  3.53it/s]
 95%|█████████▍| 351280/371472 [5:47:41<1:38:23,  3.42it/s]
                                                           
{'loss': 2.5538, 'learning_rate': 1.4894740244349154e-07, 'epoch': 15.13}
+
 95%|█████████▍| 351280/371472 [5:47:41<1:38:23,  3.42it/s]
 95%|█████████▍| 351281/371472 [5:47:42<1:38:22,  3.42it/s]
 95%|█████████▍| 351282/371472 [5:47:42<1:40:26,  3.35it/s]
 95%|█████████▍| 351283/371472 [5:47:42<1:43:36,  3.25it/s]
 95%|█████████▍| 351284/371472 [5:47:43<1:47:36,  3.13it/s]
 95%|█████████▍| 351285/371472 [5:47:43<1:39:38,  3.38it/s]
 95%|█████████▍| 351286/371472 [5:47:43<1:36:03,  3.50it/s]
 95%|█████████▍| 351287/371472 [5:47:44<1:35:57,  3.51it/s]
 95%|█████████▍| 351288/371472 [5:47:44<1:35:57,  3.51it/s]
 95%|█████████▍| 351289/371472 [5:47:44<1:34:49,  3.55it/s]
 95%|█████████▍| 351290/371472 [5:47:44<1:38:58,  3.40it/s]
 95%|█████████▍| 351291/371472 [5:47:45<1:35:38,  3.52it/s]
 95%|█████████▍| 351292/371472 [5:47:45<1:31:54,  3.66it/s]
 95%|█████████▍| 351293/371472 [5:47:45<1:31:26,  3.68it/s]
 95%|█████████▍| 351294/371472 [5:47:46<1:34:11,  3.57it/s]
 95%|█████████▍| 351295/371472 [5:47:46<1:38:24,  3.42it/s]
 95%|█████████▍| 351296/371472 [5:47:46<1:34:16,  3.57it/s]
 95%|█████████▍| 351297/371472 [5:47:46<1:43:12,  3.26it/s]
 95%|█████████▍| 351298/371472 [5:47:47<1:43:41,  3.24it/s]
 95%|█████████▍| 351299/371472 [5:47:47<1:39:07,  3.39it/s]
 95%|█████████▍| 351300/371472 [5:47:47<1:45:30,  3.19it/s]
                                                           
{'loss': 2.6139, 'learning_rate': 1.488989204680127e-07, 'epoch': 15.13}
+
 95%|█████████▍| 351300/371472 [5:47:47<1:45:30,  3.19it/s]
 95%|█████████▍| 351301/371472 [5:47:48<1:43:34,  3.25it/s]
 95%|█████████▍| 351302/371472 [5:47:48<1:40:53,  3.33it/s]
 95%|█████████▍| 351303/371472 [5:47:48<1:41:08,  3.32it/s]
 95%|█████████▍| 351304/371472 [5:47:49<1:41:49,  3.30it/s]
 95%|█████████▍| 351305/371472 [5:47:49<1:36:58,  3.47it/s]
 95%|█████████▍| 351306/371472 [5:47:49<1:34:55,  3.54it/s]
 95%|█████████▍| 351307/371472 [5:47:49<1:35:20,  3.52it/s]
 95%|█████████▍| 351308/371472 [5:47:50<1:34:50,  3.54it/s]
 95%|█████████▍| 351309/371472 [5:47:50<1:34:24,  3.56it/s]
 95%|█████████▍| 351310/371472 [5:47:50<1:33:51,  3.58it/s]
 95%|█████████▍| 351311/371472 [5:47:51<1:38:50,  3.40it/s]
 95%|█████████▍| 351312/371472 [5:47:51<1:36:13,  3.49it/s]
 95%|█████████▍| 351313/371472 [5:47:51<1:38:06,  3.42it/s]
 95%|█████████▍| 351314/371472 [5:47:51<1:37:00,  3.46it/s]
 95%|█████████▍| 351315/371472 [5:47:52<1:38:20,  3.42it/s]
 95%|█████████▍| 351316/371472 [5:47:52<1:42:11,  3.29it/s]
 95%|█████████▍| 351317/371472 [5:47:52<1:43:20,  3.25it/s]
 95%|█████████▍| 351318/371472 [5:47:53<1:43:51,  3.23it/s]
 95%|█████████▍| 351319/371472 [5:47:53<1:38:25,  3.41it/s]
 95%|█████████▍| 351320/371472 [5:47:53<1:38:26,  3.41it/s]
                                                           
{'loss': 2.5182, 'learning_rate': 1.4885043849253376e-07, 'epoch': 15.13}
+
 95%|█████████▍| 351320/371472 [5:47:53<1:38:26,  3.41it/s]
 95%|█████████▍| 351321/371472 [5:47:53<1:36:53,  3.47it/s]
 95%|█████████▍| 351322/371472 [5:47:54<1:33:15,  3.60it/s]
 95%|█████████▍| 351323/371472 [5:47:54<1:31:29,  3.67it/s]
 95%|█████████▍| 351324/371472 [5:47:54<1:30:15,  3.72it/s]
 95%|█████████▍| 351325/371472 [5:47:55<1:29:56,  3.73it/s]
 95%|█████████▍| 351326/371472 [5:47:55<1:31:44,  3.66it/s]
 95%|█████████▍| 351327/371472 [5:47:55<1:34:15,  3.56it/s]
 95%|█████████▍| 351328/371472 [5:47:55<1:32:14,  3.64it/s]
 95%|█████████▍| 351329/371472 [5:47:56<1:32:51,  3.62it/s]
 95%|█████████▍| 351330/371472 [5:47:56<1:32:17,  3.64it/s]
 95%|█████████▍| 351331/371472 [5:47:56<1:33:16,  3.60it/s]
 95%|█████████▍| 351332/371472 [5:47:57<1:35:50,  3.50it/s]
 95%|█████████▍| 351333/371472 [5:47:57<1:31:16,  3.68it/s]
 95%|█████████▍| 351334/371472 [5:47:57<1:36:30,  3.48it/s]
 95%|█████████▍| 351335/371472 [5:47:57<1:33:49,  3.58it/s]
 95%|█████████▍| 351336/371472 [5:47:58<1:34:59,  3.53it/s]
 95%|█████████▍| 351337/371472 [5:47:58<1:36:05,  3.49it/s]
 95%|█████████▍| 351338/371472 [5:47:58<1:39:05,  3.39it/s]
 95%|█████████▍| 351339/371472 [5:47:59<1:37:49,  3.43it/s]
 95%|█████████▍| 351340/371472 [5:47:59<1:36:40,  3.47it/s]
                                                           
{'loss': 2.582, 'learning_rate': 1.4880195651705491e-07, 'epoch': 15.13}
+
 95%|█████████▍| 351340/371472 [5:47:59<1:36:40,  3.47it/s]
 95%|█████████▍| 351341/371472 [5:47:59<1:42:25,  3.28it/s]
 95%|█████████▍| 351342/371472 [5:47:59<1:46:22,  3.15it/s]
 95%|█████████▍| 351343/371472 [5:48:00<1:48:56,  3.08it/s]
 95%|█████████▍| 351344/371472 [5:48:00<1:41:23,  3.31it/s]
 95%|█████████▍| 351345/371472 [5:48:00<1:44:47,  3.20it/s]
 95%|█████████▍| 351346/371472 [5:48:01<1:44:08,  3.22it/s]
 95%|█████████▍| 351347/371472 [5:48:01<1:49:01,  3.08it/s]
 95%|█████████▍| 351348/371472 [5:48:01<1:44:42,  3.20it/s]
 95%|█████████▍| 351349/371472 [5:48:02<1:42:01,  3.29it/s]
 95%|█████████▍| 351350/371472 [5:48:02<1:38:18,  3.41it/s]
 95%|█████████▍| 351351/371472 [5:48:02<1:36:48,  3.46it/s]
 95%|█████████▍| 351352/371472 [5:48:02<1:35:58,  3.49it/s]
 95%|█████████▍| 351353/371472 [5:48:03<1:36:40,  3.47it/s]
 95%|█████████▍| 351354/371472 [5:48:03<1:39:30,  3.37it/s]
 95%|█████████▍| 351355/371472 [5:48:03<1:41:31,  3.30it/s]
 95%|█████████▍| 351356/371472 [5:48:04<1:40:15,  3.34it/s]
 95%|█████████▍| 351357/371472 [5:48:04<1:40:41,  3.33it/s]
 95%|█████████▍| 351358/371472 [5:48:04<1:37:47,  3.43it/s]
 95%|█████████▍| 351359/371472 [5:48:05<1:36:23,  3.48it/s]
 95%|█████████▍| 351360/371472 [5:48:05<1:34:37,  3.54it/s]
                                                           
{'loss': 2.7352, 'learning_rate': 1.4875347454157599e-07, 'epoch': 15.13}
+
 95%|█████████▍| 351360/371472 [5:48:05<1:34:37,  3.54it/s]
 95%|█████████▍| 351361/371472 [5:48:05<1:31:14,  3.67it/s]
 95%|█████████▍| 351362/371472 [5:48:05<1:30:57,  3.68it/s]
 95%|█████████▍| 351363/371472 [5:48:06<1:42:07,  3.28it/s]
 95%|█████████▍| 351364/371472 [5:48:06<1:39:24,  3.37it/s]
 95%|█████████▍| 351365/371472 [5:48:06<1:40:58,  3.32it/s]
 95%|█████████▍| 351366/371472 [5:48:07<1:34:59,  3.53it/s]
 95%|█████████▍| 351367/371472 [5:48:07<1:32:00,  3.64it/s]
 95%|█████████▍| 351368/371472 [5:48:07<1:28:39,  3.78it/s]
 95%|█████████▍| 351369/371472 [5:48:07<1:27:15,  3.84it/s]
 95%|█████████▍| 351370/371472 [5:48:08<1:30:13,  3.71it/s]
 95%|█████████▍| 351371/371472 [5:48:08<1:30:44,  3.69it/s]
 95%|█████████▍| 351372/371472 [5:48:08<1:34:48,  3.53it/s]
 95%|█████████▍| 351373/371472 [5:48:08<1:37:12,  3.45it/s]
 95%|█████████▍| 351374/371472 [5:48:09<1:41:33,  3.30it/s]
 95%|█████████▍| 351375/371472 [5:48:09<1:38:23,  3.40it/s]
 95%|█████████▍| 351376/371472 [5:48:09<1:42:22,  3.27it/s]
 95%|█████████▍| 351377/371472 [5:48:10<1:36:16,  3.48it/s]
 95%|█████████▍| 351378/371472 [5:48:10<1:45:02,  3.19it/s]
 95%|█████████▍| 351379/371472 [5:48:10<1:41:56,  3.29it/s]
 95%|█████████▍| 351380/371472 [5:48:11<1:38:41,  3.39it/s]
                                                           
{'loss': 2.5655, 'learning_rate': 1.4870499256609714e-07, 'epoch': 15.13}
+
 95%|█████████▍| 351380/371472 [5:48:11<1:38:41,  3.39it/s]
 95%|█████████▍| 351381/371472 [5:48:11<1:37:31,  3.43it/s]
 95%|█████████▍| 351382/371472 [5:48:11<1:35:57,  3.49it/s]
 95%|█████████▍| 351383/371472 [5:48:11<1:37:03,  3.45it/s]
 95%|█████████▍| 351384/371472 [5:48:12<1:39:53,  3.35it/s]
 95%|█████████▍| 351385/371472 [5:48:12<1:38:48,  3.39it/s]
 95%|█████████▍| 351386/371472 [5:48:12<1:35:16,  3.51it/s]
 95%|█████████▍| 351387/371472 [5:48:13<1:42:47,  3.26it/s]
 95%|█████████▍| 351388/371472 [5:48:13<1:41:06,  3.31it/s]
 95%|█████████▍| 351389/371472 [5:48:13<1:36:42,  3.46it/s]
 95%|█████████▍| 351390/371472 [5:48:13<1:33:12,  3.59it/s]
 95%|█████████▍| 351391/371472 [5:48:14<1:32:41,  3.61it/s]
 95%|█████████▍| 351392/371472 [5:48:14<1:32:23,  3.62it/s]
 95%|█████████▍| 351393/371472 [5:48:14<1:31:10,  3.67it/s]
 95%|█████████▍| 351394/371472 [5:48:15<1:30:18,  3.71it/s]
 95%|█████████▍| 351395/371472 [5:48:15<1:28:15,  3.79it/s]
 95%|█████████▍| 351396/371472 [5:48:15<1:35:43,  3.50it/s]
 95%|█████████▍| 351397/371472 [5:48:15<1:41:11,  3.31it/s]
 95%|█████████▍| 351398/371472 [5:48:16<1:36:58,  3.45it/s]
 95%|█████████▍| 351399/371472 [5:48:16<1:48:33,  3.08it/s]
 95%|█████████▍| 351400/371472 [5:48:16<1:47:03,  3.12it/s]
                                                           
{'loss': 2.4846, 'learning_rate': 1.4865651059061818e-07, 'epoch': 15.14}
+
 95%|█████████▍| 351400/371472 [5:48:16<1:47:03,  3.12it/s]
 95%|█████████▍| 351401/371472 [5:48:17<1:49:36,  3.05it/s]
 95%|█████████▍| 351402/371472 [5:48:17<1:47:42,  3.11it/s]
 95%|█████████▍| 351403/371472 [5:48:17<1:43:22,  3.24it/s]
 95%|█████████▍| 351404/371472 [5:48:18<1:41:30,  3.29it/s]
 95%|█████████▍| 351405/371472 [5:48:18<1:44:45,  3.19it/s]
 95%|█████████▍| 351406/371472 [5:48:18<1:41:40,  3.29it/s]
 95%|█████████▍| 351407/371472 [5:48:19<1:39:52,  3.35it/s]
 95%|█████████▍| 351408/371472 [5:48:19<1:35:03,  3.52it/s]
 95%|█████████▍| 351409/371472 [5:48:19<1:32:10,  3.63it/s]
 95%|█████████▍| 351410/371472 [5:48:19<1:31:59,  3.63it/s]
 95%|█████████▍| 351411/371472 [5:48:20<1:28:51,  3.76it/s]
 95%|█████████▍| 351412/371472 [5:48:20<1:32:11,  3.63it/s]
 95%|█████████▍| 351413/371472 [5:48:20<1:35:09,  3.51it/s]
 95%|█████████▍| 351414/371472 [5:48:20<1:32:42,  3.61it/s]
 95%|█████████▍| 351415/371472 [5:48:21<1:30:36,  3.69it/s]
 95%|█████████▍| 351416/371472 [5:48:21<1:30:49,  3.68it/s]
 95%|█████████▍| 351417/371472 [5:48:21<1:30:17,  3.70it/s]
 95%|█████████▍| 351418/371472 [5:48:22<1:27:46,  3.81it/s]
 95%|█████���███▍| 351419/371472 [5:48:22<1:27:33,  3.82it/s]
 95%|█████████▍| 351420/371472 [5:48:22<1:34:29,  3.54it/s]
                                                           
{'loss': 2.6674, 'learning_rate': 1.4860802861513936e-07, 'epoch': 15.14}
+
 95%|█████████▍| 351420/371472 [5:48:22<1:34:29,  3.54it/s]
 95%|█████████▍| 351421/371472 [5:48:22<1:35:09,  3.51it/s]
 95%|█████████▍| 351422/371472 [5:48:23<1:34:43,  3.53it/s]
 95%|█████████▍| 351423/371472 [5:48:23<1:38:42,  3.39it/s]
 95%|█████████▍| 351424/371472 [5:48:23<1:43:01,  3.24it/s]
 95%|█████████▍| 351425/371472 [5:48:24<1:37:26,  3.43it/s]
 95%|█████████▍| 351426/371472 [5:48:24<1:44:32,  3.20it/s]
 95%|█████████▍| 351427/371472 [5:48:24<1:43:13,  3.24it/s]
 95%|█████████▍| 351428/371472 [5:48:25<1:43:36,  3.22it/s]
 95%|█████████▍| 351429/371472 [5:48:25<1:40:44,  3.32it/s]
 95%|█████████▍| 351430/371472 [5:48:25<1:42:10,  3.27it/s]
 95%|█████████▍| 351431/371472 [5:48:25<1:37:37,  3.42it/s]
 95%|█████████▍| 351432/371472 [5:48:26<1:35:11,  3.51it/s]
 95%|█████████▍| 351433/371472 [5:48:26<1:45:40,  3.16it/s]
 95%|█████████▍| 351434/371472 [5:48:26<1:41:00,  3.31it/s]
 95%|█████████▍| 351435/371472 [5:48:27<1:36:22,  3.47it/s]
 95%|█████████▍| 351436/371472 [5:48:27<1:38:25,  3.39it/s]
 95%|█████████▍| 351437/371472 [5:48:27<1:38:54,  3.38it/s]
 95%|█████████▍| 351438/371472 [5:48:28<1:41:25,  3.29it/s]
 95%|█████████▍| 351439/371472 [5:48:28<1:52:30,  2.97it/s]
 95%|█████████▍| 351440/371472 [5:48:28<1:50:19,  3.03it/s]
                                                           
{'loss': 2.5403, 'learning_rate': 1.485595466396604e-07, 'epoch': 15.14}
+
 95%|█████████▍| 351440/371472 [5:48:28<1:50:19,  3.03it/s]
 95%|█████████▍| 351441/371472 [5:48:29<1:48:51,  3.07it/s]
 95%|█████████▍| 351442/371472 [5:48:29<1:46:03,  3.15it/s]
 95%|█████████▍| 351443/371472 [5:48:29<1:39:29,  3.36it/s]
 95%|█████████▍| 351444/371472 [5:48:29<1:34:47,  3.52it/s]
 95%|█████████▍| 351445/371472 [5:48:30<1:36:17,  3.47it/s]
 95%|█████████▍| 351446/371472 [5:48:30<1:40:39,  3.32it/s]
 95%|█████████▍| 351447/371472 [5:48:30<1:43:26,  3.23it/s]
 95%|█████████▍| 351448/371472 [5:48:31<1:39:27,  3.36it/s]
 95%|█████████▍| 351449/371472 [5:48:31<1:39:12,  3.36it/s]
 95%|█████████▍| 351450/371472 [5:48:31<1:35:58,  3.48it/s]
 95%|█████████▍| 351451/371472 [5:48:31<1:32:43,  3.60it/s]
 95%|█████████▍| 351452/371472 [5:48:32<1:34:17,  3.54it/s]
 95%|█████████▍| 351453/371472 [5:48:32<1:31:46,  3.64it/s]
 95%|█████████▍| 351454/371472 [5:48:32<1:33:16,  3.58it/s]
 95%|█████████▍| 351455/371472 [5:48:33<1:34:26,  3.53it/s]
 95%|█████████▍| 351456/371472 [5:48:33<1:31:57,  3.63it/s]
 95%|█████████▍| 351457/371472 [5:48:33<1:30:34,  3.68it/s]
 95%|█████████▍| 351458/371472 [5:48:33<1:30:58,  3.67it/s]
 95%|█████████▍| 351459/371472 [5:48:34<1:33:54,  3.55it/s]
 95%|█████████▍| 351460/371472 [5:48:34<1:36:42,  3.45it/s]
                                                           
{'loss': 2.5934, 'learning_rate': 1.4851106466418145e-07, 'epoch': 15.14}
+
 95%|█████████▍| 351460/371472 [5:48:34<1:36:42,  3.45it/s]
 95%|█████████▍| 351461/371472 [5:48:34<1:35:11,  3.50it/s]
 95%|█████████▍| 351462/371472 [5:48:35<1:34:06,  3.54it/s]
 95%|█████████▍| 351463/371472 [5:48:35<1:31:17,  3.65it/s]
 95%|█████████▍| 351464/371472 [5:48:35<1:30:09,  3.70it/s]
 95%|█████████▍| 351465/371472 [5:48:35<1:33:27,  3.57it/s]
 95%|█████████▍| 351466/371472 [5:48:36<1:29:58,  3.71it/s]
 95%|█████████▍| 351467/371472 [5:48:36<1:27:22,  3.82it/s]
 95%|█████████▍| 351468/371472 [5:48:36<1:38:04,  3.40it/s]
 95%|█████████▍| 351469/371472 [5:48:36<1:35:10,  3.50it/s]
 95%|█████████▍| 351470/371472 [5:48:37<1:32:53,  3.59it/s]
 95%|█████████▍| 351471/371472 [5:48:37<1:31:15,  3.65it/s]
 95%|█████████▍| 351472/371472 [5:48:37<1:33:06,  3.58it/s]
 95%|█████████▍| 351473/371472 [5:48:38<1:30:28,  3.68it/s]
 95%|█████████▍| 351474/371472 [5:48:38<1:38:27,  3.39it/s]
 95%|█████████▍| 351475/371472 [5:48:38<1:35:36,  3.49it/s]
 95%|█████████▍| 351476/371472 [5:48:38<1:35:07,  3.50it/s]
 95%|█████████▍| 351477/371472 [5:48:39<1:32:04,  3.62it/s]
 95%|█████████▍| 351478/371472 [5:48:39<1:31:46,  3.63it/s]
 95%|█████████▍| 351479/371472 [5:48:39<1:32:44,  3.59it/s]
 95%|█████████▍| 351480/371472 [5:48:40<1:32:23,  3.61it/s]
                                                           
{'loss': 2.5181, 'learning_rate': 1.4846258268870263e-07, 'epoch': 15.14}
+
 95%|█████████▍| 351480/371472 [5:48:40<1:32:23,  3.61it/s]
 95%|█████████▍| 351481/371472 [5:48:40<1:37:03,  3.43it/s]
 95%|█████████▍| 351482/371472 [5:48:40<1:35:11,  3.50it/s]
 95%|█████████▍| 351483/371472 [5:48:40<1:30:36,  3.68it/s]
 95%|█████████▍| 351484/371472 [5:48:41<1:29:41,  3.71it/s]
 95%|█████████▍| 351485/371472 [5:48:41<1:37:17,  3.42it/s]
 95%|█████████▍| 351486/371472 [5:48:41<1:34:35,  3.52it/s]
 95%|█████████▍| 351487/371472 [5:48:41<1:33:23,  3.57it/s]
 95%|█████████▍| 351488/371472 [5:48:42<1:36:27,  3.45it/s]
 95%|█████████▍| 351489/371472 [5:48:42<1:31:46,  3.63it/s]
 95%|█████████▍| 351490/371472 [5:48:42<1:34:17,  3.53it/s]
 95%|█████████▍| 351491/371472 [5:48:43<1:36:17,  3.46it/s]
 95%|█████████▍| 351492/371472 [5:48:43<1:34:44,  3.51it/s]
 95%|█████████▍| 351493/371472 [5:48:43<1:38:39,  3.37it/s]
 95%|█████████▍| 351494/371472 [5:48:44<1:37:37,  3.41it/s]
 95%|█████████▍| 351495/371472 [5:48:44<1:35:35,  3.48it/s]
 95%|█████████▍| 351496/371472 [5:48:44<1:33:22,  3.57it/s]
 95%|█████████▍| 351497/371472 [5:48:44<1:30:16,  3.69it/s]
 95%|█████████▍| 351498/371472 [5:48:45<1:32:09,  3.61it/s]
 95%|█████████▍| 351499/371472 [5:48:45<1:29:16,  3.73it/s]
 95%|█████████▍| 351500/371472 [5:48:45<1:29:54,  3.70it/s]
                                                           
{'loss': 2.6259, 'learning_rate': 1.4841410071322367e-07, 'epoch': 15.14}
+
 95%|█████████▍| 351500/371472 [5:48:45<1:29:54,  3.70it/s]
 95%|█████████▍| 351501/371472 [5:48:45<1:28:53,  3.74it/s]
 95%|█████████▍| 351502/371472 [5:48:46<1:26:49,  3.83it/s]
 95%|█████████▍| 351503/371472 [5:48:46<1:26:45,  3.84it/s]
 95%|█████████▍| 351504/371472 [5:48:46<1:31:08,  3.65it/s]
 95%|█████████▍| 351505/371472 [5:48:46<1:29:28,  3.72it/s]
 95%|█████████▍| 351506/371472 [5:48:47<1:28:34,  3.76it/s]
 95%|█████████▍| 351507/371472 [5:48:47<1:28:55,  3.74it/s]
 95%|█████████▍| 351508/371472 [5:48:47<1:28:20,  3.77it/s]
 95%|█████████▍| 351509/371472 [5:48:48<1:28:49,  3.75it/s]
 95%|█████████▍| 351510/371472 [5:48:48<1:27:32,  3.80it/s]
 95%|█████████▍| 351511/371472 [5:48:48<1:36:59,  3.43it/s]
 95%|█████████▍| 351512/371472 [5:48:48<1:33:28,  3.56it/s]
 95%|█████████▍| 351513/371472 [5:48:49<1:34:06,  3.53it/s]
 95%|█████████▍| 351514/371472 [5:48:49<1:29:54,  3.70it/s]
 95%|█████████▍| 351515/371472 [5:48:49<1:32:05,  3.61it/s]
 95%|█████████▍| 351516/371472 [5:48:49<1:29:32,  3.71it/s]
 95%|█████████▍| 351517/371472 [5:48:50<1:29:11,  3.73it/s]
 95%|█████████▍| 351518/371472 [5:48:50<1:29:30,  3.72it/s]
 95%|█████████▍| 351519/371472 [5:48:50<1:33:28,  3.56it/s]
 95%|█████████▍| 351520/371472 [5:48:51<1:31:58,  3.62it/s]
                                                           
{'loss': 2.6567, 'learning_rate': 1.4836561873774482e-07, 'epoch': 15.14}
+
 95%|█████████▍| 351520/371472 [5:48:51<1:31:58,  3.62it/s]
 95%|█████████▍| 351521/371472 [5:48:51<1:31:03,  3.65it/s]
 95%|█████████▍| 351522/371472 [5:48:51<1:32:09,  3.61it/s]
 95%|█████████▍| 351523/371472 [5:48:51<1:29:47,  3.70it/s]
 95%|█████████▍| 351524/371472 [5:48:52<1:38:23,  3.38it/s]
 95%|█████████▍| 351525/371472 [5:48:52<1:33:41,  3.55it/s]
 95%|█████████▍| 351526/371472 [5:48:52<1:33:01,  3.57it/s]
 95%|█████████▍| 351527/371472 [5:48:53<1:30:23,  3.68it/s]
 95%|█████████▍| 351528/371472 [5:48:53<1:30:48,  3.66it/s]
 95%|█████████▍| 351529/371472 [5:48:53<1:28:53,  3.74it/s]
 95%|█████████▍| 351530/371472 [5:48:53<1:30:30,  3.67it/s]
 95%|���████████▍| 351531/371472 [5:48:54<1:29:47,  3.70it/s]
 95%|█████████▍| 351532/371472 [5:48:54<1:31:38,  3.63it/s]
 95%|█████████▍| 351533/371472 [5:48:54<1:32:37,  3.59it/s]
 95%|█████████▍| 351534/371472 [5:48:54<1:34:16,  3.53it/s]
 95%|█████████▍| 351535/371472 [5:48:55<1:47:36,  3.09it/s]
 95%|█████████▍| 351536/371472 [5:48:55<1:47:14,  3.10it/s]
 95%|█████████▍| 351537/371472 [5:48:55<1:42:37,  3.24it/s]
 95%|█████████▍| 351538/371472 [5:48:56<1:36:53,  3.43it/s]
 95%|█████████▍| 351539/371472 [5:48:56<1:33:00,  3.57it/s]
 95%|█████████▍| 351540/371472 [5:48:56<1:35:28,  3.48it/s]
                                                           
{'loss': 2.6869, 'learning_rate': 1.483171367622659e-07, 'epoch': 15.14}
+
 95%|█████████▍| 351540/371472 [5:48:56<1:35:28,  3.48it/s]
 95%|█████████▍| 351541/371472 [5:48:57<1:43:40,  3.20it/s]
 95%|█████████▍| 351542/371472 [5:48:57<1:38:45,  3.36it/s]
 95%|█████████▍| 351543/371472 [5:48:57<1:35:42,  3.47it/s]
 95%|█████████▍| 351544/371472 [5:48:57<1:32:57,  3.57it/s]
 95%|█████████▍| 351545/371472 [5:48:58<1:32:35,  3.59it/s]
 95%|█████████▍| 351546/371472 [5:48:58<1:32:08,  3.60it/s]
 95%|█████████▍| 351547/371472 [5:48:58<1:28:55,  3.73it/s]
 95%|█████████▍| 351548/371472 [5:48:59<1:32:33,  3.59it/s]
 95%|█████████▍| 351549/371472 [5:48:59<1:31:09,  3.64it/s]
 95%|█████████▍| 351550/371472 [5:48:59<1:30:17,  3.68it/s]
 95%|█████████▍| 351551/371472 [5:48:59<1:29:26,  3.71it/s]
 95%|█████████▍| 351552/371472 [5:49:00<1:27:31,  3.79it/s]
 95%|█████████▍| 351553/371472 [5:49:00<1:25:56,  3.86it/s]
 95%|█████████▍| 351554/371472 [5:49:00<1:29:16,  3.72it/s]
 95%|█████████▍| 351555/371472 [5:49:00<1:27:16,  3.80it/s]
 95%|█████████▍| 351556/371472 [5:49:01<1:27:28,  3.79it/s]
 95%|█████████▍| 351557/371472 [5:49:01<1:25:22,  3.89it/s]
 95%|█████████▍| 351558/371472 [5:49:01<1:24:50,  3.91it/s]
 95%|█████████▍| 351559/371472 [5:49:02<1:37:00,  3.42it/s]
 95%|█████████▍| 351560/371472 [5:49:02<1:34:39,  3.51it/s]
                                                           
{'loss': 2.7262, 'learning_rate': 1.4826865478678705e-07, 'epoch': 15.14}
+
 95%|█████████▍| 351560/371472 [5:49:02<1:34:39,  3.51it/s]
 95%|█████████▍| 351561/371472 [5:49:02<1:39:49,  3.32it/s]
 95%|█████████▍| 351562/371472 [5:49:02<1:35:19,  3.48it/s]
 95%|█████████▍| 351563/371472 [5:49:03<1:33:01,  3.57it/s]
 95%|█████████▍| 351564/371472 [5:49:03<1:32:42,  3.58it/s]
 95%|█████████▍| 351565/371472 [5:49:03<1:31:55,  3.61it/s]
 95%|█████████▍| 351566/371472 [5:49:03<1:33:50,  3.54it/s]
 95%|█████████▍| 351567/371472 [5:49:04<1:30:05,  3.68it/s]
 95%|█████████▍| 351568/371472 [5:49:04<1:30:04,  3.68it/s]
 95%|█████████▍| 351569/371472 [5:49:04<1:32:55,  3.57it/s]
 95%|█████████▍| 351570/371472 [5:49:05<1:30:29,  3.67it/s]
 95%|█████████▍| 351571/371472 [5:49:05<1:30:54,  3.65it/s]
 95%|█████████▍| 351572/371472 [5:49:05<1:29:26,  3.71it/s]
 95%|█████████▍| 351573/371472 [5:49:05<1:27:47,  3.78it/s]
 95%|█████████▍| 351574/371472 [5:49:06<1:30:39,  3.66it/s]
 95%|█████████▍| 351575/371472 [5:49:06<1:36:40,  3.43it/s]
 95%|█████████▍| 351576/371472 [5:49:06<1:37:43,  3.39it/s]
 95%|█████████▍| 351577/371472 [5:49:07<1:44:59,  3.16it/s]
 95%|█████████▍| 351578/371472 [5:49:07<1:39:37,  3.33it/s]
 95%|█████████▍| 351579/371472 [5:49:07<1:36:34,  3.43it/s]
 95%|█████████▍| 351580/371472 [5:49:07<1:34:46,  3.50it/s]
                                                           
{'loss': 2.6499, 'learning_rate': 1.4822017281130812e-07, 'epoch': 15.14}
+
 95%|█████████▍| 351580/371472 [5:49:07<1:34:46,  3.50it/s]
 95%|█████████▍| 351581/371472 [5:49:08<1:37:17,  3.41it/s]
 95%|█████████▍| 351582/371472 [5:49:08<1:40:37,  3.29it/s]
 95%|█████████▍| 351583/371472 [5:49:08<1:36:34,  3.43it/s]
 95%|█████████▍| 351584/371472 [5:49:09<1:37:54,  3.39it/s]
 95%|█████████▍| 351585/371472 [5:49:09<1:37:24,  3.40it/s]
 95%|█████████▍| 351586/371472 [5:49:09<1:33:31,  3.54it/s]
 95%|█████████▍| 351587/371472 [5:49:10<1:34:29,  3.51it/s]
 95%|█████████▍| 351588/371472 [5:49:10<1:38:29,  3.36it/s]
 95%|█████████▍| 351589/371472 [5:49:10<1:33:22,  3.55it/s]
 95%|█████████▍| 351590/371472 [5:49:10<1:30:52,  3.65it/s]
 95%|█████████▍| 351591/371472 [5:49:11<1:36:08,  3.45it/s]
 95%|█████████▍| 351592/371472 [5:49:11<1:38:19,  3.37it/s]
 95%|█████████▍| 351593/371472 [5:49:11<1:35:24,  3.47it/s]
 95%|█████████▍| 351594/371472 [5:49:11<1:31:40,  3.61it/s]
 95%|█████████▍| 351595/371472 [5:49:12<1:28:02,  3.76it/s]
 95%|█████████▍| 351596/371472 [5:49:12<1:27:45,  3.77it/s]
 95%|█████████▍| 351597/371472 [5:49:12<1:35:54,  3.45it/s]
 95%|█████████▍| 351598/371472 [5:49:13<1:33:23,  3.55it/s]
 95%|█████████▍| 351599/371472 [5:49:13<1:34:08,  3.52it/s]
 95%|█████████▍| 351600/371472 [5:49:13<1:31:01,  3.64it/s]
                                                           
{'loss': 2.5887, 'learning_rate': 1.4817169083582927e-07, 'epoch': 15.14}
+
 95%|█████████▍| 351600/371472 [5:49:13<1:31:01,  3.64it/s]
 95%|█████████▍| 351601/371472 [5:49:13<1:33:53,  3.53it/s]
 95%|█████████▍| 351602/371472 [5:49:14<1:29:49,  3.69it/s]
 95%|█████████▍| 351603/371472 [5:49:14<1:28:26,  3.74it/s]
 95%|█████████▍| 351604/371472 [5:49:14<1:29:47,  3.69it/s]
 95%|█████████▍| 351605/371472 [5:49:14<1:27:50,  3.77it/s]
 95%|█████████▍| 351606/371472 [5:49:15<1:27:47,  3.77it/s]
 95%|█████████▍| 351607/371472 [5:49:15<1:29:26,  3.70it/s]
 95%|█████████▍| 351608/371472 [5:49:15<1:31:27,  3.62it/s]
 95%|█████████▍| 351609/371472 [5:49:16<1:32:15,  3.59it/s]
 95%|█████████▍| 351610/371472 [5:49:16<1:29:22,  3.70it/s]
 95%|█████████▍| 351611/371472 [5:49:16<1:36:59,  3.41it/s]
 95%|█████████▍| 351612/371472 [5:49:16<1:35:01,  3.48it/s]
 95%|█████████▍| 351613/371472 [5:49:17<1:32:47,  3.57it/s]
 95%|█████████▍| 351614/371472 [5:49:17<1:42:30,  3.23it/s]
 95%|█████████▍| 351615/371472 [5:49:17<1:45:43,  3.13it/s]
 95%|█████████▍| 351616/371472 [5:49:18<1:40:27,  3.29it/s]
 95%|█████████▍| 351617/371472 [5:49:18<1:39:54,  3.31it/s]
 95%|█████████▍| 351618/371472 [5:49:18<1:34:57,  3.48it/s]
 95%|█████████▍| 351619/371472 [5:49:19<1:37:40,  3.39it/s]
 95%|█████████▍| 351620/371472 [5:49:19<1:38:05,  3.37it/s]
                                                           
{'loss': 2.5232, 'learning_rate': 1.4812320886035034e-07, 'epoch': 15.14}
+
 95%|█████████▍| 351620/371472 [5:49:19<1:38:05,  3.37it/s]
 95%|█████████▍| 351621/371472 [5:49:19<1:39:40,  3.32it/s]
 95%|█████████▍| 351622/371472 [5:49:19<1:36:26,  3.43it/s]
 95%|█████████▍| 351623/371472 [5:49:20<1:38:58,  3.34it/s]
 95%|█████████▍| 351624/371472 [5:49:20<1:44:06,  3.18it/s]
 95%|█████████▍| 351625/371472 [5:49:20<1:37:31,  3.39it/s]
 95%|█████████▍| 351626/371472 [5:49:21<1:33:49,  3.53it/s]
 95%|█████████▍| 351627/371472 [5:49:21<1:33:10,  3.55it/s]
 95%|█████████▍| 351628/371472 [5:49:21<1:43:49,  3.19it/s]
 95%|█████████▍| 351629/371472 [5:49:22<1:37:23,  3.40it/s]
 95%|█████████▍| 351630/371472 [5:49:22<1:32:42,  3.57it/s]
 95%|█████████▍| 351631/371472 [5:49:22<1:42:51,  3.22it/s]
 95%|█████████▍| 351632/371472 [5:49:22<1:42:48,  3.22it/s]
 95%|█████████▍| 351633/371472 [5:49:23<1:37:04,  3.41it/s]
 95%|█████████▍| 351634/371472 [5:49:23<1:40:03,  3.30it/s]
 95%|█████████▍| 351635/371472 [5:49:23<1:36:46,  3.42it/s]
 95%|█████████▍| 351636/371472 [5:49:24<1:31:16,  3.62it/s]
 95%|█████████▍| 351637/371472 [5:49:24<1:34:36,  3.49it/s]
 95%|█████████▍| 351638/371472 [5:49:24<1:33:30,  3.54it/s]
 95%|█████████▍| 351639/371472 [5:49:24<1:32:27,  3.57it/s]
 95%|█████████▍| 351640/371472 [5:49:25<1:35:30,  3.46it/s]
                                                           
{'loss': 2.4506, 'learning_rate': 1.480747268848715e-07, 'epoch': 15.15}
+
 95%|█████████▍| 351640/371472 [5:49:25<1:35:30,  3.46it/s]
 95%|█████████▍| 351641/371472 [5:49:25<1:36:00,  3.44it/s]
 95%|█████████▍| 351642/371472 [5:49:25<1:35:25,  3.46it/s]
 95%|█████████▍| 351643/371472 [5:49:26<1:44:44,  3.16it/s]
 95%|█████████▍| 351644/371472 [5:49:26<1:43:46,  3.18it/s]
 95%|█████████▍| 351645/371472 [5:49:26<1:38:53,  3.34it/s]
 95%|█████████▍| 351646/371472 [5:49:27<1:36:05,  3.44it/s]
 95%|█████████▍| 351647/371472 [5:49:27<1:33:06,  3.55it/s]
 95%|█████████▍| 351648/371472 [5:49:27<1:33:05,  3.55it/s]
 95%|█████████▍| 351649/371472 [5:49:27<1:30:22,  3.66it/s]
 95%|█████████▍| 351650/371472 [5:49:28<1:30:11,  3.66it/s]
 95%|█████████▍| 351651/371472 [5:49:28<1:28:15,  3.74it/s]
 95%|█████████▍| 351652/371472 [5:49:28<1:28:15,  3.74it/s]
 95%|█████████▍| 351653/371472 [5:49:28<1:27:21,  3.78it/s]
 95%|█████████▍| 351654/371472 [5:49:29<1:30:29,  3.65it/s]
 95%|█████████▍| 351655/371472 [5:49:29<1:32:11,  3.58it/s]
 95%|█████████▍| 351656/371472 [5:49:29<1:31:28,  3.61it/s]
 95%|█████████▍| 351657/371472 [5:49:30<1:29:38,  3.68it/s]
 95%|█████████▍| 351658/371472 [5:49:30<1:37:36,  3.38it/s]
 95%|█████████▍| 351659/371472 [5:49:30<1:41:01,  3.27it/s]
 95%|█████████▍| 351660/371472 [5:49:30<1:35:58,  3.44it/s]
                                                           
{'loss': 2.6012, 'learning_rate': 1.4802624490939254e-07, 'epoch': 15.15}
+
 95%|█████████▍| 351660/371472 [5:49:30<1:35:58,  3.44it/s]
 95%|█████████▍| 351661/371472 [5:49:31<1:37:32,  3.38it/s]
 95%|█████████▍| 351662/371472 [5:49:31<1:38:28,  3.35it/s]
 95%|█████████▍| 351663/371472 [5:49:31<1:38:58,  3.34it/s]
 95%|█████████▍| 351664/371472 [5:49:32<1:38:46,  3.34it/s]
 95%|█████████▍| 351665/371472 [5:49:32<1:34:53,  3.48it/s]
 95%|█████████▍| 351666/371472 [5:49:32<1:33:42,  3.52it/s]
 95%|█████████▍| 351667/371472 [5:49:33<1:38:24,  3.35it/s]
 95%|█████████▍| 351668/371472 [5:49:33<1:47:07,  3.08it/s]
 95%|█████████▍| 351669/371472 [5:49:33<1:45:38,  3.12it/s]
 95%|█████████▍| 351670/371472 [5:49:34<1:44:10,  3.17it/s]
 95%|█████████▍| 351671/371472 [5:49:34<1:37:45,  3.38it/s]
 95%|█████████▍| 351672/371472 [5:49:34<1:34:58,  3.47it/s]
 95%|█████████▍| 351673/371472 [5:49:34<1:34:26,  3.49it/s]
 95%|█████████▍| 351674/371472 [5:49:35<1:33:14,  3.54it/s]
 95%|█████████▍| 351675/371472 [5:49:35<1:34:31,  3.49it/s]
 95%|█████████▍| 351676/371472 [5:49:35<1:32:25,  3.57it/s]
 95%|█████████▍| 351677/371472 [5:49:35<1:32:12,  3.58it/s]
 95%|█████████▍| 351678/371472 [5:49:36<1:37:59,  3.37it/s]
 95%|█████████▍| 351679/371472 [5:49:36<1:33:40,  3.52it/s]
 95%|█████████▍| 351680/371472 [5:49:36<1:32:19,  3.57it/s]
                                                           
{'loss': 2.6042, 'learning_rate': 1.479777629339137e-07, 'epoch': 15.15}
+
 95%|█████████▍| 351680/371472 [5:49:36<1:32:19,  3.57it/s]
 95%|█████████▍| 351681/371472 [5:49:37<1:33:02,  3.54it/s]
 95%|█████████▍| 351682/371472 [5:49:37<1:32:52,  3.55it/s]
 95%|█████████▍| 351683/371472 [5:49:37<1:29:18,  3.69it/s]
 95%|█████████▍| 351684/371472 [5:49:37<1:30:09,  3.66it/s]
 95%|█████████▍| 351685/371472 [5:49:38<1:30:22,  3.65it/s]
 95%|█████████▍| 351686/371472 [5:49:38<1:27:28,  3.77it/s]
 95%|█████████▍| 351687/371472 [5:49:38<1:33:29,  3.53it/s]
 95%|█████████▍| 351688/371472 [5:49:39<1:31:38,  3.60it/s]
 95%|█████████▍| 351689/371472 [5:49:39<1:30:10,  3.66it/s]
 95%|█████████▍| 351690/371472 [5:49:39<1:31:15,  3.61it/s]
 95%|█████████▍| 351691/371472 [5:49:39<1:30:59,  3.62it/s]
 95%|█████████▍| 351692/371472 [5:49:40<1:28:04,  3.74it/s]
 95%|█████████▍| 351693/371472 [5:49:40<1:26:51,  3.80it/s]
 95%|█████████▍| 351694/371472 [5:49:40<1:26:19,  3.82it/s]
 95%|█████████▍| 351695/371472 [5:49:40<1:27:33,  3.76it/s]
 95%|█████████▍| 351696/371472 [5:49:41<1:28:05,  3.74it/s]
 95%|█████████▍| 351697/371472 [5:49:41<1:31:23,  3.61it/s]
 95%|█████████▍| 351698/371472 [5:49:41<1:36:23,  3.42it/s]
 95%|█████████▍| 351699/371472 [5:49:42<1:33:33,  3.52it/s]
 95%|█████████▍| 351700/371472 [5:49:42<1:33:59,  3.51it/s]
                                                           
{'loss': 2.6913, 'learning_rate': 1.4792928095843476e-07, 'epoch': 15.15}
+
 95%|█████████▍| 351700/371472 [5:49:42<1:33:59,  3.51it/s]
 95%|█████████▍| 351701/371472 [5:49:42<1:33:58,  3.51it/s]
 95%|█████████▍| 351702/371472 [5:49:42<1:29:26,  3.68it/s]
 95%|█████████▍| 351703/371472 [5:49:43<1:28:34,  3.72it/s]
 95%|█████████▍| 351704/371472 [5:49:43<1:25:59,  3.83it/s]
 95%|█████████▍| 351705/371472 [5:49:43<1:33:25,  3.53it/s]
 95%|█████████▍| 351706/371472 [5:49:43<1:34:45,  3.48it/s]
 95%|█████████▍| 351707/371472 [5:49:44<1:34:36,  3.48it/s]
 95%|█████████▍| 351708/371472 [5:49:44<1:31:47,  3.59it/s]
 95%|█████████▍| 351709/371472 [5:49:44<1:35:00,  3.47it/s]
 95%|█████████▍| 351710/371472 [5:49:45<1:33:53,  3.51it/s]
 95%|█████████▍| 351711/371472 [5:49:45<1:31:00,  3.62it/s]
 95%|█████████▍| 351712/371472 [5:49:45<1:29:40,  3.67it/s]
 95%|█████████▍| 351713/371472 [5:49:45<1:26:17,  3.82it/s]
 95%|█████████▍| 351714/371472 [5:49:46<1:31:41,  3.59it/s]
 95%|█████████▍| 351715/371472 [5:49:46<1:33:48,  3.51it/s]
 95%|█████████▍| 351716/371472 [5:49:46<1:47:09,  3.07it/s]
 95%|█████████▍| 351717/371472 [5:49:47<1:44:28,  3.15it/s]
 95%|█████████▍| 351718/371472 [5:49:47<1:45:18,  3.13it/s]
 95%|█████████▍| 351719/371472 [5:49:47<1:47:03,  3.08it/s]
 95%|█████████▍| 351720/371472 [5:49:48<1:46:43,  3.08it/s]
                                                           
{'loss': 2.4961, 'learning_rate': 1.478807989829559e-07, 'epoch': 15.15}
+
 95%|█████████▍| 351720/371472 [5:49:48<1:46:43,  3.08it/s]
 95%|█████████▍| 351721/371472 [5:49:48<1:43:25,  3.18it/s]
 95%|█████████▍| 351722/371472 [5:49:48<1:38:59,  3.33it/s]
 95%|█████████▍| 351723/371472 [5:49:49<1:38:04,  3.36it/s]
 95%|█████████▍| 351724/371472 [5:49:49<1:36:53,  3.40it/s]
 95%|█████████▍| 351725/371472 [5:49:49<1:45:22,  3.12it/s]
 95%|█████████▍| 351726/371472 [5:49:50<1:41:53,  3.23it/s]
 95%|█████████▍| 351727/371472 [5:49:50<1:35:35,  3.44it/s]
 95%|█████████▍| 351728/371472 [5:49:50<1:31:55,  3.58it/s]
 95%|█████████▍| 351729/371472 [5:49:50<1:30:55,  3.62it/s]
 95%|█████████▍| 351730/371472 [5:49:51<1:30:35,  3.63it/s]
 95%|█████████▍| 351731/371472 [5:49:51<1:34:15,  3.49it/s]
 95%|█████████▍| 351732/371472 [5:49:51<1:31:42,  3.59it/s]
 95%|█████████▍| 351733/371472 [5:49:51<1:37:49,  3.36it/s]
 95%|█████████▍| 351734/371472 [5:49:52<1:33:40,  3.51it/s]
 95%|█████████▍| 351735/371472 [5:49:52<1:30:04,  3.65it/s]
 95%|█████████▍| 351736/371472 [5:49:52<1:32:48,  3.54it/s]
 95%|█████████▍| 351737/371472 [5:49:53<1:29:16,  3.68it/s]
 95%|█████████▍| 351738/371472 [5:49:53<1:29:01,  3.69it/s]
 95%|█████████▍| 351739/371472 [5:49:53<1:35:33,  3.44it/s]
 95%|█████████▍| 351740/371472 [5:49:53<1:34:47,  3.47it/s]
                                                           
{'loss': 2.6891, 'learning_rate': 1.4783231700747698e-07, 'epoch': 15.15}
+
 95%|█████████▍| 351740/371472 [5:49:53<1:34:47,  3.47it/s]
 95%|█████████▍| 351741/371472 [5:49:54<1:31:37,  3.59it/s]
 95%|█████████▍| 351742/371472 [5:49:54<1:32:58,  3.54it/s]
 95%|█████████▍| 351743/371472 [5:49:54<1:31:37,  3.59it/s]
 95%|█████████▍| 351744/371472 [5:49:55<1:32:29,  3.55it/s]
 95%|█████████▍| 351745/371472 [5:49:55<1:32:59,  3.54it/s]
 95%|█████████▍| 351746/371472 [5:49:55<1:30:15,  3.64it/s]
 95%|█████████▍| 351747/371472 [5:49:55<1:33:05,  3.53it/s]
 95%|█████████▍| 351748/371472 [5:49:56<1:30:03,  3.65it/s]
 95%|█████████▍| 351749/371472 [5:49:56<1:39:52,  3.29it/s]
 95%|█████████▍| 351750/371472 [5:49:56<1:40:15,  3.28it/s]
 95%|█████████▍| 351751/371472 [5:49:57<1:37:46,  3.36it/s]
 95%|█████████▍| 351752/371472 [5:49:57<1:34:22,  3.48it/s]
 95%|█████████▍| 351753/371472 [5:49:57<1:33:34,  3.51it/s]
 95%|█████████▍| 351754/371472 [5:49:57<1:30:16,  3.64it/s]
 95%|█████████▍| 351755/371472 [5:49:58<1:29:17,  3.68it/s]
 95%|█████████▍| 351756/371472 [5:49:58<1:27:58,  3.74it/s]
 95%|█████████��| 351757/371472 [5:49:58<1:28:08,  3.73it/s]
 95%|█████████▍| 351758/371472 [5:49:58<1:26:40,  3.79it/s]
 95%|█████████▍| 351759/371472 [5:49:59<1:27:31,  3.75it/s]
 95%|█████████▍| 351760/371472 [5:49:59<1:25:44,  3.83it/s]
                                                           
{'loss': 2.6537, 'learning_rate': 1.4778383503199813e-07, 'epoch': 15.15}
+
 95%|█████████▍| 351760/371472 [5:49:59<1:25:44,  3.83it/s]
 95%|█████████▍| 351761/371472 [5:49:59<1:25:03,  3.86it/s]
 95%|█████████▍| 351762/371472 [5:49:59<1:24:14,  3.90it/s]
 95%|█████████▍| 351763/371472 [5:50:00<1:25:09,  3.86it/s]
 95%|█████████▍| 351764/371472 [5:50:00<1:28:37,  3.71it/s]
 95%|█████████▍| 351765/371472 [5:50:00<1:30:14,  3.64it/s]
 95%|█████████▍| 351766/371472 [5:50:01<1:30:50,  3.62it/s]
 95%|█████████▍| 351767/371472 [5:50:01<1:29:54,  3.65it/s]
 95%|█████████▍| 351768/371472 [5:50:01<1:31:52,  3.57it/s]
 95%|█████████▍| 351769/371472 [5:50:01<1:32:38,  3.54it/s]
 95%|█████████▍| 351770/371472 [5:50:02<1:40:01,  3.28it/s]
 95%|█████████▍| 351771/371472 [5:50:02<1:43:02,  3.19it/s]
 95%|█████████▍| 351772/371472 [5:50:02<1:38:42,  3.33it/s]
 95%|█████████▍| 351773/371472 [5:50:03<1:34:39,  3.47it/s]
 95%|█████████▍| 351774/371472 [5:50:03<1:34:28,  3.47it/s]
 95%|█████████▍| 351775/371472 [5:50:03<1:37:06,  3.38it/s]
 95%|█████████▍| 351776/371472 [5:50:04<1:36:29,  3.40it/s]
 95%|█████████▍| 351777/371472 [5:50:04<1:34:50,  3.46it/s]
 95%|█████████▍| 351778/371472 [5:50:04<1:36:04,  3.42it/s]
 95%|█████████▍| 351779/371472 [5:50:04<1:32:16,  3.56it/s]
 95%|█████████▍| 351780/371472 [5:50:05<1:33:32,  3.51it/s]
                                                           
{'loss': 2.6841, 'learning_rate': 1.4773535305651918e-07, 'epoch': 15.15}
+
 95%|█████████▍| 351780/371472 [5:50:05<1:33:32,  3.51it/s]
 95%|█████████▍| 351781/371472 [5:50:05<1:33:53,  3.50it/s]
 95%|█████████▍| 351782/371472 [5:50:05<1:33:37,  3.51it/s]
 95%|█████████▍| 351783/371472 [5:50:06<1:34:39,  3.47it/s]
 95%|█████████▍| 351784/371472 [5:50:06<1:35:15,  3.44it/s]
 95%|█████████▍| 351785/371472 [5:50:06<1:35:35,  3.43it/s]
 95%|█████████▍| 351786/371472 [5:50:06<1:33:30,  3.51it/s]
 95%|█████████▍| 351787/371472 [5:50:07<1:30:00,  3.65it/s]
 95%|█████████▍| 351788/371472 [5:50:07<1:35:14,  3.44it/s]
 95%|█████████▍| 351789/371472 [5:50:07<1:33:28,  3.51it/s]
 95%|█████████▍| 351790/371472 [5:50:08<1:34:29,  3.47it/s]
 95%|█████████▍| 351791/371472 [5:50:08<1:36:30,  3.40it/s]
 95%|█████████▍| 351792/371472 [5:50:08<1:40:15,  3.27it/s]
 95%|█████████▍| 351793/371472 [5:50:08<1:37:02,  3.38it/s]
 95%|█████████▍| 351794/371472 [5:50:09<1:37:25,  3.37it/s]
 95%|█████████▍| 351795/371472 [5:50:09<1:35:19,  3.44it/s]
 95%|█████████▍| 351796/371472 [5:50:09<1:34:33,  3.47it/s]
 95%|█████████▍| 351797/371472 [5:50:10<1:33:17,  3.52it/s]
 95%|█████████▍| 351798/371472 [5:50:10<1:30:55,  3.61it/s]
 95%|█████████▍| 351799/371472 [5:50:10<1:29:36,  3.66it/s]
 95%|█████████▍| 351800/371472 [5:50:10<1:30:42,  3.61it/s]
                                                           
{'loss': 2.5269, 'learning_rate': 1.4768687108104035e-07, 'epoch': 15.15}
+
 95%|█████████▍| 351800/371472 [5:50:10<1:30:42,  3.61it/s]
 95%|█████████▍| 351801/371472 [5:50:11<1:29:10,  3.68it/s]
 95%|█████████▍| 351802/371472 [5:50:11<1:34:55,  3.45it/s]
 95%|█████████▍| 351803/371472 [5:50:11<1:33:40,  3.50it/s]
 95%|█████████▍| 351804/371472 [5:50:12<1:42:22,  3.20it/s]
 95%|█████████▍| 351805/371472 [5:50:12<1:39:00,  3.31it/s]
 95%|█████████▍| 351806/371472 [5:50:12<1:34:04,  3.48it/s]
 95%|█████████▍| 351807/371472 [5:50:12<1:34:46,  3.46it/s]
 95%|█████████▍| 351808/371472 [5:50:13<1:34:47,  3.46it/s]
 95%|█████████▍| 351809/371472 [5:50:13<1:34:45,  3.46it/s]
 95%|█████████▍| 351810/371472 [5:50:13<1:31:20,  3.59it/s]
 95%|█████████▍| 351811/371472 [5:50:14<1:31:08,  3.60it/s]
 95%|█████████▍| 351812/371472 [5:50:14<1:30:11,  3.63it/s]
 95%|█████████▍| 351813/371472 [5:50:14<1:29:03,  3.68it/s]
 95%|█████████▍| 351814/371472 [5:50:14<1:29:48,  3.65it/s]
 95%|█████████▍| 351815/371472 [5:50:15<1:31:49,  3.57it/s]
 95%|█████████▍| 351816/371472 [5:50:15<1:35:24,  3.43it/s]
 95%|█████████▍| 351817/371472 [5:50:15<1:32:52,  3.53it/s]
 95%|█████████▍| 351818/371472 [5:50:15<1:30:46,  3.61it/s]
 95%|█████████▍| 351819/371472 [5:50:16<1:31:10,  3.59it/s]
 95%|█████████▍| 351820/371472 [5:50:16<1:30:05,  3.64it/s]
                                                           
{'loss': 2.615, 'learning_rate': 1.476383891055614e-07, 'epoch': 15.15}
+
 95%|█████████▍| 351820/371472 [5:50:16<1:30:05,  3.64it/s]
 95%|█████████▍| 351821/371472 [5:50:16<1:31:47,  3.57it/s]
 95%|█████████▍| 351822/371472 [5:50:17<1:30:58,  3.60it/s]
 95%|█████████▍| 351823/371472 [5:50:17<1:31:58,  3.56it/s]
 95%|█████████▍| 351824/371472 [5:50:17<1:31:09,  3.59it/s]
 95%|█████████▍| 351825/371472 [5:50:18<1:41:54,  3.21it/s]
 95%|█████████▍| 351826/371472 [5:50:18<1:38:19,  3.33it/s]
 95%|█████████▍| 351827/371472 [5:50:18<1:38:40,  3.32it/s]
 95%|█████████▍| 351828/371472 [5:50:18<1:39:27,  3.29it/s]
 95%|█████████▍| 351829/371472 [5:50:19<1:37:53,  3.34it/s]
 95%|█████████▍| 351830/371472 [5:50:19<1:38:06,  3.34it/s]
 95%|█████████▍| 351831/371472 [5:50:19<1:38:40,  3.32it/s]
 95%|█████████▍| 351832/371472 [5:50:20<1:33:09,  3.51it/s]
 95%|█████████▍| 351833/371472 [5:50:20<1:37:27,  3.36it/s]
 95%|█████████▍| 351834/371472 [5:50:20<1:31:27,  3.58it/s]
 95%|█████████▍| 351835/371472 [5:50:20<1:32:37,  3.53it/s]
 95%|█████████▍| 351836/371472 [5:50:21<1:30:49,  3.60it/s]
 95%|█████████▍| 351837/371472 [5:50:21<1:41:44,  3.22it/s]
 95%|█████████▍| 351838/371472 [5:50:21<1:38:25,  3.32it/s]
 95%|█████████▍| 351839/371472 [5:50:22<1:36:50,  3.38it/s]
 95%|█████████▍| 351840/371472 [5:50:22<1:38:57,  3.31it/s]
                                                           
{'loss': 2.6477, 'learning_rate': 1.4758990713008255e-07, 'epoch': 15.15}
+
 95%|█████████▍| 351840/371472 [5:50:22<1:38:57,  3.31it/s]
 95%|█████████▍| 351841/371472 [5:50:22<1:35:12,  3.44it/s]
 95%|█████████▍| 351842/371472 [5:50:23<1:48:03,  3.03it/s]
 95%|█████████▍| 351843/371472 [5:50:23<1:46:18,  3.08it/s]
 95%|█████████▍| 351844/371472 [5:50:23<1:45:00,  3.12it/s]
 95%|█████████▍| 351845/371472 [5:50:24<1:39:43,  3.28it/s]
 95%|█████████▍| 351846/371472 [5:50:24<1:38:27,  3.32it/s]
 95%|█████████▍| 351847/371472 [5:50:24<1:37:56,  3.34it/s]
 95%|█████████▍| 351848/371472 [5:50:24<1:36:19,  3.40it/s]
 95%|█████████▍| 351849/371472 [5:50:25<1:33:18,  3.51it/s]
 95%|█████████▍| 351850/371472 [5:50:25<1:33:17,  3.51it/s]
 95%|█████████▍| 351851/371472 [5:50:25<1:35:40,  3.42it/s]
 95%|█████████▍| 351852/371472 [5:50:26<1:34:43,  3.45it/s]
 95%|█████████▍| 351853/371472 [5:50:26<1:34:18,  3.47it/s]
 95%|█████████▍| 351854/371472 [5:50:26<1:36:08,  3.40it/s]
 95%|█████████▍| 351855/371472 [5:50:26<1:35:07,  3.44it/s]
 95%|█████████▍| 351856/371472 [5:50:27<1:32:22,  3.54it/s]
 95%|█████████▍| 351857/371472 [5:50:27<1:37:19,  3.36it/s]
 95%|█████████▍| 351858/371472 [5:50:27<1:31:23,  3.58it/s]
 95%|█████████▍| 351859/371472 [5:50:28<1:29:19,  3.66it/s]
 95%|█████████▍| 351860/371472 [5:50:28<1:27:37,  3.73it/s]
                                                           
{'loss': 2.6667, 'learning_rate': 1.4754142515460362e-07, 'epoch': 15.16}
+
 95%|█████████▍| 351860/371472 [5:50:28<1:27:37,  3.73it/s]
 95%|█████████▍| 351861/371472 [5:50:28<1:35:27,  3.42it/s]
 95%|█████████▍| 351862/371472 [5:50:28<1:40:41,  3.25it/s]
 95%|█████████▍| 351863/371472 [5:50:29<1:37:05,  3.37it/s]
 95%|█████████▍| 351864/371472 [5:50:29<1:37:02,  3.37it/s]
 95%|█████████▍| 351865/371472 [5:50:29<1:37:22,  3.36it/s]
 95%|█████████▍| 351866/371472 [5:50:30<1:32:53,  3.52it/s]
 95%|█████████▍| 351867/371472 [5:50:30<1:31:49,  3.56it/s]
 95%|█████████▍| 351868/371472 [5:50:30<1:31:51,  3.56it/s]
 95%|█████████▍| 351869/371472 [5:50:30<1:34:36,  3.45it/s]
 95%|█████████▍| 351870/371472 [5:50:31<1:29:55,  3.63it/s]
 95%|█████████▍| 351871/371472 [5:50:31<1:32:57,  3.51it/s]
 95%|█████████▍| 351872/371472 [5:50:31<1:30:14,  3.62it/s]
 95%|█████████▍| 351873/371472 [5:50:32<1:29:26,  3.65it/s]
 95%|█████████▍| 351874/371472 [5:50:32<1:32:24,  3.53it/s]
 95%|█████████▍| 351875/371472 [5:50:32<1:34:05,  3.47it/s]
 95%|█████████▍| 351876/371472 [5:50:32<1:35:50,  3.41it/s]
 95%|█████████▍| 351877/371472 [5:50:33<1:32:13,  3.54it/s]
 95%|█████████▍| 351878/371472 [5:50:33<1:40:14,  3.26it/s]
 95%|█████████▍| 351879/371472 [5:50:33<1:34:18,  3.46it/s]
 95%|█████████▍| 351880/371472 [5:50:34<1:38:00,  3.33it/s]
                                                           
{'loss': 2.5958, 'learning_rate': 1.4749294317912477e-07, 'epoch': 15.16}
+
 95%|█████████▍| 351880/371472 [5:50:34<1:38:00,  3.33it/s]
 95%|█████████▍| 351881/371472 [5:50:34<1:34:39,  3.45it/s]
 95%|█████████▍| 351882/371472 [5:50:34<1:34:20,  3.46it/s]
 95%|█████████▍| 351883/371472 [5:50:34<1:33:39,  3.49it/s]
 95%|█████████▍| 351884/371472 [5:50:35<1:31:59,  3.55it/s]
 95%|█████████▍| 351885/371472 [5:50:35<1:32:36,  3.52it/s]
 95%|█████████▍| 351886/371472 [5:50:35<1:31:24,  3.57it/s]
 95%|█████████▍| 351887/371472 [5:50:36<1:35:25,  3.42it/s]
 95%|█████████▍| 351888/371472 [5:50:36<1:33:17,  3.50it/s]
 95%|█████████▍| 351889/371472 [5:50:36<1:33:24,  3.49it/s]
 95%|█████████▍| 351890/371472 [5:50:36<1:29:49,  3.63it/s]
 95%|█████████▍| 351891/371472 [5:50:37<1:36:15,  3.39it/s]
 95%|█████████▍| 351892/371472 [5:50:37<1:35:07,  3.43it/s]
 95%|█████████▍| 351893/371472 [5:50:37<1:34:41,  3.45it/s]
 95%|█████████▍| 351894/371472 [5:50:38<1:33:25,  3.49it/s]
 95%|█████████▍| 351895/371472 [5:50:38<1:42:17,  3.19it/s]
 95%|█████████▍| 351896/371472 [5:50:38<1:36:17,  3.39it/s]
 95%|█████████▍| 351897/371472 [5:50:39<1:32:03,  3.54it/s]
 95%|█████████▍| 351898/371472 [5:50:39<1:30:28,  3.61it/s]
 95%|█████████▍| 351899/371472 [5:50:39<1:30:36,  3.60it/s]
 95%|█████████▍| 351900/371472 [5:50:39<1:27:01,  3.75it/s]
                                                           
{'loss': 2.5292, 'learning_rate': 1.4744446120364582e-07, 'epoch': 15.16}
+
 95%|█████████▍| 351900/371472 [5:50:39<1:27:01,  3.75it/s]
 95%|█████████▍| 351901/371472 [5:50:40<1:28:35,  3.68it/s]
 95%|█████████▍| 351902/371472 [5:50:40<1:27:18,  3.74it/s]
 95%|█████████▍| 351903/371472 [5:50:40<1:34:30,  3.45it/s]
 95%|█████████▍| 351904/371472 [5:50:40<1:31:01,  3.58it/s]
 95%|█████████▍| 351905/371472 [5:50:41<1:31:28,  3.57it/s]
 95%|█████████▍| 351906/371472 [5:50:41<1:38:00,  3.33it/s]
 95%|█████████▍| 351907/371472 [5:50:41<1:35:54,  3.40it/s]
 95%|█████████▍| 351908/371472 [5:50:42<1:36:18,  3.39it/s]
 95%|█████████▍| 351909/371472 [5:50:42<1:34:17,  3.46it/s]
 95%|█████████▍| 351910/371472 [5:50:42<1:30:01,  3.62it/s]
 95%|█████████▍| 351911/371472 [5:50:42<1:27:46,  3.71it/s]
 95%|█████████▍| 351912/371472 [5:50:43<1:26:00,  3.79it/s]
 95%|█████████▍| 351913/371472 [5:50:43<1:30:07,  3.62it/s]
 95%|█████████▍| 351914/371472 [5:50:43<1:30:40,  3.59it/s]
 95%|█████████▍| 351915/371472 [5:50:44<1:30:58,  3.58it/s]
 95%|█████████▍| 351916/371472 [5:50:44<1:28:33,  3.68it/s]
 95%|█████████▍| 351917/371472 [5:50:44<1:27:49,  3.71it/s]
 95%|█████████▍| 351918/371472 [5:50:44<1:25:52,  3.80it/s]
 95%|█████████▍| 351919/371472 [5:50:45<1:25:54,  3.79it/s]
 95%|█████████▍| 351920/371472 [5:50:45<1:24:09,  3.87it/s]
                                                           
{'loss': 2.5468, 'learning_rate': 1.47395979228167e-07, 'epoch': 15.16}
+
 95%|█████████▍| 351920/371472 [5:50:45<1:24:09,  3.87it/s]
 95%|█████████▍| 351921/371472 [5:50:45<1:29:30,  3.64it/s]
 95%|█████████▍| 351922/371472 [5:50:45<1:30:57,  3.58it/s]
 95%|█████████▍| 351923/371472 [5:50:46<1:31:24,  3.56it/s]
 95%|█████████▍| 351924/371472 [5:50:46<1:30:50,  3.59it/s]
 95%|█████████▍| 351925/371472 [5:50:46<1:38:01,  3.32it/s]
 95%|█████████▍| 351926/371472 [5:50:47<1:34:55,  3.43it/s]
 95%|█████████▍| 351927/371472 [5:50:47<1:38:43,  3.30it/s]
 95%|█████████▍| 351928/371472 [5:50:47<1:40:12,  3.25it/s]
 95%|█████████▍| 351929/371472 [5:50:48<1:40:30,  3.24it/s]
 95%|█████████▍| 351930/371472 [5:50:48<1:42:27,  3.18it/s]
 95%|█████████▍| 351931/371472 [5:50:48<1:48:53,  2.99it/s]
 95%|█████████▍| 351932/371472 [5:50:49<1:45:03,  3.10it/s]
 95%|█████████▍| 351933/371472 [5:50:49<1:39:18,  3.28it/s]
 95%|█████████▍| 351934/371472 [5:50:49<1:33:23,  3.49it/s]
 95%|█████████▍| 351935/371472 [5:50:49<1:29:29,  3.64it/s]
 95%|█████████▍| 351936/371472 [5:50:50<1:31:27,  3.56it/s]
 95%|█████████▍| 351937/371472 [5:50:50<1:33:28,  3.48it/s]
 95%|█████████▍| 351938/371472 [5:50:50<1:32:46,  3.51it/s]
 95%|█████████▍| 351939/371472 [5:50:50<1:30:37,  3.59it/s]
 95%|█████████▍| 351940/371472 [5:50:51<1:29:53,  3.62it/s]
                                                           
{'loss': 2.6184, 'learning_rate': 1.4734749725268807e-07, 'epoch': 15.16}
+
 95%|█████████▍| 351940/371472 [5:50:51<1:29:53,  3.62it/s]
 95%|█████████▍| 351941/371472 [5:50:51<1:34:35,  3.44it/s]
 95%|█████████▍| 351942/371472 [5:50:51<1:33:22,  3.49it/s]
 95%|█████████▍| 351943/371472 [5:50:52<1:32:01,  3.54it/s]
 95%|█████████▍| 351944/371472 [5:50:52<1:34:08,  3.46it/s]
 95%|█████████▍| 351945/371472 [5:50:52<1:31:52,  3.54it/s]
 95%|█████████▍| 351946/371472 [5:50:52<1:31:02,  3.57it/s]
 95%|█████████▍| 351947/371472 [5:50:53<1:32:25,  3.52it/s]
 95%|█████████▍| 351948/371472 [5:50:53<1:38:02,  3.32it/s]
 95%|█████████▍| 351949/371472 [5:50:53<1:48:05,  3.01it/s]
 95%|█████████▍| 351950/371472 [5:50:54<1:42:37,  3.17it/s]
 95%|█████████▍| 351951/371472 [5:50:54<1:38:11,  3.31it/s]
 95%|█████████▍| 351952/371472 [5:50:54<1:37:49,  3.33it/s]
 95%|█████████▍| 351953/371472 [5:50:55<1:35:42,  3.40it/s]
 95%|█████████▍| 351954/371472 [5:50:55<1:33:06,  3.49it/s]
 95%|█████████▍| 351955/371472 [5:50:55<1:32:05,  3.53it/s]
 95%|█████████▍| 351956/371472 [5:50:55<1:32:26,  3.52it/s]
 95%|█████████▍| 351957/371472 [5:50:56<1:31:14,  3.56it/s]
 95%|█████████▍| 351958/371472 [5:50:56<1:29:02,  3.65it/s]
 95%|█████████▍| 351959/371472 [5:50:56<1:37:12,  3.35it/s]
 95%|█████████▍| 351960/371472 [5:50:57<1:40:07,  3.25it/s]
                                                           
{'loss': 2.4557, 'learning_rate': 1.472990152772092e-07, 'epoch': 15.16}
+
 95%|█████████▍| 351960/371472 [5:50:57<1:40:07,  3.25it/s]
 95%|█████████▍| 351961/371472 [5:50:57<1:36:31,  3.37it/s]
 95%|█████████▍| 351962/371472 [5:50:57<1:36:13,  3.38it/s]
 95%|█████████▍| 351963/371472 [5:50:57<1:34:47,  3.43it/s]
 95%|█████████▍| 351964/371472 [5:50:58<1:32:16,  3.52it/s]
 95%|█████████▍| 351965/371472 [5:50:58<1:33:27,  3.48it/s]
 95%|█████████▍| 351966/371472 [5:50:58<1:37:16,  3.34it/s]
 95%|█████████▍| 351967/371472 [5:50:59<1:39:02,  3.28it/s]
 95%|█████████▍| 351968/371472 [5:50:59<1:36:32,  3.37it/s]
 95%|█████████▍| 351969/371472 [5:50:59<1:37:32,  3.33it/s]
 95%|█████████▍| 351970/371472 [5:51:00<1:36:10,  3.38it/s]
 95%|█████████▍| 351971/371472 [5:51:00<1:39:49,  3.26it/s]
 95%|█████████▍| 351972/371472 [5:51:00<1:39:44,  3.26it/s]
 95%|█████████▍| 351973/371472 [5:51:00<1:36:01,  3.38it/s]
 95%|█████████▍| 351974/371472 [5:51:01<1:30:27,  3.59it/s]
 95%|█████████▍| 351975/371472 [5:51:01<1:49:17,  2.97it/s]
 95%|█████████▍| 351976/371472 [5:51:01<1:45:02,  3.09it/s]
 95%|█████████▍| 351977/371472 [5:51:02<1:38:47,  3.29it/s]
 95%|█████████▍| 351978/371472 [5:51:02<1:39:28,  3.27it/s]
 95%|█████████▍| 351979/371472 [5:51:02<1:40:40,  3.23it/s]
 95%|█████████▍| 351980/371472 [5:51:03<1:41:03,  3.21it/s]
                                                           
{'loss': 2.601, 'learning_rate': 1.4725053330173026e-07, 'epoch': 15.16}
+
 95%|█████████▍| 351980/371472 [5:51:03<1:41:03,  3.21it/s]
 95%|█████████▍| 351981/371472 [5:51:03<1:36:15,  3.37it/s]
 95%|█████████▍| 351982/371472 [5:51:03<1:34:12,  3.45it/s]
 95%|█████████▍| 351983/371472 [5:51:04<1:36:50,  3.35it/s]
 95%|█████████▍| 351984/371472 [5:51:04<1:38:30,  3.30it/s]
 95%|█████████▍| 351985/371472 [5:51:04<1:35:57,  3.38it/s]
 95%|█████████▍| 351986/371472 [5:51:04<1:33:55,  3.46it/s]
 95%|█████████▍| 351987/371472 [5:51:05<1:34:39,  3.43it/s]
 95%|█████████▍| 351988/371472 [5:51:05<1:33:25,  3.48it/s]
 95%|█████████▍| 351989/371472 [5:51:05<1:30:33,  3.59it/s]
 95%|█████████▍| 351990/371472 [5:51:06<1:29:01,  3.65it/s]
 95%|█████████▍| 351991/371472 [5:51:06<1:32:11,  3.52it/s]
 95%|█████████▍| 351992/371472 [5:51:06<1:32:33,  3.51it/s]
 95%|█████████▍| 351993/371472 [5:51:06<1:29:26,  3.63it/s]
 95%|█████████▍| 351994/371472 [5:51:07<1:26:56,  3.73it/s]
 95%|█████████▍| 351995/371472 [5:51:07<1:29:25,  3.63it/s]
 95%|█████████▍| 351996/371472 [5:51:07<1:32:30,  3.51it/s]
 95%|█████████▍| 351997/371472 [5:51:07<1:30:59,  3.57it/s]
 95%|█████████▍| 351998/371472 [5:51:08<1:29:49,  3.61it/s]
 95%|█████████▍| 351999/371472 [5:51:08<1:31:34,  3.54it/s]
 95%|█████████▍| 352000/371472 [5:51:08<1:32:00,  3.53it/s]
                                                           
{'loss': 2.5618, 'learning_rate': 1.4720205132625133e-07, 'epoch': 15.16}
+
 95%|█████████▍| 352000/371472 [5:51:08<1:32:00,  3.53it/s]
 95%|█████████▍| 352001/371472 [5:51:09<1:36:13,  3.37it/s]
 95%|█████████▍| 352002/371472 [5:51:09<1:35:06,  3.41it/s]
 95%|█████████▍| 352003/371472 [5:51:09<1:31:12,  3.56it/s]
 95%|█████████▍| 352004/371472 [5:51:09<1:27:39,  3.70it/s]
 95%|█████████▍| 352005/371472 [5:51:10<1:25:32,  3.79it/s]
 95%|█████████▍| 352006/371472 [5:51:10<1:26:29,  3.75it/s]
 95%|█████████▍| 352007/371472 [5:51:10<1:33:52,  3.46it/s]
 95%|█████████▍| 352008/371472 [5:51:11<1:30:50,  3.57it/s]
 95%|█████████▍| 352009/371472 [5:51:11<1:29:45,  3.61it/s]
 95%|█████████▍| 352010/371472 [5:51:11<1:30:21,  3.59it/s]
 95%|█████████▍| 352011/371472 [5:51:11<1:27:24,  3.71it/s]
 95%|█████████▍| 352012/371472 [5:51:12<1:27:33,  3.70it/s]
 95%|█████████▍| 352013/371472 [5:51:12<1:30:34,  3.58it/s]
 95%|█████████▍| 352014/371472 [5:51:12<1:27:48,  3.69it/s]
 95%|█████████▍| 352015/371472 [5:51:12<1:28:20,  3.67it/s]
 95%|█████████▍| 352016/371472 [5:51:13<1:28:08,  3.68it/s]
 95%|█████████▍| 352017/371472 [5:51:13<1:27:41,  3.70it/s]
 95%|█████████▍| 352018/371472 [5:51:13<1:29:55,  3.61it/s]
 95%|█████████▍| 352019/371472 [5:51:14<1:30:46,  3.57it/s]
 95%|█████████▍| 352020/371472 [5:51:14<1:31:31,  3.54it/s]
                                                           
{'loss': 2.5784, 'learning_rate': 1.4715356935077248e-07, 'epoch': 15.16}
+
 95%|█████████▍| 352020/371472 [5:51:14<1:31:31,  3.54it/s]
 95%|█████████▍| 352021/371472 [5:51:14<1:32:23,  3.51it/s]
 95%|█████████▍| 352022/371472 [5:51:14<1:29:56,  3.60it/s]
 95%|█████████▍| 352023/371472 [5:51:15<1:27:47,  3.69it/s]
 95%|█████████▍| 352024/371472 [5:51:15<1:25:28,  3.79it/s]
 95%|█████████▍| 352025/371472 [5:51:15<1:33:01,  3.48it/s]
 95%|█████████▍| 352026/371472 [5:51:16<1:33:19,  3.47it/s]
 95%|█████████▍| 352027/371472 [5:51:16<1:31:07,  3.56it/s]
 95%|█████████▍| 352028/371472 [5:51:16<1:29:35,  3.62it/s]
 95%|█████████▍| 352029/371472 [5:51:16<1:29:06,  3.64it/s]
 95%|█████████▍| 352030/371472 [5:51:17<1:32:31,  3.50it/s]
 95%|█████████▍| 352031/371472 [5:51:17<1:36:58,  3.34it/s]
 95%|█████████▍| 352032/371472 [5:51:17<1:33:35,  3.46it/s]
 95%|█████████▍| 352033/371472 [5:51:18<1:37:06,  3.34it/s]
 95%|█████████▍| 352034/371472 [5:51:18<1:38:27,  3.29it/s]
 95%|█████████▍| 352035/371472 [5:51:18<1:33:10,  3.48it/s]
 95%|█████████▍| 352036/371472 [5:51:18<1:32:38,  3.50it/s]
 95%|█████████▍| 352037/371472 [5:51:19<1:30:26,  3.58it/s]
 95%|█████████▍| 352038/371472 [5:51:19<1:30:36,  3.57it/s]
 95%|█████████▍| 352039/371472 [5:51:19<1:29:36,  3.61it/s]
 95%|█████████▍| 352040/371472 [5:51:20<1:31:08,  3.55it/s]
                                                           
{'loss': 2.4493, 'learning_rate': 1.4710508737529353e-07, 'epoch': 15.16}
+
 95%|█████████▍| 352040/371472 [5:51:20<1:31:08,  3.55it/s]
 95%|█████████▍| 352041/371472 [5:51:20<1:38:10,  3.30it/s]
 95%|█████████▍| 352042/371472 [5:51:20<1:37:52,  3.31it/s]
 95%|█████████▍| 352043/371472 [5:51:20<1:33:15,  3.47it/s]
 95%|█████████▍| 352044/371472 [5:51:21<1:32:55,  3.48it/s]
 95%|█████████▍| 352045/371472 [5:51:21<1:31:58,  3.52it/s]
 95%|█████████▍| 352046/371472 [5:51:21<1:29:38,  3.61it/s]
 95%|█████████▍| 352047/371472 [5:51:22<1:35:41,  3.38it/s]
 95%|█████████▍| 352048/371472 [5:51:22<1:31:19,  3.54it/s]
 95%|█████████▍| 352049/371472 [5:51:22<1:32:18,  3.51it/s]
 95%|█████████▍| 352050/371472 [5:51:22<1:30:15,  3.59it/s]
 95%|█████████▍| 352051/371472 [5:51:23<1:35:11,  3.40it/s]
 95%|█████████▍| 352052/371472 [5:51:23<1:34:30,  3.42it/s]
 95%|█████████▍| 352053/371472 [5:51:23<1:31:35,  3.53it/s]
 95%|█████████▍| 352054/371472 [5:51:24<1:29:27,  3.62it/s]
 95%|█████████▍| 352055/371472 [5:51:24<1:30:10,  3.59it/s]
 95%|█████████▍| 352056/371472 [5:51:24<1:34:38,  3.42it/s]
 95%|█████████▍| 352057/371472 [5:51:24<1:38:50,  3.27it/s]
 95%|█████████▍| 352058/371472 [5:51:25<1:39:55,  3.24it/s]
 95%|█████████▍| 352059/371472 [5:51:25<1:38:05,  3.30it/s]
 95%|█████████▍| 352060/371472 [5:51:25<1:39:54,  3.24it/s]
                                                           
{'loss': 2.47, 'learning_rate': 1.470566053998147e-07, 'epoch': 15.16}
+
 95%|█████████▍| 352060/371472 [5:51:25<1:39:54,  3.24it/s]
 95%|█████████▍| 352061/371472 [5:51:26<1:41:16,  3.19it/s]
 95%|█████████▍| 352062/371472 [5:51:26<1:36:39,  3.35it/s]
 95%|█████████▍| 352063/371472 [5:51:26<1:40:53,  3.21it/s]
 95%|█████████▍| 352064/371472 [5:51:27<1:36:24,  3.36it/s]
 95%|█████████▍| 352065/371472 [5:51:27<1:39:25,  3.25it/s]
 95%|█████████▍| 352066/371472 [5:51:27<1:37:04,  3.33it/s]
 95%|█████████▍| 352067/371472 [5:51:28<1:49:11,  2.96it/s]
 95%|█████████▍| 352068/371472 [5:51:28<1:46:19,  3.04it/s]
 95%|█████████▍| 352069/371472 [5:51:28<1:40:45,  3.21it/s]
 95%|█████████▍| 352070/371472 [5:51:29<1:46:08,  3.05it/s]
 95%|█████████▍| 352071/371472 [5:51:29<1:41:13,  3.19it/s]
 95%|█████████▍| 352072/371472 [5:51:29<1:35:38,  3.38it/s]
 95%|█████████▍| 352073/371472 [5:51:29<1:32:49,  3.48it/s]
 95%|█████████▍| 352074/371472 [5:51:30<1:31:55,  3.52it/s]
 95%|█████████▍| 352075/371472 [5:51:30<1:34:32,  3.42it/s]
 95%|█████████▍| 352076/371472 [5:51:30<1:35:10,  3.40it/s]
 95%|█████████▍| 352077/371472 [5:51:31<1:39:58,  3.23it/s]
 95%|█████████▍| 352078/371472 [5:51:31<1:34:39,  3.41it/s]
 95%|█████████▍| 352079/371472 [5:51:31<1:32:52,  3.48it/s]
 95%|█████████▍| 352080/371472 [5:51:32<1:37:44,  3.31it/s]
                                                           
{'loss': 2.5451, 'learning_rate': 1.4700812342433575e-07, 'epoch': 15.16}
+
 95%|█████████▍| 352080/371472 [5:51:32<1:37:44,  3.31it/s]
 95%|█████████▍| 352081/371472 [5:51:32<1:35:58,  3.37it/s]
 95%|█████████▍| 352082/371472 [5:51:32<1:34:06,  3.43it/s]
 95%|█████████▍| 352083/371472 [5:51:32<1:32:09,  3.51it/s]
 95%|█████████▍| 352084/371472 [5:51:33<1:34:42,  3.41it/s]
 95%|█████████▍| 352085/371472 [5:51:33<1:31:54,  3.52it/s]
 95%|█████████▍| 352086/371472 [5:51:33<1:32:08,  3.51it/s]
 95%|█████████▍| 352087/371472 [5:51:33<1:33:23,  3.46it/s]
 95%|█████████▍| 352088/371472 [5:51:34<1:34:04,  3.43it/s]
 95%|█████████▍| 352089/371472 [5:51:34<1:33:06,  3.47it/s]
 95%|█████████▍| 352090/371472 [5:51:34<1:31:48,  3.52it/s]
 95%|█████████▍| 352091/371472 [5:51:35<1:31:11,  3.54it/s]
 95%|█████████▍| 352092/371472 [5:51:35<1:30:47,  3.56it/s]
 95%|█████████▍| 352093/371472 [5:51:35<1:30:50,  3.56it/s]
 95%|█████████▍| 352094/371472 [5:51:35<1:28:23,  3.65it/s]
 95%|█████████▍| 352095/371472 [5:51:36<1:30:36,  3.56it/s]
 95%|█████████▍| 352096/371472 [5:51:36<1:29:25,  3.61it/s]
 95%|█████████▍| 352097/371472 [5:51:36<1:35:56,  3.37it/s]
 95%|█████████▍| 352098/371472 [5:51:37<1:37:06,  3.33it/s]
 95%|█████████▍| 352099/371472 [5:51:37<1:34:41,  3.41it/s]
 95%|█████████▍| 352100/371472 [5:51:37<1:32:33,  3.49it/s]
                                                           
{'loss': 2.6072, 'learning_rate': 1.469596414488569e-07, 'epoch': 15.17}
+
 95%|█████████▍| 352100/371472 [5:51:37<1:32:33,  3.49it/s]
 95%|█████████▍| 352101/371472 [5:51:38<1:35:51,  3.37it/s]
 95%|█████████▍| 352102/371472 [5:51:38<1:46:45,  3.02it/s]
 95%|█████████▍| 352103/371472 [5:51:38<1:41:07,  3.19it/s]
 95%|█████████▍| 352104/371472 [5:51:38<1:35:10,  3.39it/s]
 95%|█████████▍| 352105/371472 [5:51:39<1:32:30,  3.49it/s]
 95%|█████████▍| 352106/371472 [5:51:39<1:30:27,  3.57it/s]
 95%|█████████▍| 352107/371472 [5:51:39<1:37:54,  3.30it/s]
 95%|█████████▍| 352108/371472 [5:51:40<1:37:32,  3.31it/s]
 95%|█████████▍| 352109/371472 [5:51:40<1:37:46,  3.30it/s]
 95%|█████████▍| 352110/371472 [5:51:40<1:38:51,  3.26it/s]
 95%|█████████▍| 352111/371472 [5:51:41<1:36:30,  3.34it/s]
 95%|█████████▍| 352112/371472 [5:51:41<1:35:53,  3.36it/s]
 95%|█████████▍| 352113/371472 [5:51:41<1:39:57,  3.23it/s]
 95%|█████████▍| 352114/371472 [5:51:41<1:39:03,  3.26it/s]
 95%|█████████▍| 352115/371472 [5:51:42<1:34:10,  3.43it/s]
 95%|█████████▍| 352116/371472 [5:51:42<1:31:28,  3.53it/s]
 95%|█████████▍| 352117/371472 [5:51:42<1:30:38,  3.56it/s]
 95%|█████████▍| 352118/371472 [5:51:43<1:32:09,  3.50it/s]
 95%|█████████▍| 352119/371472 [5:51:43<1:27:46,  3.67it/s]
 95%|█████████▍| 352120/371472 [5:51:43<1:25:21,  3.78it/s]
                                                           
{'loss': 2.5157, 'learning_rate': 1.4691115947337797e-07, 'epoch': 15.17}
+
 95%|█████████▍| 352120/371472 [5:51:43<1:25:21,  3.78it/s]
 95%|█████████▍| 352121/371472 [5:51:43<1:25:14,  3.78it/s]
 95%|█████████▍| 352122/371472 [5:51:44<1:22:27,  3.91it/s]
 95%|█████████▍| 352123/371472 [5:51:44<1:30:25,  3.57it/s]
 95%|█████████▍| 352124/371472 [5:51:44<1:29:00,  3.62it/s]
 95%|█████████▍| 352125/371472 [5:51:44<1:32:20,  3.49it/s]
 95%|█████████▍| 352126/371472 [5:51:45<1:30:37,  3.56it/s]
 95%|█████████▍| 352127/371472 [5:51:45<1:27:34,  3.68it/s]
 95%|█████████▍| 352128/371472 [5:51:45<1:26:58,  3.71it/s]
 95%|█████████▍| 352129/371472 [5:51:46<1:30:03,  3.58it/s]
 95%|█████████▍| 352130/371472 [5:51:46<1:28:04,  3.66it/s]
 95%|█████████▍| 352131/371472 [5:51:46<1:30:46,  3.55it/s]
 95%|█████████▍| 352132/371472 [5:51:46<1:30:32,  3.56it/s]
 95%|█████████▍| 352133/371472 [5:51:47<1:28:06,  3.66it/s]
 95%|█████████▍| 352134/371472 [5:51:47<1:37:40,  3.30it/s]
 95%|█████████▍| 352135/371472 [5:51:47<1:33:07,  3.46it/s]
 95%|█████████▍| 352136/371472 [5:51:48<1:33:00,  3.46it/s]
 95%|█████████▍| 352137/371472 [5:51:48<1:29:35,  3.60it/s]
 95%|█████████▍| 352138/371472 [5:51:48<1:28:25,  3.64it/s]
 95%|█████████▍| 352139/371472 [5:51:48<1:25:19,  3.78it/s]
 95%|█████████▍| 352140/371472 [5:51:49<1:24:11,  3.83it/s]
                                                           
{'loss': 2.5283, 'learning_rate': 1.4686267749789912e-07, 'epoch': 15.17}
+
 95%|█████████▍| 352140/371472 [5:51:49<1:24:11,  3.83it/s]
 95%|█████████▍| 352141/371472 [5:51:49<1:23:46,  3.85it/s]
 95%|█████████▍| 352142/371472 [5:51:49<1:29:33,  3.60it/s]
 95%|█████████▍| 352143/371472 [5:51:49<1:28:34,  3.64it/s]
 95%|█████████▍| 352144/371472 [5:51:50<1:27:06,  3.70it/s]
 95%|█████████▍| 352145/371472 [5:51:50<1:27:35,  3.68it/s]
 95%|█████████▍| 352146/371472 [5:51:50<1:28:28,  3.64it/s]
 95%|█████████▍| 352147/371472 [5:51:51<1:27:19,  3.69it/s]
 95%|█████████▍| 352148/371472 [5:51:51<1:28:48,  3.63it/s]
 95%|█████████▍| 352149/371472 [5:51:51<1:28:40,  3.63it/s]
 95%|█████████▍| 352150/371472 [5:51:51<1:28:20,  3.65it/s]
 95%|█████████▍| 352151/371472 [5:51:52<1:28:40,  3.63it/s]
 95%|█████████▍| 352152/371472 [5:51:52<1:30:29,  3.56it/s]
 95%|█████████▍| 352153/371472 [5:51:52<1:34:32,  3.41it/s]
 95%|█████████▍| 352154/371472 [5:51:53<1:36:32,  3.34it/s]
 95%|█████████▍| 352155/371472 [5:51:53<1:34:02,  3.42it/s]
 95%|█████████▍| 352156/371472 [5:51:53<1:34:32,  3.41it/s]
 95%|█████████▍| 352157/371472 [5:51:53<1:31:47,  3.51it/s]
 95%|█████████▍| 352158/371472 [5:51:54<1:30:55,  3.54it/s]
 95%|█████████▍| 352159/371472 [5:51:54<1:30:57,  3.54it/s]
 95%|█████████▍| 352160/371472 [5:51:54<1:28:27,  3.64it/s]
                                                           
{'loss': 2.4667, 'learning_rate': 1.4681419552242017e-07, 'epoch': 15.17}
+
 95%|█████████▍| 352160/371472 [5:51:54<1:28:27,  3.64it/s]
 95%|█████████▍| 352161/371472 [5:51:54<1:27:08,  3.69it/s]
 95%|█████████▍| 352162/371472 [5:51:55<1:28:34,  3.63it/s]
 95%|█████████▍| 352163/371472 [5:51:55<1:28:11,  3.65it/s]
 95%|█████████▍| 352164/371472 [5:51:55<1:28:52,  3.62it/s]
 95%|█████████▍| 352165/371472 [5:51:56<1:25:13,  3.78it/s]
 95%|█████████▍| 352166/371472 [5:51:56<1:31:12,  3.53it/s]
 95%|█████████▍| 352167/371472 [5:51:56<1:46:44,  3.01it/s]
 95%|█████████▍| 352168/371472 [5:51:57<1:43:40,  3.10it/s]
 95%|█████████▍| 352169/371472 [5:51:57<1:37:58,  3.28it/s]
 95%|█████████▍| 352170/371472 [5:51:57<1:36:38,  3.33it/s]
 95%|█████████▍| 352171/371472 [5:51:58<1:46:38,  3.02it/s]
 95%|█████████▍| 352172/371472 [5:51:58<1:38:55,  3.25it/s]
 95%|█████████▍| 352173/371472 [5:51:58<1:43:36,  3.10it/s]
 95%|█████████▍| 352174/371472 [5:51:58<1:39:37,  3.23it/s]
 95%|█████████▍| 352175/371472 [5:51:59<1:40:12,  3.21it/s]
 95%|█████████▍| 352176/371472 [5:51:59<1:36:25,  3.34it/s]
 95%|█████████▍| 352177/371472 [5:51:59<1:32:56,  3.46it/s]
 95%|█████████▍| 352178/371472 [5:52:00<1:31:31,  3.51it/s]
 95%|█████████▍| 352179/371472 [5:52:00<1:33:08,  3.45it/s]
 95%|█████████▍| 352180/371472 [5:52:00<1:32:18,  3.48it/s]
                                                           
{'loss': 2.4131, 'learning_rate': 1.4676571354694135e-07, 'epoch': 15.17}
+
 95%|█████████▍| 352180/371472 [5:52:00<1:32:18,  3.48it/s]
 95%|█████████▍| 352181/371472 [5:52:00<1:30:58,  3.53it/s]
 95%|█████████▍| 352182/371472 [5:52:01<1:31:18,  3.52it/s]
 95%|█████████▍| 352183/371472 [5:52:01<1:36:05,  3.35it/s]
 95%|█████████▍| 352184/371472 [5:52:01<1:39:30,  3.23it/s]
 95%|█████████▍| 352185/371472 [5:52:02<1:36:35,  3.33it/s]
 95%|█████████▍| 352186/371472 [5:52:02<1:36:34,  3.33it/s]
 95%|█████████▍| 352187/371472 [5:52:02<1:33:40,  3.43it/s]
 95%|█████████▍| 352188/371472 [5:52:03<1:31:39,  3.51it/s]
 95%|█████████▍| 352189/371472 [5:52:03<1:29:04,  3.61it/s]
 95%|█████████▍| 352190/371472 [5:52:03<1:28:47,  3.62it/s]
 95%|█████████▍| 352191/371472 [5:52:03<1:30:49,  3.54it/s]
 95%|█████████▍| 352192/371472 [5:52:04<1:26:23,  3.72it/s]
 95%|█████████▍| 352193/371472 [5:52:04<1:32:20,  3.48it/s]
 95%|█████████▍| 352194/371472 [5:52:04<1:32:10,  3.49it/s]
 95%|█████████▍| 352195/371472 [5:52:05<1:34:02,  3.42it/s]
 95%|█████████▍| 352196/371472 [5:52:05<1:33:30,  3.44it/s]
 95%|█████████▍| 352197/371472 [5:52:05<1:33:43,  3.43it/s]
 95%|█████████▍| 352198/371472 [5:52:05<1:32:26,  3.47it/s]
 95%|█████████▍| 352199/371472 [5:52:06<1:30:37,  3.54it/s]
 95%|█████████▍| 352200/371472 [5:52:06<1:28:38,  3.62it/s]
                                                           
{'loss': 2.7205, 'learning_rate': 1.467172315714624e-07, 'epoch': 15.17}
+
 95%|█████████▍| 352200/371472 [5:52:06<1:28:38,  3.62it/s]
 95%|█████████▍| 352201/371472 [5:52:06<1:30:18,  3.56it/s]
 95%|█████████▍| 352202/371472 [5:52:07<1:32:53,  3.46it/s]
 95%|█████████▍| 352203/371472 [5:52:07<1:29:12,  3.60it/s]
 95%|█████████▍| 352204/371472 [5:52:07<1:27:43,  3.66it/s]
 95%|█████████▍| 352205/371472 [5:52:07<1:27:00,  3.69it/s]
 95%|█████████▍| 352206/371472 [5:52:08<1:27:35,  3.67it/s]
 95%|█████████▍| 352207/371472 [5:52:08<1:25:10,  3.77it/s]
 95%|█████████▍| 352208/371472 [5:52:08<1:29:12,  3.60it/s]
 95%|█████████▍| 352209/371472 [5:52:08<1:32:03,  3.49it/s]
 95%|█████████▍| 352210/371472 [5:52:09<1:30:51,  3.53it/s]
 95%|█████████▍| 352211/371472 [5:52:09<1:28:48,  3.61it/s]
 95%|█████████▍| 352212/371472 [5:52:09<1:26:56,  3.69it/s]
 95%|█████████▍| 352213/371472 [5:52:10<1:38:28,  3.26it/s]
 95%|█████████▍| 352214/371472 [5:52:10<1:34:43,  3.39it/s]
 95%|█████████▍| 352215/371472 [5:52:10<1:32:45,  3.46it/s]
 95%|█████████▍| 352216/371472 [5:52:10<1:29:54,  3.57it/s]
 95%|█████████▍| 352217/371472 [5:52:11<1:27:35,  3.66it/s]
 95%|█████████▍| 352218/371472 [5:52:11<1:27:36,  3.66it/s]
 95%|█████████▍| 352219/371472 [5:52:11<1:27:23,  3.67it/s]
 95%|█████████▍| 352220/371472 [5:52:12<1:33:52,  3.42it/s]
                                                           
{'loss': 2.5593, 'learning_rate': 1.4666874959598354e-07, 'epoch': 15.17}
+
 95%|█████████▍| 352220/371472 [5:52:12<1:33:52,  3.42it/s]
 95%|█████████▍| 352221/371472 [5:52:12<1:36:29,  3.33it/s]
 95%|█████████▍| 352222/371472 [5:52:12<1:31:44,  3.50it/s]
 95%|█████████▍| 352223/371472 [5:52:12<1:37:04,  3.30it/s]
 95%|█████████▍| 352224/371472 [5:52:13<1:34:42,  3.39it/s]
 95%|█████████▍| 352225/371472 [5:52:13<1:32:20,  3.47it/s]
 95%|█████████▍| 352226/371472 [5:52:13<1:30:59,  3.53it/s]
 95%|█████████▍| 352227/371472 [5:52:14<1:38:56,  3.24it/s]
 95%|█████████▍| 352228/371472 [5:52:14<1:40:49,  3.18it/s]
 95%|█████████▍| 352229/371472 [5:52:14<1:36:46,  3.31it/s]
 95%|█████████▍| 352230/371472 [5:52:15<1:31:41,  3.50it/s]
 95%|█████████▍| 352231/371472 [5:52:15<1:28:17,  3.63it/s]
 95%|█████████▍| 352232/371472 [5:52:15<1:27:01,  3.68it/s]
 95%|█████████▍| 352233/371472 [5:52:15<1:26:27,  3.71it/s]
 95%|█████████▍| 352234/371472 [5:52:16<1:26:37,  3.70it/s]
 95%|█████████▍| 352235/371472 [5:52:16<1:26:30,  3.71it/s]
 95%|█████████▍| 352236/371472 [5:52:16<1:24:45,  3.78it/s]
 95%|█████████▍| 352237/371472 [5:52:16<1:24:00,  3.82it/s]
 95%|█████████▍| 352238/371472 [5:52:17<1:26:05,  3.72it/s]
 95%|█████████▍| 352239/371472 [5:52:17<1:25:46,  3.74it/s]
 95%|█████████▍| 352240/371472 [5:52:17<1:26:22,  3.71it/s]
                                                           
{'loss': 2.5573, 'learning_rate': 1.4662026762050461e-07, 'epoch': 15.17}
+
 95%|█████████▍| 352240/371472 [5:52:17<1:26:22,  3.71it/s]
 95%|█████████▍| 352241/371472 [5:52:17<1:27:22,  3.67it/s]
 95%|█████████▍| 352242/371472 [5:52:18<1:26:40,  3.70it/s]
 95%|█████████▍| 352243/371472 [5:52:18<1:24:39,  3.79it/s]
 95%|█████████▍| 352244/371472 [5:52:18<1:23:27,  3.84it/s]
 95%|█████████▍| 352245/371472 [5:52:18<1:23:32,  3.84it/s]
 95%|█████████▍| 352246/371472 [5:52:19<1:23:54,  3.82it/s]
 95%|█████████▍| 352247/371472 [5:52:19<1:22:53,  3.87it/s]
 95%|█████████▍| 352248/371472 [5:52:19<1:30:10,  3.55it/s]
 95%|█████████▍| 352249/371472 [5:52:20<1:31:25,  3.50it/s]
 95%|█████████▍| 352250/371472 [5:52:20<1:42:32,  3.12it/s]
 95%|█████████▍| 352251/371472 [5:52:20<1:37:58,  3.27it/s]
 95%|█████████▍| 352252/371472 [5:52:21<1:33:34,  3.42it/s]
 95%|█████████▍| 352253/371472 [5:52:21<1:35:39,  3.35it/s]
 95%|█████████▍| 352254/371472 [5:52:21<1:35:51,  3.34it/s]
 95%|█████████▍| 352255/371472 [5:52:21<1:36:39,  3.31it/s]
 95%|█████████▍| 352256/371472 [5:52:22<1:34:30,  3.39it/s]
 95%|█████████▍| 352257/371472 [5:52:22<1:31:46,  3.49it/s]
 95%|█████████▍| 352258/371472 [5:52:22<1:28:54,  3.60it/s]
 95%|█████████▍| 352259/371472 [5:52:23<1:27:45,  3.65it/s]
 95%|█████████▍| 352260/371472 [5:52:23<1:29:24,  3.58it/s]
                                                           
{'loss': 2.6522, 'learning_rate': 1.4657178564502577e-07, 'epoch': 15.17}
+
 95%|█████████▍| 352260/371472 [5:52:23<1:29:24,  3.58it/s]
 95%|█████████▍| 352261/371472 [5:52:23<1:31:48,  3.49it/s]
 95%|█████████▍| 352262/371472 [5:52:23<1:28:23,  3.62it/s]
 95%|█████████▍| 352263/371472 [5:52:24<1:27:02,  3.68it/s]
 95%|█████████▍| 352264/371472 [5:52:24<1:26:04,  3.72it/s]
 95%|█████████▍| 352265/371472 [5:52:24<1:27:54,  3.64it/s]
 95%|█████████▍| 352266/371472 [5:52:25<1:33:32,  3.42it/s]
 95%|█████████▍| 352267/371472 [5:52:25<1:30:05,  3.55it/s]
 95%|█████████▍| 352268/371472 [5:52:25<1:28:15,  3.63it/s]
 95%|█████████▍| 352269/371472 [5:52:25<1:28:46,  3.61it/s]
 95%|█████████▍| 352270/371472 [5:52:26<1:28:26,  3.62it/s]
 95%|█████████▍| 352271/371472 [5:52:26<1:28:22,  3.62it/s]
 95%|█████████▍| 352272/371472 [5:52:26<1:32:43,  3.45it/s]
 95%|█████████▍| 352273/371472 [5:52:26<1:32:39,  3.45it/s]
 95%|█████████▍| 352274/371472 [5:52:27<1:34:51,  3.37it/s]
 95%|█████████▍| 352275/371472 [5:52:27<1:43:42,  3.09it/s]
 95%|█████████▍| 352276/371472 [5:52:27<1:41:58,  3.14it/s]
 95%|█████████▍| 352277/371472 [5:52:28<1:38:41,  3.24it/s]
 95%|█████████▍| 352278/371472 [5:52:28<1:38:40,  3.24it/s]
 95%|█████████▍| 352279/371472 [5:52:28<1:35:27,  3.35it/s]
 95%|█████████▍| 352280/371472 [5:52:29<1:43:36,  3.09it/s]
                                                           
{'loss': 2.6779, 'learning_rate': 1.4652330366954684e-07, 'epoch': 15.17}
+
 95%|█████████▍| 352280/371472 [5:52:29<1:43:36,  3.09it/s]
 95%|█████████▍| 352281/371472 [5:52:29<1:43:24,  3.09it/s]
 95%|█████████▍| 352282/371472 [5:52:29<1:41:49,  3.14it/s]
 95%|█████████▍| 352283/371472 [5:52:30<1:34:08,  3.40it/s]
 95%|█████████▍| 352284/371472 [5:52:30<1:31:09,  3.51it/s]
 95%|█████████▍| 352285/371472 [5:52:30<1:28:38,  3.61it/s]
 95%|█████████▍| 352286/371472 [5:52:30<1:26:44,  3.69it/s]
 95%|█████████▍| 352287/371472 [5:52:31<1:27:34,  3.65it/s]
 95%|█████████▍| 352288/371472 [5:52:31<1:35:16,  3.36it/s]
 95%|█████████▍| 352289/371472 [5:52:31<1:30:12,  3.54it/s]
 95%|█████████▍| 352290/371472 [5:52:32<1:28:43,  3.60it/s]
 95%|█████████▍| 352291/371472 [5:52:32<1:34:18,  3.39it/s]
 95%|█████████▍| 352292/371472 [5:52:32<1:40:00,  3.20it/s]
 95%|█████████▍| 352293/371472 [5:52:32<1:34:13,  3.39it/s]
 95%|█████████▍| 352294/371472 [5:52:33<1:31:03,  3.51it/s]
 95%|█████████▍| 352295/371472 [5:52:33<1:27:46,  3.64it/s]
 95%|█████████▍| 352296/371472 [5:52:33<1:27:42,  3.64it/s]
 95%|█████████▍| 352297/371472 [5:52:34<1:25:25,  3.74it/s]
 95%|█████████▍| 352298/371472 [5:52:34<1:23:21,  3.83it/s]
 95%|█████████▍| 352299/371472 [5:52:34<1:28:59,  3.59it/s]
 95%|█████████▍| 352300/371472 [5:52:34<1:29:48,  3.56it/s]
                                                           
{'loss': 2.5279, 'learning_rate': 1.46474821694068e-07, 'epoch': 15.17}
+
 95%|█████████▍| 352300/371472 [5:52:34<1:29:48,  3.56it/s]
 95%|█████████▍| 352301/371472 [5:52:35<1:30:04,  3.55it/s]
 95%|█████████▍| 352302/371472 [5:52:35<1:27:48,  3.64it/s]
 95%|█████████▍| 352303/371472 [5:52:35<1:26:01,  3.71it/s]
 95%|█████████▍| 352304/371472 [5:52:35<1:26:33,  3.69it/s]
 95%|█████████▍| 352305/371472 [5:52:36<1:28:55,  3.59it/s]
 95%|█████████▍| 352306/371472 [5:52:36<1:29:00,  3.59it/s]
 95%|█████████▍| 352307/371472 [5:52:36<1:37:22,  3.28it/s]
 95%|█████████▍| 352308/371472 [5:52:37<1:36:31,  3.31it/s]
 95%|█████████▍| 352309/371472 [5:52:37<1:33:09,  3.43it/s]
 95%|█████████▍| 352310/371472 [5:52:37<1:38:50,  3.23it/s]
 95%|█████████▍| 352311/371472 [5:52:38<1:38:23,  3.25it/s]
 95%|█████████▍| 352312/371472 [5:52:38<1:32:59,  3.43it/s]
 95%|█████████▍| 352313/371472 [5:52:38<1:34:17,  3.39it/s]
 95%|█████████▍| 352314/371472 [5:52:38<1:30:15,  3.54it/s]
 95%|█████████▍| 352315/371472 [5:52:39<1:33:02,  3.43it/s]
 95%|█████████▍| 352316/371472 [5:52:39<1:29:58,  3.55it/s]
 95%|█████████▍| 352317/371472 [5:52:39<1:26:44,  3.68it/s]
 95%|█████████▍| 352318/371472 [5:52:40<1:32:59,  3.43it/s]
 95%|█████████▍| 352319/371472 [5:52:40<1:32:29,  3.45it/s]
 95%|█████████▍| 352320/371472 [5:52:40<1:35:19,  3.35it/s]
                                                           
{'loss': 2.4927, 'learning_rate': 1.4642633971858906e-07, 'epoch': 15.18}
+
 95%|█████████▍| 352320/371472 [5:52:40<1:35:19,  3.35it/s]
 95%|█████████▍| 352321/371472 [5:52:41<1:39:15,  3.22it/s]
 95%|█████████▍| 352322/371472 [5:52:41<1:35:32,  3.34it/s]
 95%|█████████▍| 352323/371472 [5:52:41<1:32:08,  3.46it/s]
 95%|█████████▍| 352324/371472 [5:52:41<1:30:34,  3.52it/s]
 95%|█████████▍| 352325/371472 [5:52:42<1:39:06,  3.22it/s]
 95%|█████████▍| 352326/371472 [5:52:42<1:36:07,  3.32it/s]
 95%|█████████▍| 352327/371472 [5:52:42<1:33:57,  3.40it/s]
 95%|█████████▍| 352328/371472 [5:52:43<1:31:41,  3.48it/s]
 95%|█████████▍| 352329/371472 [5:52:43<1:33:21,  3.42it/s]
 95%|█████████▍| 352330/371472 [5:52:43<1:31:14,  3.50it/s]
 95%|█████████▍| 352331/371472 [5:52:43<1:30:12,  3.54it/s]
 95%|█████████▍| 352332/371472 [5:52:44<1:27:32,  3.64it/s]
 95%|█████████▍| 352333/371472 [5:52:44<1:32:25,  3.45it/s]
 95%|█████████▍| 352334/371472 [5:52:44<1:33:01,  3.43it/s]
 95%|█████████▍| 352335/371472 [5:52:45<1:33:14,  3.42it/s]
 95%|█████████▍| 352336/371472 [5:52:45<1:29:30,  3.56it/s]
 95%|█████████▍| 352337/371472 [5:52:45<1:27:16,  3.65it/s]
 95%|█████████▍| 352338/371472 [5:52:45<1:28:44,  3.59it/s]
 95%|█████████▍| 352339/371472 [5:52:46<1:28:01,  3.62it/s]
 95%|█████████▍| 352340/371472 [5:52:46<1:26:54,  3.67it/s]
                                                           
{'loss': 2.731, 'learning_rate': 1.463778577431102e-07, 'epoch': 15.18}
+
 95%|█████████▍| 352340/371472 [5:52:46<1:26:54,  3.67it/s]
 95%|█████████▍| 352341/371472 [5:52:46<1:28:20,  3.61it/s]
 95%|█████████▍| 352342/371472 [5:52:46<1:33:31,  3.41it/s]
 95%|█████████▍| 352343/371472 [5:52:47<1:30:51,  3.51it/s]
 95%|█████████▍| 352344/371472 [5:52:47<1:30:52,  3.51it/s]
 95%|█████████▍| 352345/371472 [5:52:47<1:29:58,  3.54it/s]
 95%|█████████▍| 352346/371472 [5:52:48<1:26:21,  3.69it/s]
 95%|█████████▍| 352347/371472 [5:52:48<1:28:49,  3.59it/s]
 95%|█████████▍| 352348/371472 [5:52:48<1:26:16,  3.69it/s]
 95%|█████████▍| 352349/371472 [5:52:48<1:25:09,  3.74it/s]
 95%|█████████▍| 352350/371472 [5:52:49<1:23:34,  3.81it/s]
 95%|█████████▍| 352351/371472 [5:52:49<1:23:33,  3.81it/s]
 95%|█████████▍| 352352/371472 [5:52:49<1:22:33,  3.86it/s]
 95%|█████████▍| 352353/371472 [5:52:49<1:22:43,  3.85it/s]
 95%|█████████▍| 352354/371472 [5:52:50<1:29:31,  3.56it/s]
 95%|█████████▍| 352355/371472 [5:52:50<1:27:19,  3.65it/s]
 95%|█████████▍| 352356/371472 [5:52:50<1:26:35,  3.68it/s]
 95%|█████████▍| 352357/371472 [5:52:50<1:23:50,  3.80it/s]
 95%|█████████▍| 352358/371472 [5:52:51<1:24:07,  3.79it/s]
 95%|█████████▍| 352359/371472 [5:52:51<1:25:34,  3.72it/s]
 95%|█████████▍| 352360/371472 [5:52:51<1:24:34,  3.77it/s]
                                                           
{'loss': 2.5462, 'learning_rate': 1.4632937576763123e-07, 'epoch': 15.18}
+
 95%|█████████▍| 352360/371472 [5:52:51<1:24:34,  3.77it/s]
 95%|█████████▍| 352361/371472 [5:52:52<1:24:40,  3.76it/s]
 95%|█████████▍| 352362/371472 [5:52:52<1:29:52,  3.54it/s]
 95%|█████████▍| 352363/371472 [5:52:52<1:30:29,  3.52it/s]
 95%|█████████▍| 352364/371472 [5:52:53<1:49:32,  2.91it/s]
 95%|█████████▍| 352365/371472 [5:52:53<1:44:17,  3.05it/s]
 95%|█████████▍| 352366/371472 [5:52:53<1:40:15,  3.18it/s]
 95%|█████████▍| 352367/371472 [5:52:54<1:35:42,  3.33it/s]
 95%|█████████▍| 352368/371472 [5:52:54<1:34:53,  3.36it/s]
 95%|█████████▍| 352369/371472 [5:52:54<1:33:23,  3.41it/s]
 95%|█████████▍| 352370/371472 [5:52:54<1:34:51,  3.36it/s]
 95%|█████████▍| 352371/371472 [5:52:55<1:31:10,  3.49it/s]
 95%|█████████▍| 352372/371472 [5:52:55<1:30:24,  3.52it/s]
 95%|█████████▍| 352373/371472 [5:52:55<1:29:17,  3.57it/s]
 95%|█████████▍| 352374/371472 [5:52:55<1:25:19,  3.73it/s]
 95%|█████████▍| 352375/371472 [5:52:56<1:26:07,  3.70it/s]
 95%|█████████▍| 352376/371472 [5:52:56<1:27:14,  3.65it/s]
 95%|█████████▍| 352377/371472 [5:52:56<1:30:28,  3.52it/s]
 95%|█████████▍| 352378/371472 [5:52:57<1:28:16,  3.60it/s]
 95%|█████████▍| 352379/371472 [5:52:57<1:24:33,  3.76it/s]
 95%|█████████▍| 352380/371472 [5:52:57<1:24:49,  3.75it/s]
                                                           
{'loss': 2.4691, 'learning_rate': 1.462808937921524e-07, 'epoch': 15.18}
+
 95%|█████████▍| 352380/371472 [5:52:57<1:24:49,  3.75it/s]
 95%|█████████▍| 352381/371472 [5:52:57<1:27:02,  3.66it/s]
 95%|█████████▍| 352382/371472 [5:52:58<1:24:54,  3.75it/s]
 95%|█████████▍| 352383/371472 [5:52:58<1:26:25,  3.68it/s]
 95%|█████████▍| 352384/371472 [5:52:58<1:23:22,  3.82it/s]
 95%|█████████▍| 352385/371472 [5:52:58<1:22:57,  3.83it/s]
 95%|█████████▍| 352386/371472 [5:52:59<1:25:10,  3.73it/s]
 95%|█████████▍| 352387/371472 [5:52:59<1:24:14,  3.78it/s]
 95%|█████████▍| 352388/371472 [5:52:59<1:23:48,  3.80it/s]
 95%|█████████▍| 352389/371472 [5:52:59<1:25:29,  3.72it/s]
 95%|█████████▍| 352390/371472 [5:53:00<1:25:57,  3.70it/s]
 95%|█████████▍| 352391/371472 [5:53:00<1:27:35,  3.63it/s]
 95%|█████████▍| 352392/371472 [5:53:00<1:30:17,  3.52it/s]
 95%|█████████▍| 352393/371472 [5:53:01<1:26:24,  3.68it/s]
 95%|█████████▍| 352394/371472 [5:53:01<1:25:08,  3.73it/s]
 95%|█████████▍| 352395/371472 [5:53:01<1:24:46,  3.75it/s]
 95%|█████████▍| 352396/371472 [5:53:01<1:31:08,  3.49it/s]
 95%|█████████▍| 352397/371472 [5:53:02<1:28:06,  3.61it/s]
 95%|█████████▍| 352398/371472 [5:53:02<1:27:48,  3.62it/s]
 95%|█████████▍| 352399/371472 [5:53:02<1:27:00,  3.65it/s]
 95%|█████████▍| 352400/371472 [5:53:03<1:26:37,  3.67it/s]
                                                           
{'loss': 2.7852, 'learning_rate': 1.4623241181667348e-07, 'epoch': 15.18}
+
 95%|█████████▍| 352400/371472 [5:53:03<1:26:37,  3.67it/s]
 95%|█████████▍| 352401/371472 [5:53:03<1:23:53,  3.79it/s]
 95%|█████████▍| 352402/371472 [5:53:03<1:22:09,  3.87it/s]
 95%|█████████▍| 352403/371472 [5:53:03<1:25:00,  3.74it/s]
 95%|█████████▍| 352404/371472 [5:53:04<1:33:35,  3.40it/s]
 95%|█████████▍| 352405/371472 [5:53:04<1:33:12,  3.41it/s]
 95%|█████████▍| 352406/371472 [5:53:04<1:31:07,  3.49it/s]
 95%|█████████▍| 352407/371472 [5:53:05<1:34:00,  3.38it/s]
 95%|█████████▍| 352408/371472 [5:53:05<1:32:40,  3.43it/s]
 95%|█████████▍| 352409/371472 [5:53:05<1:39:47,  3.18it/s]
 95%|█████████▍| 352410/371472 [5:53:05<1:36:39,  3.29it/s]
 95%|█████████▍| 352411/371472 [5:53:06<1:35:53,  3.31it/s]
 95%|█████████▍| 352412/371472 [5:53:06<1:35:14,  3.34it/s]
 95%|█████████▍| 352413/371472 [5:53:06<1:37:51,  3.25it/s]
 95%|█████████▍| 352414/371472 [5:53:07<1:33:39,  3.39it/s]
 95%|█████████▍| 352415/371472 [5:53:07<1:33:11,  3.41it/s]
 95%|█████████▍| 352416/371472 [5:53:07<1:30:45,  3.50it/s]
 95%|█████████▍| 352417/371472 [5:53:07<1:29:11,  3.56it/s]
 95%|█████████▍| 352418/371472 [5:53:08<1:26:06,  3.69it/s]
 95%|█████████▍| 352419/371472 [5:53:08<1:25:42,  3.70it/s]
 95%|█████████▍| 352420/371472 [5:53:08<1:26:51,  3.66it/s]
                                                           
{'loss': 2.5497, 'learning_rate': 1.4618392984119463e-07, 'epoch': 15.18}
+
 95%|█████████▍| 352420/371472 [5:53:08<1:26:51,  3.66it/s]
 95%|█████████▍| 352421/371472 [5:53:09<1:29:03,  3.57it/s]
 95%|█████████▍| 352422/371472 [5:53:09<1:30:06,  3.52it/s]
 95%|█████████▍| 352423/371472 [5:53:09<1:30:07,  3.52it/s]
 95%|█████████▍| 352424/371472 [5:53:09<1:32:05,  3.45it/s]
 95%|█████████▍| 352425/371472 [5:53:10<1:29:46,  3.54it/s]
 95%|█████████▍| 352426/371472 [5:53:10<1:32:44,  3.42it/s]
 95%|█████████▍| 352427/371472 [5:53:10<1:34:48,  3.35it/s]
 95%|█████████▍| 352428/371472 [5:53:11<1:31:06,  3.48it/s]
 95%|█████████▍| 352429/371472 [5:53:11<1:31:38,  3.46it/s]
 95%|█████████▍| 352430/371472 [5:53:11<1:36:59,  3.27it/s]
 95%|█████████▍| 352431/371472 [5:53:12<1:36:10,  3.30it/s]
 95%|█████████▍| 352432/371472 [5:53:12<1:36:47,  3.28it/s]
 95%|█████████▍| 352433/371472 [5:53:12<1:31:30,  3.47it/s]
 95%|█████████▍| 352434/371472 [5:53:12<1:30:12,  3.52it/s]
 95%|█████████▍| 352435/371472 [5:53:13<1:30:13,  3.52it/s]
 95%|█████████▍| 352436/371472 [5:53:13<1:28:39,  3.58it/s]
 95%|█████████▍| 352437/371472 [5:53:13<1:28:22,  3.59it/s]
 95%|█████████▍| 352438/371472 [5:53:13<1:31:02,  3.48it/s]
 95%|█████████▍| 352439/371472 [5:53:14<1:31:21,  3.47it/s]
 95%|█████████▍| 352440/371472 [5:53:14<1:30:20,  3.51it/s]
                                                           
{'loss': 2.5512, 'learning_rate': 1.461354478657157e-07, 'epoch': 15.18}
+
 95%|█████████▍| 352440/371472 [5:53:14<1:30:20,  3.51it/s]
 95%|█████████▍| 352441/371472 [5:53:14<1:38:46,  3.21it/s]
 95%|█████████▍| 352442/371472 [5:53:15<1:33:29,  3.39it/s]
 95%|█████████▍| 352443/371472 [5:53:15<1:32:50,  3.42it/s]
 95%|█████████▍| 352444/371472 [5:53:15<1:31:03,  3.48it/s]
 95%|█████████▍| 352445/371472 [5:53:16<1:29:34,  3.54it/s]
 95%|█████████▍| 352446/371472 [5:53:16<1:54:48,  2.76it/s]
 95%|█████████▍| 352447/371472 [5:53:16<1:46:50,  2.97it/s]
 95%|█████████▍| 352448/371472 [5:53:17<1:39:14,  3.19it/s]
 95%|█████████▍| 352449/371472 [5:53:17<1:34:31,  3.35it/s]
 95%|█████████▍| 352450/371472 [5:53:17<1:35:10,  3.33it/s]
 95%|█████████▍| 352451/371472 [5:53:17<1:33:45,  3.38it/s]
 95%|█████████▍| 352452/371472 [5:53:18<1:30:00,  3.52it/s]
 95%|█████████▍| 352453/371472 [5:53:18<1:31:56,  3.45it/s]
 95%|█████████▍| 352454/371472 [5:53:18<1:30:43,  3.49it/s]
 95%|█████████▍| 352455/371472 [5:53:19<1:30:17,  3.51it/s]
 95%|█████████▍| 352456/371472 [5:53:19<1:28:04,  3.60it/s]
 95%|█████████▍| 352457/371472 [5:53:19<1:45:53,  2.99it/s]
 95%|█████████▍| 352458/371472 [5:53:20<1:42:08,  3.10it/s]
 95%|█████████▍| 352459/371472 [5:53:20<1:38:48,  3.21it/s]
 95%|█████████▍| 352460/371472 [5:53:20<1:37:31,  3.25it/s]
                                                           
{'loss': 2.6826, 'learning_rate': 1.4608696589023685e-07, 'epoch': 15.18}
+
 95%|█████████▍| 352460/371472 [5:53:20<1:37:31,  3.25it/s]
 95%|█████████▍| 352461/371472 [5:53:20<1:35:35,  3.31it/s]
 95%|█████████▍| 352462/371472 [5:53:21<1:31:36,  3.46it/s]
 95%|█████████▍| 352463/371472 [5:53:21<1:32:15,  3.43it/s]
 95%|█████████▍| 352464/371472 [5:53:21<1:37:33,  3.25it/s]
 95%|█████████▍| 352465/371472 [5:53:22<1:34:57,  3.34it/s]
 95%|█████████▍| 352466/371472 [5:53:22<1:30:56,  3.48it/s]
 95%|█████████▍| 352467/371472 [5:53:22<1:28:49,  3.57it/s]
 95%|█████████▍| 352468/371472 [5:53:22<1:26:07,  3.68it/s]
 95%|█████████▍| 352469/371472 [5:53:23<1:32:29,  3.42it/s]
 95%|█████████▍| 352470/371472 [5:53:23<1:31:25,  3.46it/s]
 95%|█████████▍| 352471/371472 [5:53:23<1:29:46,  3.53it/s]
 95%|█████████▍| 352472/371472 [5:53:24<1:30:34,  3.50it/s]
 95%|█████████▍| 352473/371472 [5:53:24<1:29:41,  3.53it/s]
 95%|█████████▍| 352474/371472 [5:53:24<1:28:19,  3.58it/s]
 95%|█████████▍| 352475/371472 [5:53:24<1:28:03,  3.60it/s]
 95%|█████████▍| 352476/371472 [5:53:25<1:27:18,  3.63it/s]
 95%|█████████▍| 352477/371472 [5:53:25<1:24:57,  3.73it/s]
 95%|█████████▍| 352478/371472 [5:53:25<1:26:09,  3.67it/s]
 95%|█████████▍| 352479/371472 [5:53:26<1:25:04,  3.72it/s]
 95%|█████████▍| 352480/371472 [5:53:26<1:27:57,  3.60it/s]
                                                           
{'loss': 2.5204, 'learning_rate': 1.460384839147579e-07, 'epoch': 15.18}
+
 95%|█████████▍| 352480/371472 [5:53:26<1:27:57,  3.60it/s]
 95%|█████████▍| 352481/371472 [5:53:26<1:28:33,  3.57it/s]
 95%|█████████▍| 352482/371472 [5:53:26<1:27:08,  3.63it/s]
 95%|█████████▍| 352483/371472 [5:53:27<1:27:13,  3.63it/s]
 95%|█████████▍| 352484/371472 [5:53:27<1:28:47,  3.56it/s]
 95%|█████████▍| 352485/371472 [5:53:27<1:26:45,  3.65it/s]
 95%|█████████▍| 352486/371472 [5:53:27<1:26:13,  3.67it/s]
 95%|█████████▍| 352487/371472 [5:53:28<1:27:47,  3.60it/s]
 95%|█████████▍| 352488/371472 [5:53:28<1:30:19,  3.50it/s]
 95%|█████████▍| 352489/371472 [5:53:28<1:26:51,  3.64it/s]
 95%|█████████▍| 352490/371472 [5:53:29<1:25:36,  3.70it/s]
 95%|█████████▍| 352491/371472 [5:53:29<1:25:56,  3.68it/s]
 95%|█████████▍| 352492/371472 [5:53:29<1:41:59,  3.10it/s]
 95%|█████████▍| 352493/371472 [5:53:30<1:37:22,  3.25it/s]
 95%|█████████▍| 352494/371472 [5:53:30<1:35:29,  3.31it/s]
 95%|█████████▍| 352495/371472 [5:53:30<1:30:13,  3.51it/s]
 95%|█████████▍| 352496/371472 [5:53:30<1:30:18,  3.50it/s]
 95%|█████████▍| 352497/371472 [5:53:31<1:27:16,  3.62it/s]
 95%|█████████▍| 352498/371472 [5:53:31<1:26:46,  3.64it/s]
 95%|█████████▍| 352499/371472 [5:53:31<1:27:28,  3.61it/s]
 95%|█████████▍| 352500/371472 [5:53:31<1:27:31,  3.61it/s]
                                                           
{'loss': 2.5001, 'learning_rate': 1.4599000193927907e-07, 'epoch': 15.18}
+
 95%|█████████▍| 352500/371472 [5:53:31<1:27:31,  3.61it/s]
 95%|█████████▍| 352501/371472 [5:53:32<1:38:42,  3.20it/s]
 95%|█████████▍| 352502/371472 [5:53:32<1:32:42,  3.41it/s]
 95%|█████████▍| 352503/371472 [5:53:32<1:35:59,  3.29it/s]
 95%|█████████▍| 352504/371472 [5:53:33<1:30:30,  3.49it/s]
 95%|█████████▍| 352505/371472 [5:53:33<1:30:22,  3.50it/s]
 95%|█████████▍| 352506/371472 [5:53:33<1:27:55,  3.59it/s]
 95%|█████████▍| 352507/371472 [5:53:33<1:28:19,  3.58it/s]
 95%|█████████▍| 352508/371472 [5:53:34<1:25:47,  3.68it/s]
 95%|█████████▍| 352509/371472 [5:53:34<1:29:42,  3.52it/s]
 95%|█████████▍| 352510/371472 [5:53:34<1:30:25,  3.49it/s]
 95%|█████████▍| 352511/371472 [5:53:35<1:28:38,  3.57it/s]
 95%|█████████▍| 352512/371472 [5:53:35<1:26:54,  3.64it/s]
 95%|█████████▍| 352513/371472 [5:53:35<1:26:45,  3.64it/s]
 95%|█████████▍| 352514/371472 [5:53:36<1:35:21,  3.31it/s]
 95%|█████████▍| 352515/371472 [5:53:36<1:31:30,  3.45it/s]
 95%|█████████▍| 352516/371472 [5:53:36<1:29:59,  3.51it/s]
 95%|█████████▍| 352517/371472 [5:53:36<1:30:54,  3.48it/s]
 95%|█████████▍| 352518/371472 [5:53:37<1:28:38,  3.56it/s]
 95%|█████████▍| 352519/371472 [5:53:37<1:27:55,  3.59it/s]
 95%|█████████▍| 352520/371472 [5:53:37<1:31:03,  3.47it/s]
                                                           
{'loss': 2.5178, 'learning_rate': 1.4594151996380012e-07, 'epoch': 15.18}
+
 95%|█████████▍| 352520/371472 [5:53:37<1:31:03,  3.47it/s]
 95%|█████████▍| 352521/371472 [5:53:37<1:29:53,  3.51it/s]
 95%|█████████▍| 352522/371472 [5:53:38<1:32:14,  3.42it/s]
 95%|█████████▍| 352523/371472 [5:53:38<1:33:35,  3.37it/s]
 95%|█████████▍| 352524/371472 [5:53:38<1:31:42,  3.44it/s]
 95%|█████████▍| 352525/371472 [5:53:39<1:31:28,  3.45it/s]
 95%|█████████▍| 352526/371472 [5:53:39<1:29:16,  3.54it/s]
 95%|█████████▍| 352527/371472 [5:53:39<1:30:56,  3.47it/s]
 95%|█████████▍| 352528/371472 [5:53:39<1:29:38,  3.52it/s]
 95%|█████████▍| 352529/371472 [5:53:40<1:27:36,  3.60it/s]
 95%|█████████▍| 352530/371472 [5:53:40<1:24:17,  3.75it/s]
 95%|█████████▍| 352531/371472 [5:53:40<1:22:27,  3.83it/s]
 95%|█████████▍| 352532/371472 [5:53:40<1:20:39,  3.91it/s]
 95%|█████████▍| 352533/371472 [5:53:41<1:20:45,  3.91it/s]
 95%|█████████▍| 352534/371472 [5:53:41<1:21:37,  3.87it/s]
 95%|█████████▍| 352535/371472 [5:53:41<1:21:55,  3.85it/s]
 95%|█████████▍| 352536/371472 [5:53:42<1:22:55,  3.81it/s]
 95%|█████████▍| 352537/371472 [5:53:42<1:24:03,  3.75it/s]
 95%|█████████▍| 352538/371472 [5:53:42<1:24:53,  3.72it/s]
 95%|█████████▍| 352539/371472 [5:53:42<1:32:15,  3.42it/s]
 95%|█████████▍| 352540/371472 [5:53:43<1:29:46,  3.51it/s]
                                                           
{'loss': 2.4805, 'learning_rate': 1.4589303798832116e-07, 'epoch': 15.18}
+
 95%|█████████▍| 352540/371472 [5:53:43<1:29:46,  3.51it/s]
 95%|█████████▍| 352541/371472 [5:53:43<1:27:42,  3.60it/s]
 95%|█████████▍| 352542/371472 [5:53:43<1:28:49,  3.55it/s]
 95%|█████████▍| 352543/371472 [5:53:44<1:27:08,  3.62it/s]
 95%|█████████▍| 352544/371472 [5:53:44<1:26:52,  3.63it/s]
 95%|█████████▍| 352545/371472 [5:53:44<1:25:25,  3.69it/s]
 95%|█████████▍| 352546/371472 [5:53:44<1:39:32,  3.17it/s]
 95%|█████████▍| 352547/371472 [5:53:45<1:33:47,  3.36it/s]
 95%|█████████▍| 352548/371472 [5:53:45<1:34:38,  3.33it/s]
 95%|█████████▍| 352549/371472 [5:53:45<1:35:53,  3.29it/s]
 95%|█████████▍| 352550/371472 [5:53:46<1:33:46,  3.36it/s]
 95%|█████████▍| 352551/371472 [5:53:46<1:28:57,  3.55it/s]
 95%|█████████▍| 352552/371472 [5:53:46<1:30:07,  3.50it/s]
 95%|█████████▍| 352553/371472 [5:53:46<1:29:29,  3.52it/s]
 95%|█████████▍| 352554/371472 [5:53:47<1:27:33,  3.60it/s]
 95%|█████████▍| 352555/371472 [5:53:47<1:25:59,  3.67it/s]
 95%|█████████▍| 352556/371472 [5:53:47<1:32:13,  3.42it/s]
 95%|█████████▍| 352557/371472 [5:53:48<1:27:18,  3.61it/s]
 95%|█████████▍| 352558/371472 [5:53:48<1:27:58,  3.58it/s]
 95%|█████████▍| 352559/371472 [5:53:48<1:24:38,  3.72it/s]
 95%|█████████▍| 352560/371472 [5:53:48<1:27:09,  3.62it/s]
                                                           
{'loss': 2.5173, 'learning_rate': 1.4584455601284234e-07, 'epoch': 15.19}
+
 95%|█████████▍| 352560/371472 [5:53:48<1:27:09,  3.62it/s]
 95%|█████████▍| 352561/371472 [5:53:49<1:23:27,  3.78it/s]
 95%|█████████▍| 352562/371472 [5:53:49<1:25:19,  3.69it/s]
 95%|█████████▍| 352563/371472 [5:53:49<1:29:56,  3.50it/s]
 95%|█████████▍| 352564/371472 [5:53:49<1:25:37,  3.68it/s]
 95%|█████████▍| 352565/371472 [5:53:50<1:28:22,  3.57it/s]
 95%|█████████▍| 352566/371472 [5:53:50<1:25:03,  3.70it/s]
 95%|█████████▍| 352567/371472 [5:53:50<1:25:16,  3.70it/s]
 95%|█████████▍| 352568/371472 [5:53:51<1:27:20,  3.61it/s]
 95%|█████████▍| 352569/371472 [5:53:51<1:24:26,  3.73it/s]
 95%|█████████▍| 352570/371472 [5:53:51<1:28:38,  3.55it/s]
 95%|█████████▍| 352571/371472 [5:53:51<1:31:24,  3.45it/s]
 95%|█████████▍| 352572/371472 [5:53:52<1:31:43,  3.43it/s]
 95%|█████████▍| 352573/371472 [5:53:52<1:27:04,  3.62it/s]
 95%|█████████▍| 352574/371472 [5:53:52<1:25:47,  3.67it/s]
 95%|█████████▍| 352575/371472 [5:53:53<1:35:32,  3.30it/s]
 95%|█████████▍| 352576/371472 [5:53:53<1:33:24,  3.37it/s]
 95%|█████████▍| 352577/371472 [5:53:53<1:30:15,  3.49it/s]
 95%|█████████▍| 352578/371472 [5:53:53<1:28:07,  3.57it/s]
 95%|█████████▍| 352579/371472 [5:53:54<1:27:24,  3.60it/s]
 95%|█████████▍| 352580/371472 [5:53:54<1:45:37,  2.98it/s]
                                                           
{'loss': 2.435, 'learning_rate': 1.4579607403736339e-07, 'epoch': 15.19}
+
 95%|█████████▍| 352580/371472 [5:53:54<1:45:37,  2.98it/s]
 95%|█████████▍| 352581/371472 [5:53:54<1:43:12,  3.05it/s]
 95%|█████████▍| 352582/371472 [5:53:55<1:35:15,  3.31it/s]
 95%|█████████▍| 352583/371472 [5:53:55<1:30:03,  3.50it/s]
 95%|█████████▍| 352584/371472 [5:53:55<1:26:13,  3.65it/s]
 95%|█████████▍| 352585/371472 [5:53:56<1:32:18,  3.41it/s]
 95%|█████████▍| 352586/371472 [5:53:56<1:30:48,  3.47it/s]
 95%|█████████▍| 352587/371472 [5:53:56<1:30:14,  3.49it/s]
 95%|█████████▍| 352588/371472 [5:53:56<1:38:58,  3.18it/s]
 95%|█████████▍| 352589/371472 [5:53:57<1:41:18,  3.11it/s]
 95%|█████████▍| 352590/371472 [5:53:57<1:42:08,  3.08it/s]
 95%|█████████▍| 352591/371472 [5:53:57<1:39:23,  3.17it/s]
 95%|█████████▍| 352592/371472 [5:53:58<1:37:13,  3.24it/s]
 95%|█████████▍| 352593/371472 [5:53:58<1:34:29,  3.33it/s]
 95%|█████████▍| 352594/371472 [5:53:58<1:32:47,  3.39it/s]
 95%|█████████▍| 352595/371472 [5:53:59<1:31:50,  3.43it/s]
 95%|█████████▍| 352596/371472 [5:53:59<1:34:23,  3.33it/s]
 95%|█████████▍| 352597/371472 [5:53:59<1:32:19,  3.41it/s]
 95%|█████████▍| 352598/371472 [5:53:59<1:31:09,  3.45it/s]
 95%|█████████▍| 352599/371472 [5:54:00<1:31:29,  3.44it/s]
 95%|█████████▍| 352600/371472 [5:54:00<1:30:13,  3.49it/s]
                                                           
{'loss': 2.5328, 'learning_rate': 1.4574759206188454e-07, 'epoch': 15.19}
+
 95%|█████████▍| 352600/371472 [5:54:00<1:30:13,  3.49it/s]
 95%|█████████▍| 352601/371472 [5:54:00<1:27:50,  3.58it/s]
 95%|█████████▍| 352602/371472 [5:54:01<1:25:52,  3.66it/s]
 95%|█████████▍| 352603/371472 [5:54:01<1:25:46,  3.67it/s]
 95%|█████████▍| 352604/371472 [5:54:01<1:23:38,  3.76it/s]
 95%|█████████▍| 352605/371472 [5:54:01<1:29:07,  3.53it/s]
 95%|█████████▍| 352606/371472 [5:54:02<1:29:06,  3.53it/s]
 95%|█████████▍| 352607/371472 [5:54:02<1:32:38,  3.39it/s]
 95%|█████████▍| 352608/371472 [5:54:02<1:27:41,  3.59it/s]
 95%|█████████▍| 352609/371472 [5:54:03<1:26:42,  3.63it/s]
 95%|█████████▍| 352610/371472 [5:54:03<1:25:53,  3.66it/s]
 95%|█████████▍| 352611/371472 [5:54:03<1:25:25,  3.68it/s]
 95%|█████████▍| 352612/371472 [5:54:03<1:28:42,  3.54it/s]
 95%|█████████▍| 352613/371472 [5:54:04<1:36:25,  3.26it/s]
 95%|█████████▍| 352614/371472 [5:54:04<1:32:03,  3.41it/s]
 95%|█████████▍| 352615/371472 [5:54:04<1:28:37,  3.55it/s]
 95%|█████████▍| 352616/371472 [5:54:05<1:29:13,  3.52it/s]
 95%|█████████▍| 352617/371472 [5:54:05<1:27:50,  3.58it/s]
 95%|█████████▍| 352618/371472 [5:54:05<1:32:32,  3.40it/s]
 95%|█████████▍| 352619/371472 [5:54:05<1:30:03,  3.49it/s]
 95%|█████████▍| 352620/371472 [5:54:06<1:28:09,  3.56it/s]
                                                           
{'loss': 2.6025, 'learning_rate': 1.456991100864056e-07, 'epoch': 15.19}
+
 95%|█████████▍| 352620/371472 [5:54:06<1:28:09,  3.56it/s]
 95%|█████████▍| 352621/371472 [5:54:06<1:24:38,  3.71it/s]
 95%|█████████▍| 352622/371472 [5:54:06<1:23:55,  3.74it/s]
 95%|█████████▍| 352623/371472 [5:54:06<1:23:36,  3.76it/s]
 95%|█████████▍| 352624/371472 [5:54:07<1:26:21,  3.64it/s]
 95%|█████████▍| 352625/371472 [5:54:07<1:24:32,  3.72it/s]
 95%|█████████▍| 352626/371472 [5:54:07<1:23:07,  3.78it/s]
 95%|█████████▍| 352627/371472 [5:54:08<1:22:37,  3.80it/s]
 95%|█████████▍| 352628/371472 [5:54:08<1:21:48,  3.84it/s]
 95%|█████████▍| 352629/371472 [5:54:08<1:24:07,  3.73it/s]
 95%|█████████▍| 352630/371472 [5:54:08<1:25:46,  3.66it/s]
 95%|█████████▍| 352631/371472 [5:54:09<1:29:26,  3.51it/s]
 95%|█████████▍| 352632/371472 [5:54:09<1:29:58,  3.49it/s]
 95%|█████████▍| 352633/371472 [5:54:09<1:29:54,  3.49it/s]
 95%|█████████▍| 352634/371472 [5:54:09<1:26:15,  3.64it/s]
 95%|█████████▍| 352635/371472 [5:54:10<1:24:40,  3.71it/s]
 95%|█████████▍| 352636/371472 [5:54:10<1:26:38,  3.62it/s]
 95%|█████████▍| 352637/371472 [5:54:10<1:27:18,  3.60it/s]
 95%|█████████▍| 352638/371472 [5:54:11<1:26:44,  3.62it/s]
 95%|█████████▍| 352639/371472 [5:54:11<1:27:55,  3.57it/s]
 95%|█████████▍| 352640/371472 [5:54:11<1:30:22,  3.47it/s]
                                                           
{'loss': 2.5481, 'learning_rate': 1.4565062811092676e-07, 'epoch': 15.19}
+
 95%|█████████▍| 352640/371472 [5:54:11<1:30:22,  3.47it/s]
 95%|█████████▍| 352641/371472 [5:54:11<1:27:26,  3.59it/s]
 95%|█████████▍| 352642/371472 [5:54:12<1:26:21,  3.63it/s]
 95%|█████████▍| 352643/371472 [5:54:12<1:25:23,  3.68it/s]
 95%|█████████▍| 352644/371472 [5:54:12<1:25:25,  3.67it/s]
 95%|█████████▍| 352645/371472 [5:54:13<1:28:33,  3.54it/s]
 95%|█████████▍| 352646/371472 [5:54:13<1:26:25,  3.63it/s]
 95%|█████████▍| 352647/371472 [5:54:13<1:27:22,  3.59it/s]
 95%|█████████▍| 352648/371472 [5:54:13<1:30:41,  3.46it/s]
 95%|█████████▍| 352649/371472 [5:54:14<1:35:31,  3.28it/s]
 95%|█████████▍| 352650/371472 [5:54:14<1:34:53,  3.31it/s]
 95%|█████████▍| 352651/371472 [5:54:14<1:32:05,  3.41it/s]
 95%|█████████▍| 352652/371472 [5:54:15<1:28:10,  3.56it/s]
 95%|█████████▍| 352653/371472 [5:54:15<1:27:35,  3.58it/s]
 95%|█████████▍| 352654/371472 [5:54:15<1:25:42,  3.66it/s]
 95%|█████████▍| 352655/371472 [5:54:15<1:25:08,  3.68it/s]
 95%|█████████▍| 352656/371472 [5:54:16<1:27:15,  3.59it/s]
 95%|█████████▍| 352657/371472 [5:54:16<1:25:33,  3.66it/s]
 95%|█████████▍| 352658/371472 [5:54:16<1:30:57,  3.45it/s]
 95%|█████████▍| 352659/371472 [5:54:17<1:37:09,  3.23it/s]
 95%|█████████▍| 352660/371472 [5:54:17<1:47:43,  2.91it/s]
                                                           
{'loss': 2.5107, 'learning_rate': 1.4560214613544783e-07, 'epoch': 15.19}
+
 95%|█████████▍| 352660/371472 [5:54:17<1:47:43,  2.91it/s]
 95%|█████████▍| 352661/371472 [5:54:17<1:42:14,  3.07it/s]
 95%|█████████▍| 352662/371472 [5:54:18<1:36:07,  3.26it/s]
 95%|█████████▍| 352663/371472 [5:54:18<1:31:28,  3.43it/s]
 95%|█████████▍| 352664/371472 [5:54:18<1:29:20,  3.51it/s]
 95%|█████████▍| 352665/371472 [5:54:18<1:32:25,  3.39it/s]
 95%|█████████▍| 352666/371472 [5:54:19<1:28:16,  3.55it/s]
 95%|█████████▍| 352667/371472 [5:54:19<1:25:50,  3.65it/s]
 95%|█████████▍| 352668/371472 [5:54:19<1:31:39,  3.42it/s]
 95%|█████████▍| 352669/371472 [5:54:20<1:31:39,  3.42it/s]
 95%|█████████▍| 352670/371472 [5:54:20<1:29:21,  3.51it/s]
 95%|█████████▍| 352671/371472 [5:54:20<1:30:14,  3.47it/s]
 95%|█████████▍| 352672/371472 [5:54:21<1:42:00,  3.07it/s]
 95%|█████████▍| 352673/371472 [5:54:21<1:39:23,  3.15it/s]
 95%|█████████▍| 352674/371472 [5:54:21<1:33:48,  3.34it/s]
 95%|█████████▍| 352675/371472 [5:54:21<1:33:05,  3.37it/s]
 95%|█████████▍| 352676/371472 [5:54:22<1:28:40,  3.53it/s]
 95%|█████████▍| 352677/371472 [5:54:22<1:28:59,  3.52it/s]
 95%|█████████▍| 352678/371472 [5:54:22<1:25:45,  3.65it/s]
 95%|█████████▍| 352679/371472 [5:54:22<1:27:28,  3.58it/s]
 95%|█████████▍| 352680/371472 [5:54:23<1:27:58,  3.56it/s]
                                                           
{'loss': 2.5927, 'learning_rate': 1.4555366415996898e-07, 'epoch': 15.19}
+
 95%|█████████▍| 352680/371472 [5:54:23<1:27:58,  3.56it/s]
 95%|█████████▍| 352681/371472 [5:54:23<1:27:15,  3.59it/s]
 95%|█████████▍| 352682/371472 [5:54:23<1:28:14,  3.55it/s]
 95%|█████████▍| 352683/371472 [5:54:24<1:25:28,  3.66it/s]
 95%|█████████▍| 352684/371472 [5:54:24<1:31:09,  3.44it/s]
 95%|█████████▍| 352685/371472 [5:54:24<1:30:18,  3.47it/s]
 95%|█████████▍| 352686/371472 [5:54:24<1:29:34,  3.50it/s]
 95%|█████████▍| 352687/371472 [5:54:25<1:28:14,  3.55it/s]
 95%|█████████▍| 352688/371472 [5:54:25<1:33:35,  3.34it/s]
 95%|█████████▍| 352689/371472 [5:54:25<1:32:08,  3.40it/s]
 95%|█████████▍| 352690/371472 [5:54:26<1:27:57,  3.56it/s]
 95%|█████████▍| 352691/371472 [5:54:26<1:26:18,  3.63it/s]
 95%|█████████▍| 352692/371472 [5:54:26<1:42:28,  3.05it/s]
 95%|█████████▍| 352693/371472 [5:54:27<1:41:49,  3.07it/s]
 95%|█████████▍| 352694/371472 [5:54:27<1:34:22,  3.32it/s]
 95%|█████████▍| 352695/371472 [5:54:27<1:35:20,  3.28it/s]
 95%|█████████▍| 352696/371472 [5:54:27<1:31:16,  3.43it/s]
 95%|█████████▍| 352697/371472 [5:54:28<1:37:38,  3.20it/s]
 95%|█████████▍| 352698/371472 [5:54:28<1:33:35,  3.34it/s]
 95%|█████████▍| 352699/371472 [5:54:28<1:28:54,  3.52it/s]
 95%|█████████▍| 352700/371472 [5:54:29<1:30:45,  3.45it/s]
                                                           
{'loss': 2.5984, 'learning_rate': 1.4550518218449008e-07, 'epoch': 15.19}
+
 95%|█████████▍| 352700/371472 [5:54:29<1:30:45,  3.45it/s]
 95%|█████████▍| 352701/371472 [5:54:29<1:28:42,  3.53it/s]
 95%|█████████▍| 352702/371472 [5:54:29<1:27:11,  3.59it/s]
 95%|█████████▍| 352703/371472 [5:54:29<1:26:13,  3.63it/s]
 95%|█████████▍| 352704/371472 [5:54:30<1:23:17,  3.76it/s]
 95%|█████████▍| 352705/371472 [5:54:30<1:21:40,  3.83it/s]
 95%|█████████▍| 352706/371472 [5:54:30<1:20:35,  3.88it/s]
 95%|█████████▍| 352707/371472 [5:54:31<1:28:47,  3.52it/s]
 95%|█████████▍| 352708/371472 [5:54:31<1:29:46,  3.48it/s]
 95%|█████████▍| 352709/371472 [5:54:31<1:28:52,  3.52it/s]
 95%|█████████▍| 352710/371472 [5:54:31<1:34:58,  3.29it/s]
 95%|█████████▍| 352711/371472 [5:54:32<1:37:32,  3.21it/s]
 95%|█████████▍| 352712/371472 [5:54:32<1:33:57,  3.33it/s]
 95%|█████████▍| 352713/371472 [5:54:32<1:30:53,  3.44it/s]
 95%|█████████▍| 352714/371472 [5:54:33<1:33:15,  3.35it/s]
 95%|█████████▍| 352715/371472 [5:54:33<1:29:24,  3.50it/s]
 95%|���████████▍| 352716/371472 [5:54:33<1:39:30,  3.14it/s]
 95%|█████████▍| 352717/371472 [5:54:34<1:38:12,  3.18it/s]
 95%|█████████▍| 352718/371472 [5:54:34<1:34:33,  3.31it/s]
 95%|█████████▍| 352719/371472 [5:54:34<1:33:40,  3.34it/s]
 95%|█████████▍| 352720/371472 [5:54:34<1:29:21,  3.50it/s]
                                                           
{'loss': 2.6067, 'learning_rate': 1.454567002090112e-07, 'epoch': 15.19}
+
 95%|█████████▍| 352720/371472 [5:54:34<1:29:21,  3.50it/s]
 95%|█████████▍| 352721/371472 [5:54:35<1:29:48,  3.48it/s]
 95%|█████████▍| 352722/371472 [5:54:35<1:27:13,  3.58it/s]
 95%|█████████▍| 352723/371472 [5:54:35<1:26:36,  3.61it/s]
 95%|█████████▍| 352724/371472 [5:54:35<1:26:26,  3.62it/s]
 95%|█████████▍| 352725/371472 [5:54:36<1:26:14,  3.62it/s]
 95%|█████████▍| 352726/371472 [5:54:36<1:23:56,  3.72it/s]
 95%|█████████▍| 352727/371472 [5:54:36<1:25:31,  3.65it/s]
 95%|█████████▍| 352728/371472 [5:54:37<1:25:02,  3.67it/s]
 95%|█████████▍| 352729/371472 [5:54:37<1:23:42,  3.73it/s]
 95%|█████████▍| 352730/371472 [5:54:37<1:42:47,  3.04it/s]
 95%|█████████▍| 352731/371472 [5:54:38<1:42:53,  3.04it/s]
 95%|█████████▍| 352732/371472 [5:54:38<1:36:48,  3.23it/s]
 95%|█████████▍| 352733/371472 [5:54:38<1:37:11,  3.21it/s]
 95%|█████████▍| 352734/371472 [5:54:38<1:32:32,  3.37it/s]
 95%|█████████▍| 352735/371472 [5:54:39<1:30:32,  3.45it/s]
 95%|█████████▍| 352736/371472 [5:54:39<1:27:09,  3.58it/s]
 95%|█████████▍| 352737/371472 [5:54:39<1:30:35,  3.45it/s]
 95%|█████████▍| 352738/371472 [5:54:40<1:28:51,  3.51it/s]
 95%|█████████▍| 352739/371472 [5:54:40<1:27:23,  3.57it/s]
 95%|█████████▍| 352740/371472 [5:54:40<1:28:43,  3.52it/s]
                                                           
{'loss': 2.7445, 'learning_rate': 1.4540821823353225e-07, 'epoch': 15.19}
+
 95%|█████████▍| 352740/371472 [5:54:40<1:28:43,  3.52it/s]
 95%|█████████▍| 352741/371472 [5:54:40<1:27:22,  3.57it/s]
 95%|█████████▍| 352742/371472 [5:54:41<1:23:48,  3.72it/s]
 95%|█████████▍| 352743/371472 [5:54:41<1:31:45,  3.40it/s]
 95%|█████████▍| 352744/371472 [5:54:41<1:30:21,  3.45it/s]
 95%|█████████▍| 352745/371472 [5:54:42<1:27:29,  3.57it/s]
 95%|█████████▍| 352746/371472 [5:54:42<1:27:57,  3.55it/s]
 95%|█████████▍| 352747/371472 [5:54:42<1:25:00,  3.67it/s]
 95%|█████████▍| 352748/371472 [5:54:42<1:30:34,  3.45it/s]
 95%|█████████▍| 352749/371472 [5:54:43<1:27:43,  3.56it/s]
 95%|█████████▍| 352750/371472 [5:54:43<1:27:17,  3.57it/s]
 95%|█████████▍| 352751/371472 [5:54:43<1:28:20,  3.53it/s]
 95%|█████████▍| 352752/371472 [5:54:44<1:27:38,  3.56it/s]
 95%|█████████▍| 352753/371472 [5:54:44<1:26:34,  3.60it/s]
 95%|█████████▍| 352754/371472 [5:54:44<1:25:53,  3.63it/s]
 95%|█████████▍| 352755/371472 [5:54:44<1:28:57,  3.51it/s]
 95%|█████████▍| 352756/371472 [5:54:45<1:34:04,  3.32it/s]
 95%|█████████▍| 352757/371472 [5:54:45<1:35:38,  3.26it/s]
 95%|█████████▍| 352758/371472 [5:54:45<1:31:28,  3.41it/s]
 95%|█████████▍| 352759/371472 [5:54:46<1:31:50,  3.40it/s]
 95%|█████████▍| 352760/371472 [5:54:46<1:29:06,  3.50it/s]
                                                           
{'loss': 2.6974, 'learning_rate': 1.4535973625805343e-07, 'epoch': 15.19}
+
 95%|█████████▍| 352760/371472 [5:54:46<1:29:06,  3.50it/s]
 95%|█████████▍| 352761/371472 [5:54:46<1:26:48,  3.59it/s]
 95%|█████████▍| 352762/371472 [5:54:46<1:28:09,  3.54it/s]
 95%|█████████▍| 352763/371472 [5:54:47<1:33:43,  3.33it/s]
 95%|█████████▍| 352764/371472 [5:54:47<1:32:42,  3.36it/s]
 95%|█████████▍| 352765/371472 [5:54:47<1:28:19,  3.53it/s]
 95%|█████████▍| 352766/371472 [5:54:48<1:28:27,  3.52it/s]
 95%|█████████▍| 352767/371472 [5:54:48<1:29:31,  3.48it/s]
 95%|█████████▍| 352768/371472 [5:54:48<1:29:46,  3.47it/s]
 95%|█████████▍| 352769/371472 [5:54:48<1:26:12,  3.62it/s]
 95%|█████████▍| 352770/371472 [5:54:49<1:29:46,  3.47it/s]
 95%|█████████▍| 352771/371472 [5:54:49<1:26:24,  3.61it/s]
 95%|█████████▍| 352772/371472 [5:54:49<1:27:22,  3.57it/s]
 95%|█████████▍| 352773/371472 [5:54:50<1:30:25,  3.45it/s]
 95%|█████████▍| 352774/371472 [5:54:50<1:28:42,  3.51it/s]
 95%|█████████▍| 352775/371472 [5:54:50<1:26:14,  3.61it/s]
 95%|█████████▍| 352776/371472 [5:54:50<1:26:05,  3.62it/s]
 95%|█████████▍| 352777/371472 [5:54:51<1:30:01,  3.46it/s]
 95%|█████████▍| 352778/371472 [5:54:51<1:32:46,  3.36it/s]
 95%|█████████▍| 352779/371472 [5:54:51<1:31:13,  3.42it/s]
 95%|█████████▍| 352780/371472 [5:54:52<1:30:07,  3.46it/s]
                                                           
{'loss': 2.5571, 'learning_rate': 1.4531125428257447e-07, 'epoch': 15.19}
+
 95%|█████████▍| 352780/371472 [5:54:52<1:30:07,  3.46it/s]
 95%|█████████▍| 352781/371472 [5:54:52<1:31:13,  3.41it/s]
 95%|█████████▍| 352782/371472 [5:54:52<1:28:27,  3.52it/s]
 95%|█████████▍| 352783/371472 [5:54:52<1:29:18,  3.49it/s]
 95%|█████████▍| 352784/371472 [5:54:53<1:27:22,  3.56it/s]
 95%|█████████▍| 352785/371472 [5:54:53<1:26:27,  3.60it/s]
 95%|█████████▍| 352786/371472 [5:54:53<1:24:52,  3.67it/s]
 95%|█████████▍| 352787/371472 [5:54:53<1:22:58,  3.75it/s]
 95%|█████████▍| 352788/371472 [5:54:54<1:23:08,  3.75it/s]
 95%|█████████▍| 352789/371472 [5:54:54<1:25:32,  3.64it/s]
 95%|█████████▍| 352790/371472 [5:54:54<1:25:11,  3.65it/s]
 95%|█████████▍| 352791/371472 [5:54:55<1:25:02,  3.66it/s]
 95%|█████████▍| 352792/371472 [5:54:55<1:32:24,  3.37it/s]
 95%|█████████▍| 352793/371472 [5:54:55<1:33:41,  3.32it/s]
 95%|█████████▍| 352794/371472 [5:54:55<1:27:43,  3.55it/s]
 95%|█████████▍| 352795/371472 [5:54:56<1:32:36,  3.36it/s]
 95%|█████████▍| 352796/371472 [5:54:56<1:33:38,  3.32it/s]
 95%|█████████▍| 352797/371472 [5:54:56<1:28:02,  3.54it/s]
 95%|█████████▍| 352798/371472 [5:54:57<1:41:29,  3.07it/s]
 95%|█████████▍| 352799/371472 [5:54:57<1:34:55,  3.28it/s]
 95%|█████████▍| 352800/371472 [5:54:57<1:28:59,  3.50it/s]
                                                           
{'loss': 2.6046, 'learning_rate': 1.4526277230709562e-07, 'epoch': 15.2}
+
 95%|█████████▍| 352800/371472 [5:54:57<1:28:59,  3.50it/s]
 95%|█████████▍| 352801/371472 [5:54:58<1:29:10,  3.49it/s]
 95%|█████████▍| 352802/371472 [5:54:58<1:27:57,  3.54it/s]
 95%|█████████▍| 352803/371472 [5:54:58<1:29:27,  3.48it/s]
 95%|█████████▍| 352804/371472 [5:54:58<1:28:17,  3.52it/s]
 95%|█████████▍| 352805/371472 [5:54:59<1:32:38,  3.36it/s]
 95%|█████████▍| 352806/371472 [5:54:59<1:30:23,  3.44it/s]
 95%|█████████▍| 352807/371472 [5:54:59<1:30:23,  3.44it/s]
 95%|█████████▍| 352808/371472 [5:55:00<1:32:25,  3.37it/s]
 95%|█████████▍| 352809/371472 [5:55:00<1:33:23,  3.33it/s]
 95%|█████████▍| 352810/371472 [5:55:00<1:28:55,  3.50it/s]
 95%|█████████▍| 352811/371472 [5:55:01<1:32:42,  3.35it/s]
 95%|█████████▍| 352812/371472 [5:55:01<1:31:57,  3.38it/s]
 95%|█████████▍| 352813/371472 [5:55:01<1:30:40,  3.43it/s]
 95%|█████████▍| 352814/371472 [5:55:01<1:29:26,  3.48it/s]
 95%|█████████▍| 352815/371472 [5:55:02<1:26:37,  3.59it/s]
 95%|█████████▍| 352816/371472 [5:55:02<1:30:29,  3.44it/s]
 95%|█████████▍| 352817/371472 [5:55:02<1:30:43,  3.43it/s]
 95%|█████████▍| 352818/371472 [5:55:03<1:30:23,  3.44it/s]
 95%|█████████▍| 352819/371472 [5:55:03<1:32:17,  3.37it/s]
 95%|█████████▍| 352820/371472 [5:55:03<1:29:25,  3.48it/s]
                                                           
{'loss': 2.5581, 'learning_rate': 1.452142903316167e-07, 'epoch': 15.2}
+
 95%|█████████▍| 352820/371472 [5:55:03<1:29:25,  3.48it/s]
 95%|█████████▍| 352821/371472 [5:55:03<1:26:41,  3.59it/s]
 95%|█████████▍| 352822/371472 [5:55:04<1:27:56,  3.53it/s]
 95%|█████████▍| 352823/371472 [5:55:04<1:25:09,  3.65it/s]
 95%|█████████▍| 352824/371472 [5:55:04<1:26:58,  3.57it/s]
 95%|█████████▍| 352825/371472 [5:55:05<1:28:09,  3.53it/s]
 95%|█████████▍| 352826/371472 [5:55:05<1:33:22,  3.33it/s]
 95%|█████████▍| 352827/371472 [5:55:05<1:29:42,  3.46it/s]
 95%|█████████▍| 352828/371472 [5:55:05<1:25:51,  3.62it/s]
 95%|█████████▍| 352829/371472 [5:55:06<1:28:49,  3.50it/s]
 95%|█████████▍| 352830/371472 [5:55:06<1:32:05,  3.37it/s]
 95%|█████████▍| 352831/371472 [5:55:06<1:31:08,  3.41it/s]
 95%|█████████▍| 352832/371472 [5:55:07<1:29:17,  3.48it/s]
 95%|█████████▍| 352833/371472 [5:55:07<1:25:00,  3.65it/s]
 95%|█████████▍| 352834/371472 [5:55:07<1:25:36,  3.63it/s]
 95%|█████████▍| 352835/371472 [5:55:07<1:22:08,  3.78it/s]
 95%|█████████▍| 352836/371472 [5:55:08<1:21:23,  3.82it/s]
 95%|█████████▍| 352837/371472 [5:55:08<1:22:50,  3.75it/s]
 95%|█████████▍| 352838/371472 [5:55:08<1:29:47,  3.46it/s]
 95%|█████████▍| 352839/371472 [5:55:08<1:28:23,  3.51it/s]
 95%|█████████▍| 352840/371472 [5:55:09<1:26:00,  3.61it/s]
                                                           
{'loss': 2.6565, 'learning_rate': 1.4516580835613784e-07, 'epoch': 15.2}
+
 95%|█████████▍| 352840/371472 [5:55:09<1:26:00,  3.61it/s]
 95%|█████████▍| 352841/371472 [5:55:09<1:31:37,  3.39it/s]
 95%|█████████▍| 352842/371472 [5:55:09<1:28:21,  3.51it/s]
 95%|█████████▍| 352843/371472 [5:55:10<1:27:06,  3.56it/s]
 95%|█████████▍| 352844/371472 [5:55:10<1:30:30,  3.43it/s]
 95%|█████████▍| 352845/371472 [5:55:10<1:41:10,  3.07it/s]
 95%|█████████▍| 352846/371472 [5:55:11<1:34:42,  3.28it/s]
 95%|█████████▍| 352847/371472 [5:55:11<1:32:32,  3.35it/s]
 95%|█████████▍| 352848/371472 [5:55:11<1:30:47,  3.42it/s]
 95%|█████████▍| 352849/371472 [5:55:11<1:30:54,  3.41it/s]
 95%|█████████▍| 352850/371472 [5:55:12<1:28:15,  3.52it/s]
 95%|█████████▍| 352851/371472 [5:55:12<1:28:20,  3.51it/s]
 95%|█████████▍| 352852/371472 [5:55:12<1:25:54,  3.61it/s]
 95%|█████████▍| 352853/371472 [5:55:13<1:28:38,  3.50it/s]
 95%|█████████▍| 352854/371472 [5:55:13<1:27:25,  3.55it/s]
 95%|█████████▍| 352855/371472 [5:55:13<1:27:07,  3.56it/s]
 95%|█████████▍| 352856/371472 [5:55:13<1:25:18,  3.64it/s]
 95%|█████████▍| 352857/371472 [5:55:14<1:29:26,  3.47it/s]
 95%|█████████▍| 352858/371472 [5:55:14<1:26:15,  3.60it/s]
 95%|█████████▍| 352859/371472 [5:55:14<1:28:57,  3.49it/s]
 95%|█████████▍| 352860/371472 [5:55:15<1:37:15,  3.19it/s]
                                                           
{'loss': 2.5179, 'learning_rate': 1.451173263806589e-07, 'epoch': 15.2}
+
 95%|█████████▍| 352860/371472 [5:55:15<1:37:15,  3.19it/s]
 95%|█████████▍| 352861/371472 [5:55:15<1:32:59,  3.34it/s]
 95%|█████████▍| 352862/371472 [5:55:15<1:31:01,  3.41it/s]
 95%|█████████▍| 352863/371472 [5:55:15<1:30:22,  3.43it/s]
 95%|█████████▍| 352864/371472 [5:55:16<1:29:08,  3.48it/s]
 95%|█████████▍| 352865/371472 [5:55:16<1:31:02,  3.41it/s]
 95%|█████████▍| 352866/371472 [5:55:16<1:32:13,  3.36it/s]
 95%|█████████▍| 352867/371472 [5:55:17<1:29:50,  3.45it/s]
 95%|█████████▍| 352868/371472 [5:55:17<1:35:47,  3.24it/s]
 95%|█████████▍| 352869/371472 [5:55:17<1:36:02,  3.23it/s]
 95%|█████████▍| 352870/371472 [5:55:18<1:35:20,  3.25it/s]
 95%|█████████▍| 352871/371472 [5:55:18<1:28:58,  3.48it/s]
 95%|█████████▍| 352872/371472 [5:55:18<1:27:02,  3.56it/s]
 95%|█████████▍| 352873/371472 [5:55:18<1:27:55,  3.53it/s]
 95%|█████████▍| 352874/371472 [5:55:19<1:25:00,  3.65it/s]
 95%|█████████▍| 352875/371472 [5:55:19<1:27:11,  3.56it/s]
 95%|█████████▍| 352876/371472 [5:55:19<1:28:33,  3.50it/s]
 95%|█████████▍| 352877/371472 [5:55:20<1:28:45,  3.49it/s]
 95%|█████████▍| 352878/371472 [5:55:20<1:28:48,  3.49it/s]
 95%|█████████▍| 352879/371472 [5:55:20<1:36:04,  3.23it/s]
 95%|█████████▍| 352880/371472 [5:55:20<1:36:15,  3.22it/s]
                                                           
{'loss': 2.4706, 'learning_rate': 1.4506884440518007e-07, 'epoch': 15.2}
+
 95%|█████████▍| 352880/371472 [5:55:20<1:36:15,  3.22it/s]
 95%|█████████▍| 352881/371472 [5:55:21<1:31:10,  3.40it/s]
 95%|█████████▍| 352882/371472 [5:55:21<1:34:36,  3.27it/s]
 95%|█████████▍| 352883/371472 [5:55:21<1:35:22,  3.25it/s]
 95%|█████████▍| 352884/371472 [5:55:22<1:38:20,  3.15it/s]
 95%|█████████▍| 352885/371472 [5:55:22<1:33:16,  3.32it/s]
 95%|█████████▍| 352886/371472 [5:55:22<1:31:51,  3.37it/s]
 95%|█████████▍| 352887/371472 [5:55:23<1:31:19,  3.39it/s]
 95%|█████████▍| 352888/371472 [5:55:23<1:28:33,  3.50it/s]
 95%|█████████▍| 352889/371472 [5:55:23<1:26:26,  3.58it/s]
 95%|█████████▍| 352890/371472 [5:55:23<1:29:44,  3.45it/s]
 95%|█████████▍| 352891/371472 [5:55:24<1:32:23,  3.35it/s]
 95%|█████████▍| 352892/371472 [5:55:24<1:37:16,  3.18it/s]
 95%|█████████▍| 352893/371472 [5:55:24<1:36:10,  3.22it/s]
 95%|█████████▍| 352894/371472 [5:55:25<1:33:11,  3.32it/s]
 95%|█████████▍| 352895/371472 [5:55:25<1:30:39,  3.42it/s]
 95%|█████████▍| 352896/371472 [5:55:25<1:27:20,  3.54it/s]
 95%|█████████▍| 352897/371472 [5:55:25<1:25:45,  3.61it/s]
 95%|█████████▍| 352898/371472 [5:55:26<1:27:47,  3.53it/s]
 95%|█████████▌| 352899/371472 [5:55:26<1:27:31,  3.54it/s]
 95%|█████████▌| 352900/371472 [5:55:26<1:29:45,  3.45it/s]
                                                           
{'loss': 2.6417, 'learning_rate': 1.450203624297011e-07, 'epoch': 15.2}
+
 95%|█████████▌| 352900/371472 [5:55:26<1:29:45,  3.45it/s]
 95%|█████████▌| 352901/371472 [5:55:27<1:36:23,  3.21it/s]
 95%|█████████▌| 352902/371472 [5:55:27<1:32:24,  3.35it/s]
 95%|█████████▌| 352903/371472 [5:55:27<1:30:49,  3.41it/s]
 95%|█████████▌| 352904/371472 [5:55:27<1:27:46,  3.53it/s]
 95%|█████████▌| 352905/371472 [5:55:28<1:27:44,  3.53it/s]
 95%|█████████▌| 352906/371472 [5:55:28<1:31:30,  3.38it/s]
 95%|█████████▌| 352907/371472 [5:55:28<1:26:17,  3.59it/s]
 95%|█████████▌| 352908/371472 [5:55:29<1:29:52,  3.44it/s]
 95%|█████████▌| 352909/371472 [5:55:29<1:26:19,  3.58it/s]
 95%|█████████▌| 352910/371472 [5:55:29<1:23:54,  3.69it/s]
 95%|█████████▌| 352911/371472 [5:55:29<1:25:34,  3.62it/s]
 95%|█████████▌| 352912/371472 [5:55:30<1:27:19,  3.54it/s]
 95%|█████████▌| 352913/371472 [5:55:30<1:24:46,  3.65it/s]
 95%|█████████▌| 352914/371472 [5:55:30<1:26:39,  3.57it/s]
 95%|█████████▌| 352915/371472 [5:55:31<1:26:48,  3.56it/s]
 95%|█████████▌| 352916/371472 [5:55:31<1:25:11,  3.63it/s]
 95%|█████████▌| 352917/371472 [5:55:31<1:26:58,  3.56it/s]
 95%|█████████▌| 352918/371472 [5:55:31<1:24:49,  3.65it/s]
 95%|█████████▌| 352919/371472 [5:55:32<1:24:23,  3.66it/s]
 95%|█████████▌| 352920/371472 [5:55:32<1:23:17,  3.71it/s]
                                                           
{'loss': 2.6903, 'learning_rate': 1.4497188045422226e-07, 'epoch': 15.2}
+
 95%|█████████▌| 352920/371472 [5:55:32<1:23:17,  3.71it/s]
 95%|█████████▌| 352921/371472 [5:55:32<1:24:38,  3.65it/s]
 95%|█████████▌| 352922/371472 [5:55:32<1:24:37,  3.65it/s]
 95%|█████████▌| 352923/371472 [5:55:33<1:24:01,  3.68it/s]
 95%|█████████▌| 352924/371472 [5:55:33<1:24:02,  3.68it/s]
 95%|█████████▌| 352925/371472 [5:55:33<1:21:24,  3.80it/s]
 95%|█████████▌| 352926/371472 [5:55:34<1:26:11,  3.59it/s]
 95%|█████████▌| 352927/371472 [5:55:34<1:27:56,  3.51it/s]
 95%|█████████▌| 352928/371472 [5:55:34<1:26:59,  3.55it/s]
 95%|█████████▌| 352929/371472 [5:55:34<1:23:59,  3.68it/s]
 95%|█████████▌| 352930/371472 [5:55:35<1:22:06,  3.76it/s]
 95%|█████████▌| 352931/371472 [5:55:35<1:21:49,  3.78it/s]
 95%|█████████▌| 352932/371472 [5:55:35<1:22:40,  3.74it/s]
 95%|█████████▌| 352933/371472 [5:55:35<1:19:45,  3.87it/s]
 95%|█████████▌| 352934/371472 [5:55:36<1:20:00,  3.86it/s]
 95%|█████████▌| 352935/371472 [5:55:36<1:22:27,  3.75it/s]
 95%|█████████▌| 352936/371472 [5:55:36<1:27:24,  3.53it/s]
 95%|█████████▌| 352937/371472 [5:55:37<1:26:30,  3.57it/s]
 95%|█████████▌| 352938/371472 [5:55:37<1:27:00,  3.55it/s]
 95%|█████████▌| 352939/371472 [5:55:37<1:24:11,  3.67it/s]
 95%|█████████▌| 352940/371472 [5:55:37<1:26:05,  3.59it/s]
                                                           
{'loss': 2.59, 'learning_rate': 1.4492339847874333e-07, 'epoch': 15.2}
+
 95%|█████████▌| 352940/371472 [5:55:37<1:26:05,  3.59it/s]
 95%|█████████▌| 352941/371472 [5:55:38<1:26:10,  3.58it/s]
 95%|█████████▌| 352942/371472 [5:55:38<1:25:56,  3.59it/s]
 95%|█████████▌| 352943/371472 [5:55:38<1:22:52,  3.73it/s]
 95%|█████████▌| 352944/371472 [5:55:38<1:23:51,  3.68it/s]
 95%|█████████▌| 352945/371472 [5:55:39<1:23:28,  3.70it/s]
 95%|█████████▌| 352946/371472 [5:55:39<1:25:53,  3.59it/s]
 95%|█████████▌| 352947/371472 [5:55:39<1:26:02,  3.59it/s]
 95%|█████████▌| 352948/371472 [5:55:40<1:30:02,  3.43it/s]
 95%|█████████▌| 352949/371472 [5:55:40<1:28:38,  3.48it/s]
 95%|█████████▌| 352950/371472 [5:55:40<1:26:18,  3.58it/s]
 95%|█████████▌| 352951/371472 [5:55:40<1:27:23,  3.53it/s]
 95%|█████████▌| 352952/371472 [5:55:41<1:34:45,  3.26it/s]
 95%|█████████▌| 352953/371472 [5:55:41<1:35:53,  3.22it/s]
 95%|█████████▌| 352954/371472 [5:55:41<1:31:48,  3.36it/s]
 95%|█████████▌| 352955/371472 [5:55:42<1:29:30,  3.45it/s]
 95%|█████████▌| 352956/371472 [5:55:42<1:28:08,  3.50it/s]
 95%|█████████▌| 352957/371472 [5:55:42<1:26:32,  3.57it/s]
 95%|█████████▌| 352958/371472 [5:55:43<1:25:34,  3.61it/s]
 95%|█████████▌| 352959/371472 [5:55:43<1:28:48,  3.47it/s]
 95%|█████████▌| 352960/371472 [5:55:43<1:24:59,  3.63it/s]
                                                           
{'loss': 2.6392, 'learning_rate': 1.4487491650326446e-07, 'epoch': 15.2}
+
 95%|█████████▌| 352960/371472 [5:55:43<1:24:59,  3.63it/s]
 95%|█████████▌| 352961/371472 [5:55:43<1:27:47,  3.51it/s]
 95%|█████████▌| 352962/371472 [5:55:44<1:27:39,  3.52it/s]
 95%|█████████▌| 352963/371472 [5:55:44<1:34:49,  3.25it/s]
 95%|█████████▌| 352964/371472 [5:55:44<1:32:15,  3.34it/s]
 95%|█████████▌| 352965/371472 [5:55:45<1:28:08,  3.50it/s]
 95%|█████████▌| 352966/371472 [5:55:45<1:31:06,  3.39it/s]
 95%|█████████▌| 352967/371472 [5:55:45<1:30:40,  3.40it/s]
 95%|█████████▌| 352968/371472 [5:55:45<1:30:00,  3.43it/s]
 95%|█████████▌| 352969/371472 [5:55:46<1:28:28,  3.49it/s]
 95%|█████████▌| 352970/371472 [5:55:46<1:31:47,  3.36it/s]
 95%|█████████▌| 352971/371472 [5:55:46<1:35:43,  3.22it/s]
 95%|█████████▌| 352972/371472 [5:55:47<1:31:56,  3.35it/s]
 95%|█████████▌| 352973/371472 [5:55:47<1:31:18,  3.38it/s]
 95%|█████████▌| 352974/371472 [5:55:47<1:30:13,  3.42it/s]
 95%|█████████▌| 352975/371472 [5:55:47<1:26:51,  3.55it/s]
 95%|█████████▌| 352976/371472 [5:55:48<1:24:19,  3.66it/s]
 95%|█████████▌| 352977/371472 [5:55:48<1:28:09,  3.50it/s]
 95%|█████████▌| 352978/371472 [5:55:48<1:31:45,  3.36it/s]
 95%|█████████▌| 352979/371472 [5:55:49<1:30:09,  3.42it/s]
 95%|█████████▌| 352980/371472 [5:55:49<1:28:13,  3.49it/s]
                                                           
{'loss': 2.5677, 'learning_rate': 1.4482643452778553e-07, 'epoch': 15.2}
+
 95%|█████████▌| 352980/371472 [5:55:49<1:28:13,  3.49it/s]
 95%|█████████▌| 352981/371472 [5:55:49<1:26:16,  3.57it/s]
 95%|█████████▌| 352982/371472 [5:55:49<1:22:22,  3.74it/s]
 95%|█████████▌| 352983/371472 [5:55:50<1:25:49,  3.59it/s]
 95%|█████████▌| 352984/371472 [5:55:50<1:28:34,  3.48it/s]
 95%|█████████▌| 352985/371472 [5:55:50<1:24:58,  3.63it/s]
 95%|█████████▌| 352986/371472 [5:55:51<1:24:25,  3.65it/s]
 95%|█████████▌| 352987/371472 [5:55:51<1:25:54,  3.59it/s]
 95%|█████████▌| 352988/371472 [5:55:51<1:26:22,  3.57it/s]
 95%|█████████▌| 352989/371472 [5:55:51<1:25:03,  3.62it/s]
 95%|█████████▌| 352990/371472 [5:55:52<1:30:10,  3.42it/s]
 95%|█████████▌| 352991/371472 [5:55:52<1:30:29,  3.40it/s]
 95%|█████████▌| 352992/371472 [5:55:52<1:28:35,  3.48it/s]
 95%|█████████▌| 352993/371472 [5:55:53<1:26:26,  3.56it/s]
 95%|█████████▌| 352994/371472 [5:55:53<1:27:40,  3.51it/s]
 95%|█████████▌| 352995/371472 [5:55:53<1:33:59,  3.28it/s]
 95%|█████████▌| 352996/371472 [5:55:54<1:31:52,  3.35it/s]
 95%|█████████▌| 352997/371472 [5:55:54<1:45:34,  2.92it/s]
 95%|█████████▌| 352998/371472 [5:55:54<1:38:18,  3.13it/s]
 95%|█████████▌| 352999/371472 [5:55:55<1:45:00,  2.93it/s]
 95%|█████████▌| 353000/371472 [5:55:55<1:42:04,  3.02it/s]
                                                           
{'loss': 2.5056, 'learning_rate': 1.447779525523067e-07, 'epoch': 15.2}
+
 95%|█████████▌| 353000/371472 [5:55:55<1:42:04,  3.02it/s]
 95%|█████████▌| 353001/371472 [5:55:55<1:37:49,  3.15it/s]
 95%|█████████▌| 353002/371472 [5:55:55<1:33:11,  3.30it/s]
 95%|█████████▌| 353003/371472 [5:55:56<1:29:34,  3.44it/s]
 95%|█████████▌| 353004/371472 [5:55:56<1:26:08,  3.57it/s]
 95%|█████████▌| 353005/371472 [5:55:56<1:30:06,  3.42it/s]
 95%|█████████▌| 353006/371472 [5:55:57<1:35:02,  3.24it/s]
 95%|█████████▌| 353007/371472 [5:55:57<1:31:35,  3.36it/s]
 95%|█████████▌| 353008/371472 [5:55:57<1:28:59,  3.46it/s]
 95%|█████████▌| 353009/371472 [5:55:57<1:25:04,  3.62it/s]
 95%|█████████▌| 353010/371472 [5:55:58<1:30:08,  3.41it/s]
 95%|█████████▌| 353011/371472 [5:55:58<1:30:15,  3.41it/s]
 95%|█████████▌| 353012/371472 [5:55:58<1:29:26,  3.44it/s]
 95%|█████████▌| 353013/371472 [5:55:59<1:28:34,  3.47it/s]
 95%|█████████▌| 353014/371472 [5:55:59<1:29:45,  3.43it/s]
 95%|█████████▌| 353015/371472 [5:55:59<1:26:58,  3.54it/s]
 95%|█████████▌| 353016/371472 [5:55:59<1:24:36,  3.64it/s]
 95%|█████████▌| 353017/371472 [5:56:00<1:30:41,  3.39it/s]
 95%|█████████▌| 353018/371472 [5:56:00<1:31:03,  3.38it/s]
 95%|█████████▌| 353019/371472 [5:56:00<1:30:39,  3.39it/s]
 95%|█████████▌| 353020/371472 [5:56:01<1:25:34,  3.59it/s]
                                                           
{'loss': 2.7021, 'learning_rate': 1.4472947057682775e-07, 'epoch': 15.21}
+
 95%|█████████▌| 353020/371472 [5:56:01<1:25:34,  3.59it/s]
 95%|█████████▌| 353021/371472 [5:56:01<1:26:11,  3.57it/s]
 95%|█████████▌| 353022/371472 [5:56:01<1:25:35,  3.59it/s]
 95%|█████████▌| 353023/371472 [5:56:01<1:25:00,  3.62it/s]
 95%|█████████▌| 353024/371472 [5:56:02<1:24:12,  3.65it/s]
 95%|█████████▌| 353025/371472 [5:56:02<1:27:50,  3.50it/s]
 95%|█████████▌| 353026/371472 [5:56:02<1:24:01,  3.66it/s]
 95%|█████████▌| 353027/371472 [5:56:03<1:27:27,  3.51it/s]
 95%|█████████▌| 353028/371472 [5:56:03<1:24:53,  3.62it/s]
 95%|█████████▌| 353029/371472 [5:56:03<1:27:51,  3.50it/s]
 95%|█████████▌| 353030/371472 [5:56:03<1:29:23,  3.44it/s]
 95%|█████████▌| 353031/371472 [5:56:04<1:27:55,  3.50it/s]
 95%|█████████▌| 353032/371472 [5:56:04<1:28:10,  3.49it/s]
 95%|█████████▌| 353033/371472 [5:56:04<1:26:00,  3.57it/s]
 95%|█████████▌| 353034/371472 [5:56:05<1:23:48,  3.67it/s]
 95%|█████████▌| 353035/371472 [5:56:05<1:22:30,  3.72it/s]
 95%|█████████▌| 353036/371472 [5:56:05<1:21:18,  3.78it/s]
 95%|█████████▌| 353037/371472 [5:56:05<1:20:23,  3.82it/s]
 95%|█████████▌| 353038/371472 [5:56:06<1:23:03,  3.70it/s]
 95%|█████████▌| 353039/371472 [5:56:06<1:25:18,  3.60it/s]
 95%|█████████▌| 353040/371472 [5:56:06<1:27:04,  3.53it/s]
                                                           
{'loss': 2.6101, 'learning_rate': 1.446809886013489e-07, 'epoch': 15.21}
+
 95%|█████████▌| 353040/371472 [5:56:06<1:27:04,  3.53it/s]
 95%|█████████▌| 353041/371472 [5:56:07<1:30:56,  3.38it/s]
 95%|█████████▌| 353042/371472 [5:56:07<1:29:35,  3.43it/s]
 95%|█████████▌| 353043/371472 [5:56:07<1:26:47,  3.54it/s]
 95%|█████████▌| 353044/371472 [5:56:07<1:26:39,  3.54it/s]
 95%|█████████▌| 353045/371472 [5:56:08<1:24:47,  3.62it/s]
 95%|█████████▌| 353046/371472 [5:56:08<1:23:48,  3.66it/s]
 95%|█████████▌| 353047/371472 [5:56:08<1:27:12,  3.52it/s]
 95%|█████████▌| 353048/371472 [5:56:08<1:27:11,  3.52it/s]
 95%|█████████▌| 353049/371472 [5:56:09<1:26:33,  3.55it/s]
 95%|█████████▌| 353050/371472 [5:56:09<1:30:38,  3.39it/s]
 95%|█████████▌| 353051/371472 [5:56:09<1:30:26,  3.39it/s]
 95%|█████████▌| 353052/371472 [5:56:10<1:28:34,  3.47it/s]
 95%|█████████▌| 353053/371472 [5:56:10<1:41:45,  3.02it/s]
 95%|█████████���| 353054/371472 [5:56:10<1:34:40,  3.24it/s]
 95%|█████████▌| 353055/371472 [5:56:11<1:33:50,  3.27it/s]
 95%|█████████▌| 353056/371472 [5:56:11<1:38:23,  3.12it/s]
 95%|█████████▌| 353057/371472 [5:56:11<1:33:50,  3.27it/s]
 95%|█████████▌| 353058/371472 [5:56:12<1:29:45,  3.42it/s]
 95%|█████████▌| 353059/371472 [5:56:12<1:27:32,  3.51it/s]
 95%|█████████▌| 353060/371472 [5:56:12<1:26:21,  3.55it/s]
                                                           
{'loss': 2.4156, 'learning_rate': 1.4463250662586997e-07, 'epoch': 15.21}
+
 95%|█████████▌| 353060/371472 [5:56:12<1:26:21,  3.55it/s]
 95%|█████████▌| 353061/371472 [5:56:12<1:22:41,  3.71it/s]
 95%|█████████▌| 353062/371472 [5:56:13<1:24:06,  3.65it/s]
 95%|█████████▌| 353063/371472 [5:56:13<1:27:24,  3.51it/s]
 95%|█████████▌| 353064/371472 [5:56:13<1:27:00,  3.53it/s]
 95%|█████████▌| 353065/371472 [5:56:13<1:25:14,  3.60it/s]
 95%|█████████▌| 353066/371472 [5:56:14<1:28:52,  3.45it/s]
 95%|█████████▌| 353067/371472 [5:56:14<1:27:07,  3.52it/s]
 95%|█████████▌| 353068/371472 [5:56:14<1:26:26,  3.55it/s]
 95%|█████████▌| 353069/371472 [5:56:15<1:31:53,  3.34it/s]
 95%|█████████▌| 353070/371472 [5:56:15<1:36:05,  3.19it/s]
 95%|█████████▌| 353071/371472 [5:56:15<1:30:05,  3.40it/s]
 95%|█████████▌| 353072/371472 [5:56:15<1:24:38,  3.62it/s]
 95%|█████████▌| 353073/371472 [5:56:16<1:25:46,  3.58it/s]
 95%|█████████▌| 353074/371472 [5:56:16<1:37:16,  3.15it/s]
 95%|█████████▌| 353075/371472 [5:56:16<1:29:35,  3.42it/s]
 95%|█████████▌| 353076/371472 [5:56:17<1:26:19,  3.55it/s]
 95%|█████████▌| 353077/371472 [5:56:17<1:24:35,  3.62it/s]
 95%|█████████▌| 353078/371472 [5:56:17<1:27:48,  3.49it/s]
 95%|█████████▌| 353079/371472 [5:56:17<1:24:50,  3.61it/s]
 95%|█████████▌| 353080/371472 [5:56:18<1:21:59,  3.74it/s]
                                                           
{'loss': 2.5549, 'learning_rate': 1.4458402465039105e-07, 'epoch': 15.21}
+
 95%|█████████▌| 353080/371472 [5:56:18<1:21:59,  3.74it/s]
 95%|█████████▌| 353081/371472 [5:56:18<1:22:11,  3.73it/s]
 95%|█████████▌| 353082/371472 [5:56:18<1:27:57,  3.48it/s]
 95%|█████████▌| 353083/371472 [5:56:19<1:26:52,  3.53it/s]
 95%|█████████▌| 353084/371472 [5:56:19<1:24:54,  3.61it/s]
 95%|█████████▌| 353085/371472 [5:56:19<1:27:19,  3.51it/s]
 95%|█████████▌| 353086/371472 [5:56:19<1:26:26,  3.55it/s]
 95%|█████████▌| 353087/371472 [5:56:20<1:28:10,  3.48it/s]
 95%|█████████▌| 353088/371472 [5:56:20<1:27:23,  3.51it/s]
 95%|█████████▌| 353089/371472 [5:56:20<1:24:40,  3.62it/s]
 95%|█████████▌| 353090/371472 [5:56:21<1:21:59,  3.74it/s]
 95%|█████████▌| 353091/371472 [5:56:21<1:20:37,  3.80it/s]
 95%|█████████▌| 353092/371472 [5:56:21<1:19:21,  3.86it/s]
 95%|█████████▌| 353093/371472 [5:56:21<1:26:47,  3.53it/s]
 95%|█████████▌| 353094/371472 [5:56:22<1:28:18,  3.47it/s]
 95%|█████████▌| 353095/371472 [5:56:22<1:29:28,  3.42it/s]
 95%|█████████▌| 353096/371472 [5:56:22<1:30:08,  3.40it/s]
 95%|█████████▌| 353097/371472 [5:56:23<1:28:03,  3.48it/s]
 95%|█████████▌| 353098/371472 [5:56:23<1:25:34,  3.58it/s]
 95%|█████████▌| 353099/371472 [5:56:23<1:25:02,  3.60it/s]
 95%|█████████▌| 353100/371472 [5:56:23<1:25:47,  3.57it/s]
                                                           
{'loss': 2.574, 'learning_rate': 1.445355426749122e-07, 'epoch': 15.21}
+
 95%|█████████▌| 353100/371472 [5:56:23<1:25:47,  3.57it/s]
 95%|█████████▌| 353101/371472 [5:56:24<1:26:21,  3.55it/s]
 95%|█████████▌| 353102/371472 [5:56:24<1:33:47,  3.26it/s]
 95%|█████████▌| 353103/371472 [5:56:24<1:30:59,  3.36it/s]
 95%|█████████▌| 353104/371472 [5:56:25<1:28:19,  3.47it/s]
 95%|█████████▌| 353105/371472 [5:56:25<1:32:40,  3.30it/s]
 95%|█████████▌| 353106/371472 [5:56:25<1:29:52,  3.41it/s]
 95%|█████████▌| 353107/371472 [5:56:25<1:27:58,  3.48it/s]
 95%|█████████▌| 353108/371472 [5:56:26<1:26:37,  3.53it/s]
 95%|█████████▌| 353109/371472 [5:56:26<1:27:56,  3.48it/s]
 95%|███████���█▌| 353110/371472 [5:56:26<1:29:53,  3.40it/s]
 95%|█████████▌| 353111/371472 [5:56:27<1:29:32,  3.42it/s]
 95%|█████████▌| 353112/371472 [5:56:27<1:31:49,  3.33it/s]
 95%|█████████▌| 353113/371472 [5:56:27<1:33:06,  3.29it/s]
 95%|█████████▌| 353114/371472 [5:56:28<1:31:29,  3.34it/s]
 95%|█████████▌| 353115/371472 [5:56:28<1:34:04,  3.25it/s]
 95%|█████████▌| 353116/371472 [5:56:28<1:31:50,  3.33it/s]
 95%|█████████▌| 353117/371472 [5:56:28<1:27:29,  3.50it/s]
 95%|█████████▌| 353118/371472 [5:56:29<1:26:51,  3.52it/s]
 95%|█████████▌| 353119/371472 [5:56:29<1:23:36,  3.66it/s]
 95%|█████████▌| 353120/371472 [5:56:29<1:21:57,  3.73it/s]
                                                           
{'loss': 2.5385, 'learning_rate': 1.4448706069943324e-07, 'epoch': 15.21}
+
 95%|█████████▌| 353120/371472 [5:56:29<1:21:57,  3.73it/s]
 95%|█████████▌| 353121/371472 [5:56:29<1:21:05,  3.77it/s]
 95%|█████████▌| 353122/371472 [5:56:30<1:24:09,  3.63it/s]
 95%|█████████▌| 353123/371472 [5:56:30<1:27:17,  3.50it/s]
 95%|█████████▌| 353124/371472 [5:56:30<1:24:05,  3.64it/s]
 95%|█████████▌| 353125/371472 [5:56:31<1:21:57,  3.73it/s]
 95%|█████████▌| 353126/371472 [5:56:31<1:22:24,  3.71it/s]
 95%|█████████▌| 353127/371472 [5:56:31<1:24:23,  3.62it/s]
 95%|█████████▌| 353128/371472 [5:56:31<1:26:28,  3.54it/s]
 95%|█████████▌| 353129/371472 [5:56:32<1:28:27,  3.46it/s]
 95%|█████████▌| 353130/371472 [5:56:32<1:27:48,  3.48it/s]
 95%|█████████▌| 353131/371472 [5:56:32<1:25:07,  3.59it/s]
 95%|█████████▌| 353132/371472 [5:56:33<1:22:58,  3.68it/s]
 95%|█████████▌| 353133/371472 [5:56:33<1:21:15,  3.76it/s]
 95%|█████████▌| 353134/371472 [5:56:33<1:21:36,  3.75it/s]
 95%|█████████▌| 353135/371472 [5:56:33<1:24:24,  3.62it/s]
 95%|█████████▌| 353136/371472 [5:56:34<1:24:18,  3.62it/s]
 95%|█████████▌| 353137/371472 [5:56:34<1:22:22,  3.71it/s]
 95%|█████████▌| 353138/371472 [5:56:34<1:21:23,  3.75it/s]
 95%|█████████▌| 353139/371472 [5:56:34<1:28:19,  3.46it/s]
 95%|█████████▌| 353140/371472 [5:56:35<1:25:54,  3.56it/s]
                                                           
{'loss': 2.5167, 'learning_rate': 1.4443857872395442e-07, 'epoch': 15.21}
+
 95%|█████████▌| 353140/371472 [5:56:35<1:25:54,  3.56it/s]
 95%|█████████▌| 353141/371472 [5:56:35<1:28:02,  3.47it/s]
 95%|█████████▌| 353142/371472 [5:56:35<1:23:55,  3.64it/s]
 95%|█████████▌| 353143/371472 [5:56:36<1:23:12,  3.67it/s]
 95%|█████████▌| 353144/371472 [5:56:36<1:23:18,  3.67it/s]
 95%|█████████▌| 353145/371472 [5:56:36<1:26:51,  3.52it/s]
 95%|█████████▌| 353146/371472 [5:56:36<1:33:48,  3.26it/s]
 95%|█████████▌| 353147/371472 [5:56:37<1:31:15,  3.35it/s]
 95%|█████████▌| 353148/371472 [5:56:37<1:30:54,  3.36it/s]
 95%|█████████▌| 353149/371472 [5:56:37<1:29:44,  3.40it/s]
 95%|█████████▌| 353150/371472 [5:56:38<1:29:20,  3.42it/s]
 95%|█████████▌| 353151/371472 [5:56:38<1:36:36,  3.16it/s]
 95%|█████████▌| 353152/371472 [5:56:38<1:35:10,  3.21it/s]
 95%|█████████▌| 353153/371472 [5:56:39<1:33:39,  3.26it/s]
 95%|█████████▌| 353154/371472 [5:56:39<1:30:05,  3.39it/s]
 95%|█████████▌| 353155/371472 [5:56:39<1:34:13,  3.24it/s]
 95%|█████████▌| 353156/371472 [5:56:39<1:33:18,  3.27it/s]
 95%|█████████▌| 353157/371472 [5:56:40<1:30:04,  3.39it/s]
 95%|█████████▌| 353158/371472 [5:56:40<1:29:23,  3.41it/s]
 95%|█████████▌| 353159/371472 [5:56:40<1:31:50,  3.32it/s]
 95%|█████████▌| 353160/371472 [5:56:41<1:41:35,  3.00it/s]
                                                           
{'loss': 2.5262, 'learning_rate': 1.4439009674847546e-07, 'epoch': 15.21}
+
 95%|█████████▌| 353160/371472 [5:56:41<1:41:35,  3.00it/s]
 95%|█████████▌| 353161/371472 [5:56:41<1:37:47,  3.12it/s]
 95%|█████████▌| 353162/371472 [5:56:41<1:36:57,  3.15it/s]
 95%|█████████▌| 353163/371472 [5:56:42<1:32:02,  3.32it/s]
 95%|█████████▌| 353164/371472 [5:56:42<1:32:42,  3.29it/s]
 95%|█████████▌| 353165/371472 [5:56:42<1:32:29,  3.30it/s]
 95%|████��████▌| 353166/371472 [5:56:43<1:27:45,  3.48it/s]
 95%|█████████▌| 353167/371472 [5:56:43<1:32:49,  3.29it/s]
 95%|█████████▌| 353168/371472 [5:56:43<1:37:54,  3.12it/s]
 95%|█████████▌| 353169/371472 [5:56:43<1:32:33,  3.30it/s]
 95%|█████████▌| 353170/371472 [5:56:44<1:28:22,  3.45it/s]
 95%|█████████▌| 353171/371472 [5:56:44<1:25:21,  3.57it/s]
 95%|█████████▌| 353172/371472 [5:56:44<1:23:19,  3.66it/s]
 95%|█████████▌| 353173/371472 [5:56:44<1:20:51,  3.77it/s]
 95%|█████████▌| 353174/371472 [5:56:45<1:31:31,  3.33it/s]
 95%|█████████▌| 353175/371472 [5:56:45<1:27:19,  3.49it/s]
 95%|█████████▌| 353176/371472 [5:56:45<1:25:09,  3.58it/s]
 95%|█████████▌| 353177/371472 [5:56:46<1:25:11,  3.58it/s]
 95%|█████████▌| 353178/371472 [5:56:46<1:26:00,  3.54it/s]
 95%|█████████▌| 353179/371472 [5:56:46<1:29:51,  3.39it/s]
 95%|█████████▌| 353180/371472 [5:56:47<1:35:49,  3.18it/s]
                                                           
{'loss': 2.5118, 'learning_rate': 1.4434161477299662e-07, 'epoch': 15.21}
+
 95%|█████████▌| 353180/371472 [5:56:47<1:35:49,  3.18it/s]
 95%|█████████▌| 353181/371472 [5:56:47<1:30:05,  3.38it/s]
 95%|█████████▌| 353182/371472 [5:56:47<1:29:22,  3.41it/s]
 95%|█████████▌| 353183/371472 [5:56:47<1:29:24,  3.41it/s]
 95%|█████████▌| 353184/371472 [5:56:48<1:27:03,  3.50it/s]
 95%|█████████▌| 353185/371472 [5:56:48<1:24:55,  3.59it/s]
 95%|█████████▌| 353186/371472 [5:56:48<1:26:50,  3.51it/s]
 95%|█████████▌| 353187/371472 [5:56:49<1:31:10,  3.34it/s]
 95%|█████████▌| 353188/371472 [5:56:49<1:27:11,  3.50it/s]
 95%|█████████▌| 353189/371472 [5:56:49<1:27:59,  3.46it/s]
 95%|█████████▌| 353190/371472 [5:56:50<1:33:14,  3.27it/s]
 95%|█████████▌| 353191/371472 [5:56:50<1:28:18,  3.45it/s]
 95%|█████████▌| 353192/371472 [5:56:50<1:24:54,  3.59it/s]
 95%|█████████▌| 353193/371472 [5:56:50<1:22:21,  3.70it/s]
 95%|█████████▌| 353194/371472 [5:56:51<1:21:17,  3.75it/s]
 95%|█████████▌| 353195/371472 [5:56:51<1:23:20,  3.66it/s]
 95%|█████████▌| 353196/371472 [5:56:51<1:25:03,  3.58it/s]
 95%|█████████▌| 353197/371472 [5:56:51<1:22:21,  3.70it/s]
 95%|█████████▌| 353198/371472 [5:56:52<1:23:08,  3.66it/s]
 95%|█████████▌| 353199/371472 [5:56:52<1:25:03,  3.58it/s]
 95%|█████████▌| 353200/371472 [5:56:52<1:24:30,  3.60it/s]
                                                           
{'loss': 2.5959, 'learning_rate': 1.442931327975177e-07, 'epoch': 15.21}
+
 95%|█████████▌| 353200/371472 [5:56:52<1:24:30,  3.60it/s]
 95%|█████████▌| 353201/371472 [5:56:52<1:23:14,  3.66it/s]
 95%|█████████▌| 353202/371472 [5:56:53<1:26:04,  3.54it/s]
 95%|█████████▌| 353203/371472 [5:56:53<1:24:47,  3.59it/s]
 95%|█████████▌| 353204/371472 [5:56:53<1:24:23,  3.61it/s]
 95%|█████████▌| 353205/371472 [5:56:54<1:27:36,  3.48it/s]
 95%|█████████▌| 353206/371472 [5:56:54<1:25:34,  3.56it/s]
 95%|█████████▌| 353207/371472 [5:56:54<1:25:15,  3.57it/s]
 95%|█████████▌| 353208/371472 [5:56:54<1:25:21,  3.57it/s]
 95%|█████████▌| 353209/371472 [5:56:55<1:24:15,  3.61it/s]
 95%|█████████▌| 353210/371472 [5:56:55<1:28:16,  3.45it/s]
 95%|█████████▌| 353211/371472 [5:56:55<1:30:18,  3.37it/s]
 95%|█████████▌| 353212/371472 [5:56:56<1:37:28,  3.12it/s]
 95%|█████████▌| 353213/371472 [5:56:56<1:35:58,  3.17it/s]
 95%|█████████▌| 353214/371472 [5:56:56<1:36:46,  3.14it/s]
 95%|█████████▌| 353215/371472 [5:56:57<1:34:54,  3.21it/s]
 95%|█████████▌| 353216/371472 [5:56:57<1:33:04,  3.27it/s]
 95%|█████████▌| 353217/371472 [5:56:57<1:32:36,  3.29it/s]
 95%|█████████▌| 353218/371472 [5:56:58<1:30:46,  3.35it/s]
 95%|█████████▌| 353219/371472 [5:56:58<1:30:09,  3.37it/s]
 95%|█████████▌| 353220/371472 [5:56:58<1:28:27,  3.44it/s]
                                                           
{'loss': 2.577, 'learning_rate': 1.4424465082203884e-07, 'epoch': 15.21}
+
 95%|█████████▌| 353220/371472 [5:56:58<1:28:27,  3.44it/s]
 95%|█████████▌| 353221/371472 [5:56:58<1:26:20,  3.52it/s]
 95%|██��██████▌| 353222/371472 [5:56:59<1:29:21,  3.40it/s]
 95%|█████████▌| 353223/371472 [5:56:59<1:27:42,  3.47it/s]
 95%|█████████▌| 353224/371472 [5:56:59<1:32:39,  3.28it/s]
 95%|█████████▌| 353225/371472 [5:57:00<1:29:23,  3.40it/s]
 95%|█████████▌| 353226/371472 [5:57:00<1:27:11,  3.49it/s]
 95%|█████████▌| 353227/371472 [5:57:00<1:25:47,  3.54it/s]
 95%|█████████▌| 353228/371472 [5:57:00<1:23:15,  3.65it/s]
 95%|█████████▌| 353229/371472 [5:57:01<1:21:00,  3.75it/s]
 95%|█████████▌| 353230/371472 [5:57:01<1:22:02,  3.71it/s]
 95%|█████████▌| 353231/371472 [5:57:01<1:25:22,  3.56it/s]
 95%|█████████▌| 353232/371472 [5:57:01<1:21:18,  3.74it/s]
 95%|█████████▌| 353233/371472 [5:57:02<1:19:26,  3.83it/s]
 95%|█████████▌| 353234/371472 [5:57:02<1:21:49,  3.71it/s]
 95%|█████████▌| 353235/371472 [5:57:02<1:24:50,  3.58it/s]
 95%|█████████▌| 353236/371472 [5:57:03<1:25:25,  3.56it/s]
 95%|█████████▌| 353237/371472 [5:57:03<1:25:20,  3.56it/s]
 95%|█████████▌| 353238/371472 [5:57:03<1:22:17,  3.69it/s]
 95%|█████████▌| 353239/371472 [5:57:03<1:24:55,  3.58it/s]
 95%|█████████▌| 353240/371472 [5:57:04<1:20:41,  3.77it/s]
                                                           
{'loss': 2.7245, 'learning_rate': 1.4419616884655988e-07, 'epoch': 15.21}
+
 95%|█████████▌| 353240/371472 [5:57:04<1:20:41,  3.77it/s]
 95%|█████████▌| 353241/371472 [5:57:04<1:20:17,  3.78it/s]
 95%|█████████▌| 353242/371472 [5:57:04<1:32:08,  3.30it/s]
 95%|█████████▌| 353243/371472 [5:57:05<1:33:14,  3.26it/s]
 95%|█████████▌| 353244/371472 [5:57:05<1:28:35,  3.43it/s]
 95%|█████████▌| 353245/371472 [5:57:05<1:29:29,  3.39it/s]
 95%|█████████▌| 353246/371472 [5:57:05<1:26:42,  3.50it/s]
 95%|█████████▌| 353247/371472 [5:57:06<1:26:01,  3.53it/s]
 95%|█████████▌| 353248/371472 [5:57:06<1:29:48,  3.38it/s]
 95%|█████████▌| 353249/371472 [5:57:06<1:27:59,  3.45it/s]
 95%|█████████▌| 353250/371472 [5:57:07<1:23:26,  3.64it/s]
 95%|█████████▌| 353251/371472 [5:57:07<1:23:24,  3.64it/s]
 95%|█████████▌| 353252/371472 [5:57:07<1:27:47,  3.46it/s]
 95%|█████████▌| 353253/371472 [5:57:07<1:30:23,  3.36it/s]
 95%|█████████▌| 353254/371472 [5:57:08<1:30:08,  3.37it/s]
 95%|█████████▌| 353255/371472 [5:57:08<1:24:53,  3.58it/s]
 95%|█████████▌| 353256/371472 [5:57:08<1:23:37,  3.63it/s]
 95%|█████████▌| 353257/371472 [5:57:09<1:22:13,  3.69it/s]
 95%|█████████▌| 353258/371472 [5:57:09<1:24:41,  3.58it/s]
 95%|█████████▌| 353259/371472 [5:57:09<1:24:16,  3.60it/s]
 95%|█████████▌| 353260/371472 [5:57:09<1:25:24,  3.55it/s]
                                                           
{'loss': 2.587, 'learning_rate': 1.4414768687108106e-07, 'epoch': 15.22}
+
 95%|█████████▌| 353260/371472 [5:57:09<1:25:24,  3.55it/s]
 95%|█████████▌| 353261/371472 [5:57:10<1:21:57,  3.70it/s]
 95%|█████████▌| 353262/371472 [5:57:10<1:24:48,  3.58it/s]
 95%|█████████▌| 353263/371472 [5:57:10<1:27:01,  3.49it/s]
 95%|█████████▌| 353264/371472 [5:57:11<1:29:10,  3.40it/s]
 95%|█████████▌| 353265/371472 [5:57:11<1:27:25,  3.47it/s]
 95%|█████████▌| 353266/371472 [5:57:11<1:28:23,  3.43it/s]
 95%|█████████▌| 353267/371472 [5:57:11<1:28:54,  3.41it/s]
 95%|█████████▌| 353268/371472 [5:57:12<1:26:49,  3.49it/s]
 95%|█████████▌| 353269/371472 [5:57:12<1:22:47,  3.66it/s]
 95%|█████████▌| 353270/371472 [5:57:12<1:24:17,  3.60it/s]
 95%|█████████▌| 353271/371472 [5:57:13<1:27:46,  3.46it/s]
 95%|█████████▌| 353272/371472 [5:57:13<1:29:15,  3.40it/s]
 95%|█████████▌| 353273/371472 [5:57:13<1:29:58,  3.37it/s]
 95%|█████████▌| 353274/371472 [5:57:13<1:27:16,  3.48it/s]
 95%|█████████▌| 353275/371472 [5:57:14<1:27:31,  3.47it/s]
 95%|█████████▌| 353276/371472 [5:57:14<1:25:03,  3.57it/s]
 95%|█████████▌| 353277/371472 [5:57:14<1:29:12,  3.40it/s]
 95%|█████████▌| 353278/371472 [5:57:15<1:26:04,  3.52it/s]
 95%|█████████▌| 353279/371472 [5:57:15<1:37:53,  3.10it/s]
 95%|█████████▌| 353280/371472 [5:57:15<1:33:44,  3.23it/s]
                                                           
{'loss': 2.6049, 'learning_rate': 1.440992048956021e-07, 'epoch': 15.22}
+
 95%|█████████▌| 353280/371472 [5:57:15<1:33:44,  3.23it/s]
 95%|█████████▌| 353281/371472 [5:57:16<1:29:28,  3.39it/s]
 95%|█████████▌| 353282/371472 [5:57:16<1:27:33,  3.46it/s]
 95%|█████████▌| 353283/371472 [5:57:16<1:27:23,  3.47it/s]
 95%|█████████▌| 353284/371472 [5:57:16<1:24:56,  3.57it/s]
 95%|█████████▌| 353285/371472 [5:57:17<1:25:28,  3.55it/s]
 95%|█████████▌| 353286/371472 [5:57:17<1:24:41,  3.58it/s]
 95%|█████████▌| 353287/371472 [5:57:17<1:23:18,  3.64it/s]
 95%|█████████▌| 353288/371472 [5:57:17<1:26:48,  3.49it/s]
 95%|█████████▌| 353289/371472 [5:57:18<1:25:49,  3.53it/s]
 95%|█████████▌| 353290/371472 [5:57:18<1:28:28,  3.43it/s]
 95%|█████████▌| 353291/371472 [5:57:18<1:27:20,  3.47it/s]
 95%|█████████▌| 353292/371472 [5:57:19<1:34:27,  3.21it/s]
 95%|█████████▌| 353293/371472 [5:57:19<1:29:47,  3.37it/s]
 95%|█████████▌| 353294/371472 [5:57:19<1:30:59,  3.33it/s]
 95%|█████████▌| 353295/371472 [5:57:20<1:27:41,  3.45it/s]
 95%|█████████▌| 353296/371472 [5:57:20<1:24:52,  3.57it/s]
 95%|█████████▌| 353297/371472 [5:57:20<1:23:11,  3.64it/s]
 95%|█████████▌| 353298/371472 [5:57:20<1:22:56,  3.65it/s]
 95%|█████████▌| 353299/371472 [5:57:21<1:22:48,  3.66it/s]
 95%|█████████▌| 353300/371472 [5:57:21<1:22:55,  3.65it/s]
                                                           
{'loss': 2.6307, 'learning_rate': 1.4405072292012326e-07, 'epoch': 15.22}
+
 95%|█████████▌| 353300/371472 [5:57:21<1:22:55,  3.65it/s]
 95%|█████████▌| 353301/371472 [5:57:21<1:24:48,  3.57it/s]
 95%|█████████▌| 353302/371472 [5:57:21<1:22:50,  3.66it/s]
 95%|█████████▌| 353303/371472 [5:57:22<1:24:08,  3.60it/s]
 95%|█████████▌| 353304/371472 [5:57:22<1:23:39,  3.62it/s]
 95%|█████████▌| 353305/371472 [5:57:22<1:22:20,  3.68it/s]
 95%|█████████▌| 353306/371472 [5:57:23<1:20:39,  3.75it/s]
 95%|█████████▌| 353307/371472 [5:57:23<1:22:38,  3.66it/s]
 95%|█████████▌| 353308/371472 [5:57:23<1:21:00,  3.74it/s]
 95%|█████████▌| 353309/371472 [5:57:23<1:21:47,  3.70it/s]
 95%|█████████▌| 353310/371472 [5:57:24<1:25:19,  3.55it/s]
 95%|█████████▌| 353311/371472 [5:57:24<1:21:20,  3.72it/s]
 95%|█████████▌| 353312/371472 [5:57:24<1:27:55,  3.44it/s]
 95%|█████████▌| 353313/371472 [5:57:25<1:31:01,  3.32it/s]
 95%|█████████▌| 353314/371472 [5:57:25<1:26:41,  3.49it/s]
 95%|█████████▌| 353315/371472 [5:57:25<1:26:00,  3.52it/s]
 95%|█████████▌| 353316/371472 [5:57:25<1:23:49,  3.61it/s]
 95%|█████████▌| 353317/371472 [5:57:26<1:22:42,  3.66it/s]
 95%|█████████▌| 353318/371472 [5:57:26<1:22:08,  3.68it/s]
 95%|█████████▌| 353319/371472 [5:57:26<1:22:27,  3.67it/s]
 95%|█████████▌| 353320/371472 [5:57:26<1:21:51,  3.70it/s]
                                                           
{'loss': 2.5835, 'learning_rate': 1.4400224094464433e-07, 'epoch': 15.22}
+
 95%|█████████▌| 353320/371472 [5:57:26<1:21:51,  3.70it/s]
 95%|█████████▌| 353321/371472 [5:57:27<1:20:43,  3.75it/s]
 95%|█████████▌| 353322/371472 [5:57:27<1:24:59,  3.56it/s]
 95%|█████████▌| 353323/371472 [5:57:27<1:23:40,  3.62it/s]
 95%|█████████▌| 353324/371472 [5:57:28<1:24:23,  3.58it/s]
 95%|█████████▌| 353325/371472 [5:57:28<1:21:14,  3.72it/s]
 95%|█████████▌| 353326/371472 [5:57:28<1:26:22,  3.50it/s]
 95%|█████████▌| 353327/371472 [5:57:28<1:25:22,  3.54it/s]
 95%|█████████▌| 353328/371472 [5:57:29<1:21:34,  3.71it/s]
 95%|█████████▌| 353329/371472 [5:57:29<1:20:37,  3.75it/s]
 95%|█████████▌| 353330/371472 [5:57:29<1:19:16,  3.81it/s]
 95%|█████████▌| 353331/371472 [5:57:29<1:21:24,  3.71it/s]
 95%|█████████▌| 353332/371472 [5:57:30<1:23:59,  3.60it/s]
 95%|█████████▌| 353333/371472 [5:57:30<1:21:34,  3.71it/s]
 95%|█████████▌| 353334/371472 [5:57:30<1:20:40,  3.75it/s]
 95%|█████████▌| 353335/371472 [5:57:30<1:19:07,  3.82it/s]
 95%|█████████▌| 353336/371472 [5:57:31<1:17:40,  3.89it/s]
 95%|█████████▌| 353337/371472 [5:57:31<1:19:35,  3.80it/s]
 95%|█████████▌| 353338/371472 [5:57:31<1:21:22,  3.71it/s]
 95%|█████████▌| 353339/371472 [5:57:32<1:22:20,  3.67it/s]
 95%|█████████▌| 353340/371472 [5:57:32<1:21:10,  3.72it/s]
                                                           
{'loss': 2.6179, 'learning_rate': 1.4395375896916548e-07, 'epoch': 15.22}
+
 95%|█████████▌| 353340/371472 [5:57:32<1:21:10,  3.72it/s]
 95%|█████████▌| 353341/371472 [5:57:32<1:23:49,  3.60it/s]
 95%|█████████▌| 353342/371472 [5:57:32<1:23:55,  3.60it/s]
 95%|█████████▌| 353343/371472 [5:57:33<1:23:12,  3.63it/s]
 95%|█████████▌| 353344/371472 [5:57:33<1:21:32,  3.71it/s]
 95%|█████████▌| 353345/371472 [5:57:33<1:30:45,  3.33it/s]
 95%|█████████▌| 353346/371472 [5:57:34<1:25:33,  3.53it/s]
 95%|█████████▌| 353347/371472 [5:57:34<1:23:01,  3.64it/s]
 95%|█████████▌| 353348/371472 [5:57:34<1:22:59,  3.64it/s]
 95%|█████████▌| 353349/371472 [5:57:34<1:23:25,  3.62it/s]
 95%|█████████▌| 353350/371472 [5:57:35<1:20:29,  3.75it/s]
 95%|█████████▌| 353351/371472 [5:57:35<1:20:00,  3.77it/s]
 95%|█████████▌| 353352/371472 [5:57:35<1:19:29,  3.80it/s]
 95%|█████████▌| 353353/371472 [5:57:35<1:23:21,  3.62it/s]
 95%|█████████▌| 353354/371472 [5:57:36<1:23:12,  3.63it/s]
 95%|█████████▌| 353355/371472 [5:57:36<1:23:39,  3.61it/s]
 95%|█████████▌| 353356/371472 [5:57:36<1:26:55,  3.47it/s]
 95%|█████████▌| 353357/371472 [5:57:37<1:24:39,  3.57it/s]
 95%|█████████▌| 353358/371472 [5:57:37<1:30:39,  3.33it/s]
 95%|█████████▌| 353359/371472 [5:57:37<1:28:55,  3.39it/s]
 95%|█████████▌| 353360/371472 [5:57:37<1:30:37,  3.33it/s]
                                                           
{'loss': 2.5218, 'learning_rate': 1.4390527699368655e-07, 'epoch': 15.22}
+
 95%|█████████▌| 353360/371472 [5:57:37<1:30:37,  3.33it/s]
 95%|█████████▌| 353361/371472 [5:57:38<1:26:35,  3.49it/s]
 95%|█████████▌| 353362/371472 [5:57:38<1:25:44,  3.52it/s]
 95%|█████████▌| 353363/371472 [5:57:38<1:24:25,  3.58it/s]
 95%|█████████▌| 353364/371472 [5:57:39<1:24:59,  3.55it/s]
 95%|█████████▌| 353365/371472 [5:57:39<1:23:30,  3.61it/s]
 95%|█████████▌| 353366/371472 [5:57:39<1:24:01,  3.59it/s]
 95%|█████████▌| 353367/371472 [5:57:39<1:20:53,  3.73it/s]
 95%|█████████▌| 353368/371472 [5:57:40<1:24:24,  3.57it/s]
 95%|█████████▌| 353369/371472 [5:57:40<1:21:47,  3.69it/s]
 95%|█████████▌| 353370/371472 [5:57:40<1:21:31,  3.70it/s]
 95%|█████████▌| 353371/371472 [5:57:40<1:20:46,  3.73it/s]
 95%|█████████▌| 353372/371472 [5:57:41<1:18:20,  3.85it/s]
 95%|█████████▌| 353373/371472 [5:57:41<1:20:34,  3.74it/s]
 95%|█████████▌| 353374/371472 [5:57:41<1:27:44,  3.44it/s]
 95%|█████████▌| 353375/371472 [5:57:42<1:34:37,  3.19it/s]
 95%|█████████▌| 353376/371472 [5:57:42<1:34:52,  3.18it/s]
 95%|█████████▌| 353377/371472 [5:57:42<1:31:09,  3.31it/s]
 95%|█████████▌| 353378/371472 [5:57:43<1:27:21,  3.45it/s]
 95%|█████████▌| 353379/371472 [5:57:43<1:28:17,  3.42it/s]
 95%|█████████▌| 353380/371472 [5:57:43<1:31:58,  3.28it/s]
                                                           
{'loss': 2.6177, 'learning_rate': 1.438567950182077e-07, 'epoch': 15.22}
+
 95%|█████████▌| 353380/371472 [5:57:43<1:31:58,  3.28it/s]
 95%|█████████▌| 353381/371472 [5:57:43<1:28:38,  3.40it/s]
 95%|█████████▌| 353382/371472 [5:57:44<1:26:41,  3.48it/s]
 95%|█████████▌| 353383/371472 [5:57:44<1:24:15,  3.58it/s]
 95%|█████████▌| 353384/371472 [5:57:44<1:25:39,  3.52it/s]
 95%|█████████▌| 353385/371472 [5:57:45<1:26:42,  3.48it/s]
 95%|█████████▌| 353386/371472 [5:57:45<1:23:34,  3.61it/s]
 95%|█████████▌| 353387/371472 [5:57:45<1:21:06,  3.72it/s]
 95%|█████████▌| 353388/371472 [5:57:45<1:24:53,  3.55it/s]
 95%|█████████▌| 353389/371472 [5:57:46<1:23:21,  3.62it/s]
 95%|█████████▌| 353390/371472 [5:57:46<1:23:33,  3.61it/s]
 95%|█████████▌| 353391/371472 [5:57:46<1:27:27,  3.45it/s]
 95%|█████████▌| 353392/371472 [5:57:46<1:23:13,  3.62it/s]
 95%|█████████▌| 353393/371472 [5:57:47<1:26:37,  3.48it/s]
 95%|█████████▌| 353394/371472 [5:57:47<1:23:53,  3.59it/s]
 95%|█████████▌| 353395/371472 [5:57:47<1:23:42,  3.60it/s]
 95%|█████████▌| 353396/371472 [5:57:48<1:19:31,  3.79it/s]
 95%|█████████▌| 353397/371472 [5:57:48<1:21:53,  3.68it/s]
 95%|█████████▌| 353398/371472 [5:57:48<1:23:47,  3.60it/s]
 95%|█████████▌| 353399/371472 [5:57:48<1:21:13,  3.71it/s]
 95%|█████████▌| 353400/371472 [5:57:49<1:24:06,  3.58it/s]
                                                           
{'loss': 2.697, 'learning_rate': 1.4380831304272877e-07, 'epoch': 15.22}
+
 95%|█████████▌| 353400/371472 [5:57:49<1:24:06,  3.58it/s]
 95%|█████████▌| 353401/371472 [5:57:49<1:26:11,  3.49it/s]
 95%|█████████▌| 353402/371472 [5:57:49<1:27:10,  3.45it/s]
 95%|█████████▌| 353403/371472 [5:57:50<1:28:45,  3.39it/s]
 95%|█████████▌| 353404/371472 [5:57:50<1:29:29,  3.36it/s]
 95%|█████████▌| 353405/371472 [5:57:50<1:24:59,  3.54it/s]
 95%|█████████▌| 353406/371472 [5:57:50<1:23:07,  3.62it/s]
 95%|█████████▌| 353407/371472 [5:57:51<1:20:47,  3.73it/s]
 95%|█████████▌| 353408/371472 [5:57:51<1:18:24,  3.84it/s]
 95%|█████████▌| 353409/371472 [5:57:51<1:18:42,  3.82it/s]
 95%|█████████▌| 353410/371472 [5:57:51<1:20:06,  3.76it/s]
 95%|█████████▌| 353411/371472 [5:57:52<1:20:23,  3.74it/s]
 95%|█████████▌| 353412/371472 [5:57:52<1:19:14,  3.80it/s]
 95%|█████████▌| 353413/371472 [5:57:52<1:18:35,  3.83it/s]
 95%|█████████▌| 353414/371472 [5:57:53<1:19:29,  3.79it/s]
 95%|█████████▌| 353415/371472 [5:57:53<1:22:15,  3.66it/s]
 95%|█████████▌| 353416/371472 [5:57:53<1:29:06,  3.38it/s]
 95%|█████████▌| 353417/371472 [5:57:53<1:28:50,  3.39it/s]
 95%|█████████▌| 353418/371472 [5:57:54<1:29:20,  3.37it/s]
 95%|█████████▌| 353419/371472 [5:57:54<1:26:58,  3.46it/s]
 95%|█████████▌| 353420/371472 [5:57:54<1:26:49,  3.46it/s]
                                                           
{'loss': 2.617, 'learning_rate': 1.4375983106724992e-07, 'epoch': 15.22}
+
 95%|█████████▌| 353420/371472 [5:57:54<1:26:49,  3.46it/s]
 95%|█████████▌| 353421/371472 [5:57:55<1:28:57,  3.38it/s]
 95%|█████████▌| 353422/371472 [5:57:55<1:30:09,  3.34it/s]
 95%|█████████▌| 353423/371472 [5:57:55<1:27:03,  3.46it/s]
 95%|█████████▌| 353424/371472 [5:57:56<1:29:08,  3.37it/s]
 95%|█████████▌| 353425/371472 [5:57:56<1:27:47,  3.43it/s]
 95%|█████████▌| 353426/371472 [5:57:56<1:30:51,  3.31it/s]
 95%|█████████▌| 353427/371472 [5:57:56<1:31:14,  3.30it/s]
 95%|█████████▌| 353428/371472 [5:57:57<1:29:34,  3.36it/s]
 95%|█████████▌| 353429/371472 [5:57:57<1:33:41,  3.21it/s]
 95%|█████████▌| 353430/371472 [5:57:57<1:28:12,  3.41it/s]
 95%|█████████▌| 353431/371472 [5:57:58<1:25:52,  3.50it/s]
 95%|█████████▌| 353432/371472 [5:57:58<1:23:18,  3.61it/s]
 95%|█████████▌| 353433/371472 [5:57:58<1:20:35,  3.73it/s]
 95%|█████████▌| 353434/371472 [5:57:58<1:18:32,  3.83it/s]
 95%|█████████▌| 353435/371472 [5:57:59<1:20:12,  3.75it/s]
 95%|█████████▌| 353436/371472 [5:57:59<1:22:00,  3.67it/s]
 95%|█████████▌| 353437/371472 [5:57:59<1:20:19,  3.74it/s]
 95%|█████████▌| 353438/371472 [5:57:59<1:18:06,  3.85it/s]
 95%|█████████▌| 353439/371472 [5:58:00<1:16:44,  3.92it/s]
 95%|█████████▌| 353440/371472 [5:58:00<1:21:12,  3.70it/s]
                                                           
{'loss': 2.6002, 'learning_rate': 1.4371134909177097e-07, 'epoch': 15.22}
+
 95%|█████████▌| 353440/371472 [5:58:00<1:21:12,  3.70it/s]
 95%|█████████▌| 353441/371472 [5:58:00<1:21:52,  3.67it/s]
 95%|█████████▌| 353442/371472 [5:58:00<1:18:46,  3.81it/s]
 95%|█████████▌| 353443/371472 [5:58:01<1:21:02,  3.71it/s]
 95%|█████████▌| 353444/371472 [5:58:01<1:19:27,  3.78it/s]
 95%|█████████▌| 353445/371472 [5:58:01<1:22:57,  3.62it/s]
 95%|█████████▌| 353446/371472 [5:58:02<1:21:03,  3.71it/s]
 95%|█████████▌| 353447/371472 [5:58:02<1:20:42,  3.72it/s]
 95%|█████████▌| 353448/371472 [5:58:02<1:19:47,  3.76it/s]
 95%|█████████▌| 353449/371472 [5:58:02<1:24:24,  3.56it/s]
 95%|█████████▌| 353450/371472 [5:58:03<1:25:00,  3.53it/s]
 95%|█████████▌| 353451/371472 [5:58:03<1:24:44,  3.54it/s]
 95%|█████████▌| 353452/371472 [5:58:03<1:25:29,  3.51it/s]
 95%|█████████▌| 353453/371472 [5:58:04<1:26:30,  3.47it/s]
 95%|█████████▌| 353454/371472 [5:58:04<1:33:43,  3.20it/s]
 95%|█████████▌| 353455/371472 [5:58:04<1:35:16,  3.15it/s]
 95%|█████████▌| 353456/371472 [5:58:05<1:31:11,  3.29it/s]
 95%|█████████▌| 353457/371472 [5:58:05<1:28:15,  3.40it/s]
 95%|█████████▌| 353458/371472 [5:58:05<1:25:08,  3.53it/s]
 95%|█████████▌| 353459/371472 [5:58:05<1:21:59,  3.66it/s]
 95%|█████████▌| 353460/371472 [5:58:06<1:20:50,  3.71it/s]
                                                           
{'loss': 2.5891, 'learning_rate': 1.4366286711629215e-07, 'epoch': 15.22}
+
 95%|█████████▌| 353460/371472 [5:58:06<1:20:50,  3.71it/s]
 95%|█████████▌| 353461/371472 [5:58:06<1:22:50,  3.62it/s]
 95%|█████████▌| 353462/371472 [5:58:06<1:23:04,  3.61it/s]
 95%|█████████▌| 353463/371472 [5:58:06<1:21:13,  3.70it/s]
 95%|█████████▌| 353464/371472 [5:58:07<1:18:08,  3.84it/s]
 95%|█████████▌| 353465/371472 [5:58:07<1:18:14,  3.84it/s]
 95%|█████████▌| 353466/371472 [5:58:07<1:18:40,  3.81it/s]
 95%|█████████▌| 353467/371472 [5:58:07<1:20:59,  3.71it/s]
 95%|█████████▌| 353468/371472 [5:58:08<1:30:21,  3.32it/s]
 95%|█████████▌| 353469/371472 [5:58:08<1:30:08,  3.33it/s]
 95%|█████████▌| 353470/371472 [5:58:08<1:33:09,  3.22it/s]
 95%|█████████▌| 353471/371472 [5:58:09<1:27:32,  3.43it/s]
 95%|█████████▌| 353472/371472 [5:58:09<1:23:55,  3.57it/s]
 95%|█████████▌| 353473/371472 [5:58:09<1:20:49,  3.71it/s]
 95%|█████████▌| 353474/371472 [5:58:09<1:22:00,  3.66it/s]
 95%|█████████▌| 353475/371472 [5:58:10<1:21:39,  3.67it/s]
 95%|█████████▌| 353476/371472 [5:58:10<1:20:55,  3.71it/s]
 95%|█████████▌| 353477/371472 [5:58:10<1:22:16,  3.65it/s]
 95%|█████████▌| 353478/371472 [5:58:11<1:24:15,  3.56it/s]
 95%|█████████▌| 353479/371472 [5:58:11<1:37:02,  3.09it/s]
 95%|█████████▌| 353480/371472 [5:58:11<1:32:17,  3.25it/s]
                                                           
{'loss': 2.5985, 'learning_rate': 1.436143851408132e-07, 'epoch': 15.23}
+
 95%|█████████▌| 353480/371472 [5:58:11<1:32:17,  3.25it/s]
 95%|█████████▌| 353481/371472 [5:58:12<1:31:27,  3.28it/s]
 95%|█████████▌| 353482/371472 [5:58:12<1:27:17,  3.43it/s]
 95%|█████████▌| 353483/371472 [5:58:12<1:25:18,  3.51it/s]
 95%|█████████▌| 353484/371472 [5:58:12<1:25:28,  3.51it/s]
 95%|█████████▌| 353485/371472 [5:58:13<1:22:56,  3.61it/s]
 95%|█████████▌| 353486/371472 [5:58:13<1:20:18,  3.73it/s]
 95%|█████████▌| 353487/371472 [5:58:13<1:22:42,  3.62it/s]
 95%|█████████▌| 353488/371472 [5:58:13<1:21:24,  3.68it/s]
 95%|█████████▌| 353489/371472 [5:58:14<1:20:23,  3.73it/s]
 95%|█████████▌| 353490/371472 [5:58:14<1:18:17,  3.83it/s]
 95%|█████████▌| 353491/371472 [5:58:14<1:17:52,  3.85it/s]
 95%|█████████▌| 353492/371472 [5:58:14<1:18:11,  3.83it/s]
 95%|█████████▌| 353493/371472 [5:58:15<1:25:12,  3.52it/s]
 95%|█████████▌| 353494/371472 [5:58:15<1:23:49,  3.57it/s]
 95%|█████████▌| 353495/371472 [5:58:15<1:24:07,  3.56it/s]
 95%|█████████▌| 353496/371472 [5:58:16<1:22:27,  3.63it/s]
 95%|█████████▌| 353497/371472 [5:58:16<1:21:44,  3.66it/s]
 95%|█████████▌| 353498/371472 [5:58:16<1:25:16,  3.51it/s]
 95%|█████████▌| 353499/371472 [5:58:16<1:23:20,  3.59it/s]
 95%|█████████▌| 353500/371472 [5:58:17<1:24:54,  3.53it/s]
                                                           
{'loss': 2.5123, 'learning_rate': 1.4356590316533434e-07, 'epoch': 15.23}
+
 95%|█████████▌| 353500/371472 [5:58:17<1:24:54,  3.53it/s]
 95%|█████████▌| 353501/371472 [5:58:17<1:26:58,  3.44it/s]
 95%|█████████▌| 353502/371472 [5:58:17<1:24:21,  3.55it/s]
 95%|█████████▌| 353503/371472 [5:58:18<1:23:06,  3.60it/s]
 95%|█████████▌| 353504/371472 [5:58:18<1:21:54,  3.66it/s]
 95%|█████████▌| 353505/371472 [5:58:18<1:21:31,  3.67it/s]
 95%|█████████▌| 353506/371472 [5:58:18<1:23:59,  3.57it/s]
 95%|█████████▌| 353507/371472 [5:58:19<1:25:53,  3.49it/s]
 95%|█████████▌| 353508/371472 [5:58:19<1:35:02,  3.15it/s]
 95%|█████████▌| 353509/371472 [5:58:19<1:30:14,  3.32it/s]
 95%|█████████▌| 353510/371472 [5:58:20<1:31:42,  3.26it/s]
 95%|█████████▌| 353511/371472 [5:58:20<1:33:12,  3.21it/s]
 95%|█████████▌| 353512/371472 [5:58:20<1:29:24,  3.35it/s]
 95%|█████████▌| 353513/371472 [5:58:21<1:28:05,  3.40it/s]
 95%|█████████▌| 353514/371472 [5:58:21<1:25:05,  3.52it/s]
 95%|█████████▌| 353515/371472 [5:58:21<1:23:14,  3.60it/s]
 95%|█████████▌| 353516/371472 [5:58:21<1:24:22,  3.55it/s]
 95%|█████████▌| 353517/371472 [5:58:22<1:23:28,  3.59it/s]
 95%|█████████▌| 353518/371472 [5:58:22<1:29:41,  3.34it/s]
 95%|█████████▌| 353519/371472 [5:58:22<1:27:23,  3.42it/s]
 95%|█████████▌| 353520/371472 [5:58:23<1:39:08,  3.02it/s]
                                                           
{'loss': 2.6804, 'learning_rate': 1.435174211898554e-07, 'epoch': 15.23}
+
 95%|█████████▌| 353520/371472 [5:58:23<1:39:08,  3.02it/s]
 95%|█████████▌| 353521/371472 [5:58:23<1:34:48,  3.16it/s]
 95%|█████████▌| 353522/371472 [5:58:23<1:29:11,  3.35it/s]
 95%|█████████▌| 353523/371472 [5:58:24<1:30:00,  3.32it/s]
 95%|█████████▌| 353524/371472 [5:58:24<1:29:38,  3.34it/s]
 95%|█████████▌| 353525/371472 [5:58:24<1:25:36,  3.49it/s]
 95%|█████████▌| 353526/371472 [5:58:24<1:26:01,  3.48it/s]
 95%|█████████▌| 353527/371472 [5:58:25<1:24:56,  3.52it/s]
 95%|█████████▌| 353528/371472 [5:58:25<1:22:07,  3.64it/s]
 95%|█████████▌| 353529/371472 [5:58:25<1:40:05,  2.99it/s]
 95%|█████████▌| 353530/371472 [5:58:26<1:38:06,  3.05it/s]
 95%|█████████▌| 353531/371472 [5:58:26<1:31:55,  3.25it/s]
 95%|█████████▌| 353532/371472 [5:58:26<1:35:13,  3.14it/s]
 95%|█████████▌| 353533/371472 [5:58:27<1:30:51,  3.29it/s]
 95%|█████████▌| 353534/371472 [5:58:27<1:27:36,  3.41it/s]
 95%|█████████▌| 353535/371472 [5:58:27<1:29:22,  3.34it/s]
 95%|█████████▌| 353536/371472 [5:58:27<1:30:02,  3.32it/s]
 95%|█████████▌| 353537/371472 [5:58:28<1:26:47,  3.44it/s]
 95%|█████████▌| 353538/371472 [5:58:28<1:26:42,  3.45it/s]
 95%|█████████▌| 353539/371472 [5:58:28<1:24:56,  3.52it/s]
 95%|█████████▌| 353540/371472 [5:58:29<1:25:40,  3.49it/s]
                                                           
{'loss': 2.4971, 'learning_rate': 1.4346893921437656e-07, 'epoch': 15.23}
+
 95%|█████████▌| 353540/371472 [5:58:29<1:25:40,  3.49it/s]
 95%|█████████▌| 353541/371472 [5:58:29<1:21:27,  3.67it/s]
 95%|█████████▌| 353542/371472 [5:58:29<1:23:14,  3.59it/s]
 95%|█████████▌| 353543/371472 [5:58:29<1:24:20,  3.54it/s]
 95%|█████████▌| 353544/371472 [5:58:30<1:27:09,  3.43it/s]
 95%|█████████▌| 353545/371472 [5:58:30<1:23:50,  3.56it/s]
 95%|█████████▌| 353546/371472 [5:58:30<1:22:44,  3.61it/s]
 95%|█████████▌| 353547/371472 [5:58:30<1:19:41,  3.75it/s]
 95%|█████████▌| 353548/371472 [5:58:31<1:18:24,  3.81it/s]
 95%|█████████▌| 353549/371472 [5:58:31<1:17:44,  3.84it/s]
 95%|█████████▌| 353550/371472 [5:58:31<1:19:46,  3.74it/s]
 95%|█████████▌| 353551/371472 [5:58:32<1:20:31,  3.71it/s]
 95%|█████████▌| 353552/371472 [5:58:32<1:19:40,  3.75it/s]
 95%|█████████▌| 353553/371472 [5:58:32<1:18:46,  3.79it/s]
 95%|█████████▌| 353554/371472 [5:58:32<1:17:45,  3.84it/s]
 95%|█████████▌| 353555/371472 [5:58:33<1:18:53,  3.78it/s]
 95%|█████████▌| 353556/371472 [5:58:33<1:21:41,  3.65it/s]
 95%|█████████▌| 353557/371472 [5:58:33<1:20:02,  3.73it/s]
 95%|█████████▌| 353558/371472 [5:58:33<1:20:11,  3.72it/s]
 95%|█████████▌| 353559/371472 [5:58:34<1:22:58,  3.60it/s]
 95%|█████████▌| 353560/371472 [5:58:34<1:24:10,  3.55it/s]
                                                           
{'loss': 2.5475, 'learning_rate': 1.434204572388976e-07, 'epoch': 15.23}
+
 95%|█████████▌| 353560/371472 [5:58:34<1:24:10,  3.55it/s]
 95%|█████████▌| 353561/371472 [5:58:34<1:23:31,  3.57it/s]
 95%|█████████▌| 353562/371472 [5:58:35<1:24:30,  3.53it/s]
 95%|█████████▌| 353563/371472 [5:58:35<1:22:07,  3.63it/s]
 95%|█████████▌| 353564/371472 [5:58:35<1:20:05,  3.73it/s]
 95%|█████████▌| 353565/371472 [5:58:35<1:23:11,  3.59it/s]
 95%|█████████▌| 353566/371472 [5:58:36<1:27:47,  3.40it/s]
 95%|█████████▌| 353567/371472 [5:58:36<1:28:38,  3.37it/s]
 95%|█████████▌| 353568/371472 [5:58:36<1:25:59,  3.47it/s]
 95%|█████████▌| 353569/371472 [5:58:37<1:25:20,  3.50it/s]
 95%|█████████▌| 353570/371472 [5:58:37<1:26:42,  3.44it/s]
 95%|█████████▌| 353571/371472 [5:58:37<1:27:20,  3.42it/s]
 95%|█████████▌| 353572/371472 [5:58:37<1:27:29,  3.41it/s]
 95%|█████████▌| 353573/371472 [5:58:38<1:27:33,  3.41it/s]
 95%|█████████▌| 353574/371472 [5:58:38<1:26:31,  3.45it/s]
 95%|█████████▌| 353575/371472 [5:58:38<1:27:04,  3.43it/s]
 95%|█████████▌| 353576/371472 [5:58:39<1:24:39,  3.52it/s]
 95%|█████████▌| 353577/371472 [5:58:39<1:26:26,  3.45it/s]
 95%|█████████▌| 353578/371472 [5:58:39<1:21:49,  3.64it/s]
 95%|█████████▌| 353579/371472 [5:58:39<1:19:06,  3.77it/s]
 95%|█████████▌| 353580/371472 [5:58:40<1:19:08,  3.77it/s]
                                                           
{'loss': 2.562, 'learning_rate': 1.4337197526341879e-07, 'epoch': 15.23}
+
 95%|█████████▌| 353580/371472 [5:58:40<1:19:08,  3.77it/s]
 95%|█████████▌| 353581/371472 [5:58:40<1:20:44,  3.69it/s]
 95%|█████████▌| 353582/371472 [5:58:40<1:19:57,  3.73it/s]
 95%|█████████▌| 353583/371472 [5:58:40<1:18:53,  3.78it/s]
 95%|█████████▌| 353584/371472 [5:58:41<1:21:00,  3.68it/s]
 95%|█████████▌| 353585/371472 [5:58:41<1:21:58,  3.64it/s]
 95%|█████████▌| 353586/371472 [5:58:41<1:29:40,  3.32it/s]
 95%|█████████▌| 353587/371472 [5:58:42<1:28:58,  3.35it/s]
 95%|█████████▌| 353588/371472 [5:58:42<1:25:01,  3.51it/s]
 95%|█████████▌| 353589/371472 [5:58:42<1:28:29,  3.37it/s]
 95%|█████████▌| 353590/371472 [5:58:43<1:24:45,  3.52it/s]
 95%|█████████▌| 353591/371472 [5:58:43<1:22:04,  3.63it/s]
 95%|█████████▌| 353592/371472 [5:58:43<1:29:05,  3.34it/s]
 95%|█████████▌| 353593/371472 [5:58:43<1:25:42,  3.48it/s]
 95%|█████████▌| 353594/371472 [5:58:44<1:23:54,  3.55it/s]
 95%|█████████▌| 353595/371472 [5:58:44<1:22:33,  3.61it/s]
 95%|█████████▌| 353596/371472 [5:58:44<1:24:22,  3.53it/s]
 95%|█████████▌| 353597/371472 [5:58:44<1:20:27,  3.70it/s]
 95%|█████████▌| 353598/371472 [5:58:45<1:22:07,  3.63it/s]
 95%|█████████▌| 353599/371472 [5:58:45<1:22:46,  3.60it/s]
 95%|█████████▌| 353600/371472 [5:58:45<1:21:41,  3.65it/s]
                                                           
{'loss': 2.6029, 'learning_rate': 1.4332349328793983e-07, 'epoch': 15.23}
+
 95%|█████████▌| 353600/371472 [5:58:45<1:21:41,  3.65it/s]
 95%|█████████▌| 353601/371472 [5:58:46<1:20:31,  3.70it/s]
 95%|█████████▌| 353602/371472 [5:58:46<1:20:12,  3.71it/s]
 95%|█████████▌| 353603/371472 [5:58:46<1:24:56,  3.51it/s]
 95%|█████████▌| 353604/371472 [5:58:46<1:26:05,  3.46it/s]
 95%|█████████▌| 353605/371472 [5:58:47<1:24:51,  3.51it/s]
 95%|█████████▌| 353606/371472 [5:58:47<1:23:13,  3.58it/s]
 95%|█████████▌| 353607/371472 [5:58:47<1:18:42,  3.78it/s]
 95%|█████████▌| 353608/371472 [5:58:47<1:20:22,  3.70it/s]
 95%|█████████▌| 353609/371472 [5:58:48<1:28:34,  3.36it/s]
 95%|█████████▌| 353610/371472 [5:58:48<1:26:34,  3.44it/s]
 95%|█████████▌| 353611/371472 [5:58:48<1:23:12,  3.58it/s]
 95%|█████████▌| 353612/371472 [5:58:49<1:20:11,  3.71it/s]
 95%|█████████▌| 353613/371472 [5:58:49<1:22:12,  3.62it/s]
 95%|█████████▌| 353614/371472 [5:58:49<1:20:20,  3.70it/s]
 95%|█████████▌| 353615/371472 [5:58:49<1:20:23,  3.70it/s]
 95%|█████████▌| 353616/371472 [5:58:50<1:22:53,  3.59it/s]
 95%|█████████▌| 353617/371472 [5:58:50<1:25:44,  3.47it/s]
 95%|█████████▌| 353618/371472 [5:58:50<1:22:30,  3.61it/s]
 95%|█████████▌| 353619/371472 [5:58:51<1:26:57,  3.42it/s]
 95%|█████████▌| 353620/371472 [5:58:51<1:24:41,  3.51it/s]
                                                           
{'loss': 2.5679, 'learning_rate': 1.4327501131246088e-07, 'epoch': 15.23}
+
 95%|█████████▌| 353620/371472 [5:58:51<1:24:41,  3.51it/s]
 95%|█████████▌| 353621/371472 [5:58:51<1:27:56,  3.38it/s]
 95%|█████████▌| 353622/371472 [5:58:52<1:29:16,  3.33it/s]
 95%|█████████▌| 353623/371472 [5:58:52<1:27:26,  3.40it/s]
 95%|█████████▌| 353624/371472 [5:58:52<1:30:25,  3.29it/s]
 95%|█████████▌| 353625/371472 [5:58:52<1:31:21,  3.26it/s]
 95%|█████████▌| 353626/371472 [5:58:53<1:27:37,  3.39it/s]
 95%|█████████▌| 353627/371472 [5:58:53<1:22:43,  3.59it/s]
 95%|█████████▌| 353628/371472 [5:58:53<1:24:56,  3.50it/s]
 95%|█████████▌| 353629/371472 [5:58:54<1:24:53,  3.50it/s]
 95%|█████████▌| 353630/371472 [5:58:54<1:25:48,  3.47it/s]
 95%|█████████▌| 353631/371472 [5:58:54<1:22:37,  3.60it/s]
 95%|█████████▌| 353632/371472 [5:58:54<1:24:45,  3.51it/s]
 95%|█████████▌| 353633/371472 [5:58:55<1:29:55,  3.31it/s]
 95%|█████████▌| 353634/371472 [5:58:55<1:26:45,  3.43it/s]
 95%|█████████▌| 353635/371472 [5:58:55<1:30:04,  3.30it/s]
 95%|█████████▌| 353636/371472 [5:58:56<1:25:32,  3.48it/s]
 95%|█████████▌| 353637/371472 [5:58:56<1:23:48,  3.55it/s]
 95%|█████████▌| 353638/371472 [5:58:56<1:23:54,  3.54it/s]
 95%|█████████▌| 353639/371472 [5:58:56<1:29:36,  3.32it/s]
 95%|█████████▌| 353640/371472 [5:58:57<1:27:04,  3.41it/s]
                                                           
{'loss': 2.517, 'learning_rate': 1.4322652933698205e-07, 'epoch': 15.23}
+
 95%|█████████▌| 353640/371472 [5:58:57<1:27:04,  3.41it/s]
 95%|█████████▌| 353641/371472 [5:58:57<1:24:27,  3.52it/s]
 95%|█████████▌| 353642/371472 [5:58:57<1:21:55,  3.63it/s]
 95%|█████████▌| 353643/371472 [5:58:58<1:20:10,  3.71it/s]
 95%|█████████▌| 353644/371472 [5:58:58<1:24:56,  3.50it/s]
 95%|█████████▌| 353645/371472 [5:58:58<1:20:45,  3.68it/s]
 95%|█████████▌| 353646/371472 [5:58:58<1:21:18,  3.65it/s]
 95%|█████████▌| 353647/371472 [5:58:59<1:21:24,  3.65it/s]
 95%|█████████▌| 353648/371472 [5:58:59<1:26:08,  3.45it/s]
 95%|█████████▌| 353649/371472 [5:58:59<1:26:45,  3.42it/s]
 95%|█████████▌| 353650/371472 [5:59:00<1:24:23,  3.52it/s]
 95%|█████████▌| 353651/371472 [5:59:00<1:28:37,  3.35it/s]
 95%|█████████▌| 353652/371472 [5:59:00<1:26:20,  3.44it/s]
 95%|█████████▌| 353653/371472 [5:59:00<1:30:28,  3.28it/s]
 95%|█████████▌| 353654/371472 [5:59:01<1:26:59,  3.41it/s]
 95%|█████████▌| 353655/371472 [5:59:01<1:24:27,  3.52it/s]
 95%|█████████▌| 353656/371472 [5:59:01<1:23:38,  3.55it/s]
 95%|█████████▌| 353657/371472 [5:59:02<1:22:11,  3.61it/s]
 95%|█████████▌| 353658/371472 [5:59:02<1:19:44,  3.72it/s]
 95%|█████████▌| 353659/371472 [5:59:02<1:18:34,  3.78it/s]
 95%|█████████▌| 353660/371472 [5:59:02<1:19:40,  3.73it/s]
                                                           
{'loss': 2.5771, 'learning_rate': 1.431780473615031e-07, 'epoch': 15.23}
+
 95%|█████████▌| 353660/371472 [5:59:02<1:19:40,  3.73it/s]
 95%|█████████▌| 353661/371472 [5:59:03<1:18:24,  3.79it/s]
 95%|█████████▌| 353662/371472 [5:59:03<1:18:38,  3.77it/s]
 95%|█████████▌| 353663/371472 [5:59:03<1:19:04,  3.75it/s]
 95%|█████████▌| 353664/371472 [5:59:03<1:25:11,  3.48it/s]
 95%|█████████▌| 353665/371472 [5:59:04<1:24:16,  3.52it/s]
 95%|█████████▌| 353666/371472 [5:59:04<1:24:29,  3.51it/s]
 95%|█████████▌| 353667/371472 [5:59:04<1:22:06,  3.61it/s]
 95%|█████████▌| 353668/371472 [5:59:05<1:21:59,  3.62it/s]
 95%|█████████▌| 353669/371472 [5:59:05<1:36:06,  3.09it/s]
 95%|█████████▌| 353670/371472 [5:59:05<1:29:30,  3.31it/s]
 95%|█████████▌| 353671/371472 [5:59:05<1:24:56,  3.49it/s]
 95%|█████████▌| 353672/371472 [5:59:06<1:22:59,  3.57it/s]
 95%|█████████▌| 353673/371472 [5:59:06<1:23:40,  3.55it/s]
 95%|█████████▌| 353674/371472 [5:59:06<1:22:29,  3.60it/s]
 95%|█████████▌| 353675/371472 [5:59:07<1:30:47,  3.27it/s]
 95%|█████████▌| 353676/371472 [5:59:07<1:32:21,  3.21it/s]
 95%|█████████▌| 353677/371472 [5:59:07<1:27:10,  3.40it/s]
 95%|█████████▌| 353678/371472 [5:59:08<1:22:22,  3.60it/s]
 95%|█████████▌| 353679/371472 [5:59:08<1:20:14,  3.70it/s]
 95%|█████████▌| 353680/371472 [5:59:08<1:17:57,  3.80it/s]
                                                           
{'loss': 2.7452, 'learning_rate': 1.4312956538602425e-07, 'epoch': 15.23}
+
 95%|█████████▌| 353680/371472 [5:59:08<1:17:57,  3.80it/s]
 95%|█████████▌| 353681/371472 [5:59:08<1:17:58,  3.80it/s]
 95%|█████████▌| 353682/371472 [5:59:09<1:24:21,  3.51it/s]
 95%|█████████▌| 353683/371472 [5:59:09<1:27:34,  3.39it/s]
 95%|█████████▌| 353684/371472 [5:59:09<1:25:06,  3.48it/s]
 95%|█████████▌| 353685/371472 [5:59:09<1:22:42,  3.58it/s]
 95%|█████████▌| 353686/371472 [5:59:10<1:21:11,  3.65it/s]
 95%|█████████▌| 353687/371472 [5:59:10<1:19:05,  3.75it/s]
 95%|█████████▌| 353688/371472 [5:59:10<1:20:05,  3.70it/s]
 95%|█████████▌| 353689/371472 [5:59:11<1:19:29,  3.73it/s]
 95%|█████████▌| 353690/371472 [5:59:11<1:30:20,  3.28it/s]
 95%|█████████▌| 353691/371472 [5:59:11<1:28:26,  3.35it/s]
 95%|█████████▌| 353692/371472 [5:59:11<1:26:38,  3.42it/s]
 95%|█████████▌| 353693/371472 [5:59:12<1:24:09,  3.52it/s]
 95%|█████████▌| 353694/371472 [5:59:12<1:21:29,  3.64it/s]
 95%|█████████▌| 353695/371472 [5:59:12<1:21:00,  3.66it/s]
 95%|█████████▌| 353696/371472 [5:59:13<1:22:26,  3.59it/s]
 95%|█████████▌| 353697/371472 [5:59:13<1:22:22,  3.60it/s]
 95%|█████████▌| 353698/371472 [5:59:13<1:20:53,  3.66it/s]
 95%|█████████▌| 353699/371472 [5:59:13<1:19:16,  3.74it/s]
 95%|█████████▌| 353700/371472 [5:59:14<1:18:43,  3.76it/s]
                                                           
{'loss': 2.7133, 'learning_rate': 1.4308108341054532e-07, 'epoch': 15.23}
+
 95%|█████████▌| 353700/371472 [5:59:14<1:18:43,  3.76it/s]
 95%|█████████▌| 353701/371472 [5:59:14<1:20:56,  3.66it/s]
 95%|█████████▌| 353702/371472 [5:59:14<1:24:26,  3.51it/s]
 95%|█████████▌| 353703/371472 [5:59:15<1:30:07,  3.29it/s]
 95%|█████████▌| 353704/371472 [5:59:15<1:27:24,  3.39it/s]
 95%|█████████▌| 353705/371472 [5:59:15<1:27:39,  3.38it/s]
 95%|█████████▌| 353706/371472 [5:59:15<1:25:15,  3.47it/s]
 95%|█████████▌| 353707/371472 [5:59:16<1:29:23,  3.31it/s]
 95%|█████████▌| 353708/371472 [5:59:16<1:26:31,  3.42it/s]
 95%|█████████▌| 353709/371472 [5:59:16<1:28:52,  3.33it/s]
 95%|█████████▌| 353710/371472 [5:59:17<1:25:56,  3.44it/s]
 95%|█████████▌| 353711/371472 [5:59:17<1:27:02,  3.40it/s]
 95%|█████████▌| 353712/371472 [5:59:17<1:23:19,  3.55it/s]
 95%|█████████▌| 353713/371472 [5:59:17<1:23:39,  3.54it/s]
 95%|█████████▌| 353714/371472 [5:59:18<1:22:50,  3.57it/s]
 95%|█████████▌| 353715/371472 [5:59:18<1:21:56,  3.61it/s]
 95%|█████████▌| 353716/371472 [5:59:18<1:21:49,  3.62it/s]
 95%|█████████▌| 353717/371472 [5:59:18<1:18:50,  3.75it/s]
 95%|█████████▌| 353718/371472 [5:59:19<1:17:52,  3.80it/s]
 95%|█████████▌| 353719/371472 [5:59:19<1:19:43,  3.71it/s]
 95%|█████████▌| 353720/371472 [5:59:19<1:21:20,  3.64it/s]
                                                           
{'loss': 2.569, 'learning_rate': 1.430326014350665e-07, 'epoch': 15.24}
+
 95%|█████████▌| 353720/371472 [5:59:19<1:21:20,  3.64it/s]
 95%|█████████▌| 353721/371472 [5:59:20<1:23:07,  3.56it/s]
 95%|█████████▌| 353722/371472 [5:59:20<1:23:23,  3.55it/s]
 95%|█████████▌| 353723/371472 [5:59:20<1:24:40,  3.49it/s]
 95%|█████████▌| 353724/371472 [5:59:20<1:25:37,  3.45it/s]
 95%|█████████▌| 353725/371472 [5:59:21<1:23:14,  3.55it/s]
 95%|█████████▌| 353726/371472 [5:59:21<1:20:58,  3.65it/s]
 95%|█████████▌| 353727/371472 [5:59:21<1:30:13,  3.28it/s]
 95%|█████████▌| 353728/371472 [5:59:22<1:28:30,  3.34it/s]
 95%|█████████▌| 353729/371472 [5:59:22<1:28:14,  3.35it/s]
 95%|█████████▌| 353730/371472 [5:59:22<1:25:28,  3.46it/s]
 95%|█████████▌| 353731/371472 [5:59:22<1:22:26,  3.59it/s]
 95%|█████████▌| 353732/371472 [5:59:23<1:21:34,  3.62it/s]
 95%|█████████▌| 353733/371472 [5:59:23<1:19:31,  3.72it/s]
 95%|█████████▌| 353734/371472 [5:59:23<1:24:51,  3.48it/s]
 95%|█████████▌| 353735/371472 [5:59:24<1:23:22,  3.55it/s]
 95%|█████████▌| 353736/371472 [5:59:24<1:31:10,  3.24it/s]
 95%|█████████▌| 353737/371472 [5:59:24<1:28:36,  3.34it/s]
 95%|█████████▌| 353738/371472 [5:59:25<1:31:54,  3.22it/s]
 95%|█████████▌| 353739/371472 [5:59:25<1:30:42,  3.26it/s]
 95%|█████████▌| 353740/371472 [5:59:25<1:26:48,  3.40it/s]
                                                           
{'loss': 2.6036, 'learning_rate': 1.4298411945958754e-07, 'epoch': 15.24}
+
 95%|█████████▌| 353740/371472 [5:59:25<1:26:48,  3.40it/s]
 95%|█████████▌| 353741/371472 [5:59:25<1:30:00,  3.28it/s]
 95%|█████████▌| 353742/371472 [5:59:26<1:26:48,  3.40it/s]
 95%|█████████▌| 353743/371472 [5:59:26<1:25:01,  3.48it/s]
 95%|█████████▌| 353744/371472 [5:59:26<1:24:47,  3.48it/s]
 95%|█████████▌| 353745/371472 [5:59:27<1:23:10,  3.55it/s]
 95%|█████████▌| 353746/371472 [5:59:27<1:31:42,  3.22it/s]
 95%|█████████▌| 353747/371472 [5:59:27<1:27:01,  3.39it/s]
 95%|█████████▌| 353748/371472 [5:59:27<1:23:40,  3.53it/s]
 95%|█████████▌| 353749/371472 [5:59:28<1:22:04,  3.60it/s]
 95%|█████████▌| 353750/371472 [5:59:28<1:22:45,  3.57it/s]
 95%|█████████▌| 353751/371472 [5:59:28<1:20:08,  3.69it/s]
 95%|█████████▌| 353752/371472 [5:59:29<1:22:41,  3.57it/s]
 95%|█████████▌| 353753/371472 [5:59:29<1:20:49,  3.65it/s]
 95%|█████████▌| 353754/371472 [5:59:29<1:18:31,  3.76it/s]
 95%|█████████▌| 353755/371472 [5:59:30<1:41:36,  2.91it/s]
 95%|█████████▌| 353756/371472 [5:59:30<1:36:29,  3.06it/s]
 95%|█████████▌| 353757/371472 [5:59:30<1:31:16,  3.23it/s]
 95%|█████████▌| 353758/371472 [5:59:30<1:28:42,  3.33it/s]
 95%|█████████▌| 353759/371472 [5:59:31<1:30:59,  3.24it/s]
 95%|█████████▌| 353760/371472 [5:59:31<1:29:09,  3.31it/s]
                                                           
{'loss': 2.6114, 'learning_rate': 1.429356374841087e-07, 'epoch': 15.24}
+
 95%|█████████▌| 353760/371472 [5:59:31<1:29:09,  3.31it/s]
 95%|█████████▌| 353761/371472 [5:59:31<1:23:39,  3.53it/s]
 95%|█████████▌| 353762/371472 [5:59:32<1:20:07,  3.68it/s]
 95%|█████████▌| 353763/371472 [5:59:32<1:21:29,  3.62it/s]
 95%|█████████▌| 353764/371472 [5:59:32<1:21:25,  3.62it/s]
 95%|█████████▌| 353765/371472 [5:59:32<1:19:54,  3.69it/s]
 95%|█████████▌| 353766/371472 [5:59:33<1:22:37,  3.57it/s]
 95%|█████████▌| 353767/371472 [5:59:33<1:22:14,  3.59it/s]
 95%|█████████▌| 353768/371472 [5:59:33<1:21:06,  3.64it/s]
 95%|█████████▌| 353769/371472 [5:59:33<1:22:04,  3.59it/s]
 95%|█████████▌| 353770/371472 [5:59:34<1:19:42,  3.70it/s]
 95%|█████████▌| 353771/371472 [5:59:34<1:19:50,  3.70it/s]
 95%|█████████▌| 353772/371472 [5:59:34<1:23:23,  3.54it/s]
 95%|█████████▌| 353773/371472 [5:59:35<1:24:59,  3.47it/s]
 95%|█████████▌| 353774/371472 [5:59:35<1:23:39,  3.53it/s]
 95%|█████████▌| 353775/371472 [5:59:35<1:23:42,  3.52it/s]
 95%|█████████▌| 353776/371472 [5:59:35<1:22:24,  3.58it/s]
 95%|█████████▌| 353777/371472 [5:59:36<1:22:39,  3.57it/s]
 95%|█████████▌| 353778/371472 [5:59:36<1:25:41,  3.44it/s]
 95%|█████████▌| 353779/371472 [5:59:36<1:27:17,  3.38it/s]
 95%|█████████▌| 353780/371472 [5:59:37<1:23:41,  3.52it/s]
                                                           
{'loss': 2.5741, 'learning_rate': 1.4288715550862977e-07, 'epoch': 15.24}
+
 95%|█████████▌| 353780/371472 [5:59:37<1:23:41,  3.52it/s]
 95%|█████████▌| 353781/371472 [5:59:37<1:22:51,  3.56it/s]
 95%|█████████▌| 353782/371472 [5:59:37<1:24:14,  3.50it/s]
 95%|█████████▌| 353783/371472 [5:59:37<1:23:12,  3.54it/s]
 95%|█████████▌| 353784/371472 [5:59:38<1:21:56,  3.60it/s]
 95%|█████████▌| 353785/371472 [5:59:38<1:23:08,  3.55it/s]
 95%|█████████▌| 353786/371472 [5:59:38<1:26:32,  3.41it/s]
 95%|█████████▌| 353787/371472 [5:59:39<1:25:55,  3.43it/s]
 95%|█████████▌| 353788/371472 [5:59:39<1:22:21,  3.58it/s]
 95%|█████████▌| 353789/371472 [5:59:39<1:23:33,  3.53it/s]
 95%|█████████▌| 353790/371472 [5:59:39<1:20:31,  3.66it/s]
 95%|█████████▌| 353791/371472 [5:59:40<1:20:17,  3.67it/s]
 95%|█████████▌| 353792/371472 [5:59:40<1:18:08,  3.77it/s]
 95%|█████████▌| 353793/371472 [5:59:40<1:20:16,  3.67it/s]
 95%|█████████▌| 353794/371472 [5:59:40<1:20:01,  3.68it/s]
 95%|█████████▌| 353795/371472 [5:59:41<1:28:38,  3.32it/s]
 95%|█████████▌| 353796/371472 [5:59:41<1:24:02,  3.51it/s]
 95%|█████████▌| 353797/371472 [5:59:41<1:22:09,  3.59it/s]
 95%|█████████▌| 353798/371472 [5:59:42<1:24:23,  3.49it/s]
 95%|█████████▌| 353799/371472 [5:59:42<1:24:37,  3.48it/s]
 95%|█████████▌| 353800/371472 [5:59:42<1:21:49,  3.60it/s]
                                                           
{'loss': 2.5969, 'learning_rate': 1.4283867353315092e-07, 'epoch': 15.24}
+
 95%|█████████▌| 353800/371472 [5:59:42<1:21:49,  3.60it/s]
 95%|█████████▌| 353801/371472 [5:59:42<1:19:38,  3.70it/s]
 95%|█████████▌| 353802/371472 [5:59:43<1:19:45,  3.69it/s]
 95%|█████████▌| 353803/371472 [5:59:43<1:18:10,  3.77it/s]
 95%|█████████▌| 353804/371472 [5:59:43<1:20:27,  3.66it/s]
 95%|█████████▌| 353805/371472 [5:59:44<1:21:43,  3.60it/s]
 95%|█████████▌| 353806/371472 [5:59:44<1:19:28,  3.70it/s]
 95%|█████████▌| 353807/371472 [5:59:44<1:25:54,  3.43it/s]
 95%|█████████▌| 353808/371472 [5:59:44<1:22:16,  3.58it/s]
 95%|█████████▌| 353809/371472 [5:59:45<1:20:55,  3.64it/s]
 95%|█████████▌| 353810/371472 [5:59:45<1:22:04,  3.59it/s]
 95%|█████████▌| 353811/371472 [5:59:45<1:23:56,  3.51it/s]
 95%|█████████▌| 353812/371472 [5:59:46<1:33:08,  3.16it/s]
 95%|█████████▌| 353813/371472 [5:59:46<1:28:40,  3.32it/s]
 95%|█████████▌| 353814/371472 [5:59:46<1:28:01,  3.34it/s]
 95%|█████████▌| 353815/371472 [5:59:47<1:26:39,  3.40it/s]
 95%|█████████▌| 353816/371472 [5:59:47<1:24:40,  3.48it/s]
 95%|█████████▌| 353817/371472 [5:59:47<1:28:08,  3.34it/s]
 95%|█████████▌| 353818/371472 [5:59:47<1:24:56,  3.46it/s]
 95%|█████████▌| 353819/371472 [5:59:48<1:22:26,  3.57it/s]
 95%|█████████▌| 353820/371472 [5:59:48<1:20:27,  3.66it/s]
                                                           
{'loss': 2.471, 'learning_rate': 1.4279019155767196e-07, 'epoch': 15.24}
+
 95%|█████████▌| 353820/371472 [5:59:48<1:20:27,  3.66it/s]
 95%|█████████▌| 353821/371472 [5:59:48<1:30:13,  3.26it/s]
 95%|█████████▌| 353822/371472 [5:59:49<1:26:58,  3.38it/s]
 95%|█████████▌| 353823/371472 [5:59:49<1:28:15,  3.33it/s]
 95%|█████████▌| 353824/371472 [5:59:49<1:25:53,  3.42it/s]
 95%|█████████▌| 353825/371472 [5:59:49<1:26:29,  3.40it/s]
 95%|█████████▌| 353826/371472 [5:59:50<1:25:03,  3.46it/s]
 95%|█████████▌| 353827/371472 [5:59:50<1:29:15,  3.29it/s]
 95%|█████████▌| 353828/371472 [5:59:50<1:28:30,  3.32it/s]
 95%|█████████▌| 353829/371472 [5:59:51<1:26:54,  3.38it/s]
 95%|█████████▌| 353830/371472 [5:59:51<1:24:06,  3.50it/s]
 95%|█████████▌| 353831/371472 [5:59:51<1:25:34,  3.44it/s]
 95%|█████████▌| 353832/371472 [5:59:51<1:23:36,  3.52it/s]
 95%|█████████▌| 353833/371472 [5:59:52<1:26:11,  3.41it/s]
 95%|█████████▌| 353834/371472 [5:59:52<1:22:52,  3.55it/s]
 95%|█████████▌| 353835/371472 [5:59:52<1:24:35,  3.48it/s]
 95%|█████████▌| 353836/371472 [5:59:53<1:24:08,  3.49it/s]
 95%|█████████▌| 353837/371472 [5:59:53<1:23:30,  3.52it/s]
 95%|█████████▌| 353838/371472 [5:59:53<1:22:28,  3.56it/s]
 95%|█████████▌| 353839/371472 [5:59:53<1:20:02,  3.67it/s]
 95%|█████████▌| 353840/371472 [5:59:54<1:17:35,  3.79it/s]
                                                           
{'loss': 2.5252, 'learning_rate': 1.4274170958219314e-07, 'epoch': 15.24}
+
 95%|█████████▌| 353840/371472 [5:59:54<1:17:35,  3.79it/s]
 95%|█████████▌| 353841/371472 [5:59:54<1:16:39,  3.83it/s]
 95%|█████████▌| 353842/371472 [5:59:54<1:20:04,  3.67it/s]
 95%|█████████▌| 353843/371472 [5:59:54<1:20:13,  3.66it/s]
 95%|█████████▌| 353844/371472 [5:59:55<1:25:14,  3.45it/s]
 95%|█████████▌| 353845/371472 [5:59:55<1:27:57,  3.34it/s]
 95%|█████████▌| 353846/371472 [5:59:55<1:27:11,  3.37it/s]
 95%|█████████▌| 353847/371472 [5:59:56<1:24:12,  3.49it/s]
 95%|█████████▌| 353848/371472 [5:59:56<1:24:48,  3.46it/s]
 95%|█████████▌| 353849/371472 [5:59:56<1:27:00,  3.38it/s]
 95%|█████████▌| 353850/371472 [5:59:57<1:26:43,  3.39it/s]
 95%|█████████▌| 353851/371472 [5:59:57<1:23:21,  3.52it/s]
 95%|█████████▌| 353852/371472 [5:59:57<1:22:56,  3.54it/s]
 95%|█████████▌| 353853/371472 [5:59:57<1:20:05,  3.67it/s]
 95%|█████████▌| 353854/371472 [5:59:58<1:19:38,  3.69it/s]
 95%|█████████▌| 353855/371472 [5:59:58<1:18:38,  3.73it/s]
 95%|█████████▌| 353856/371472 [5:59:58<1:18:34,  3.74it/s]
 95%|█████████▌| 353857/371472 [5:59:58<1:19:26,  3.70it/s]
 95%|█████████▌| 353858/371472 [5:59:59<1:28:08,  3.33it/s]
 95%|█████████▌| 353859/371472 [5:59:59<1:25:45,  3.42it/s]
 95%|█████████▌| 353860/371472 [5:59:59<1:24:27,  3.48it/s]
                                                           
{'loss': 2.526, 'learning_rate': 1.4269322760671418e-07, 'epoch': 15.24}
+
 95%|█████████▌| 353860/371472 [5:59:59<1:24:27,  3.48it/s]
 95%|█████████▌| 353861/371472 [6:00:00<1:21:23,  3.61it/s]
 95%|█████████▌| 353862/371472 [6:00:00<1:20:22,  3.65it/s]
 95%|█████████▌| 353863/371472 [6:00:00<1:20:41,  3.64it/s]
 95%|█████████▌| 353864/371472 [6:00:00<1:21:26,  3.60it/s]
 95%|█████████▌| 353865/371472 [6:00:01<1:30:58,  3.23it/s]
 95%|█████████▌| 353866/371472 [6:00:01<1:26:53,  3.38it/s]
 95%|█████████▌| 353867/371472 [6:00:01<1:24:42,  3.46it/s]
 95%|█████████▌| 353868/371472 [6:00:02<1:28:35,  3.31it/s]
 95%|█████████▌| 353869/371472 [6:00:02<1:27:22,  3.36it/s]
 95%|█████████▌| 353870/371472 [6:00:02<1:26:15,  3.40it/s]
 95%|█████████▌| 353871/371472 [6:00:03<1:31:44,  3.20it/s]
 95%|█████████▌| 353872/371472 [6:00:03<1:26:25,  3.39it/s]
 95%|█████████▌| 353873/371472 [6:00:03<1:23:44,  3.50it/s]
 95%|█████████▌| 353874/371472 [6:00:03<1:22:16,  3.56it/s]
 95%|█████████▌| 353875/371472 [6:00:04<1:25:06,  3.45it/s]
 95%|█████████▌| 353876/371472 [6:00:04<1:22:22,  3.56it/s]
 95%|█████████▌| 353877/371472 [6:00:04<1:23:58,  3.49it/s]
 95%|█████████▌| 353878/371472 [6:00:05<1:23:19,  3.52it/s]
 95%|█████████▌| 353879/371472 [6:00:05<1:24:41,  3.46it/s]
 95%|█████████▌| 353880/371472 [6:00:05<1:25:10,  3.44it/s]
                                                           
{'loss': 2.6591, 'learning_rate': 1.4264474563123534e-07, 'epoch': 15.24}
+
 95%|█████████▌| 353880/371472 [6:00:05<1:25:10,  3.44it/s]
 95%|█████████▌| 353881/371472 [6:00:05<1:29:15,  3.28it/s]
 95%|█████████▌| 353882/371472 [6:00:06<1:26:56,  3.37it/s]
 95%|█████████▌| 353883/371472 [6:00:06<1:36:29,  3.04it/s]
 95%|█████████▌| 353884/371472 [6:00:06<1:34:21,  3.11it/s]
 95%|█████████▌| 353885/371472 [6:00:07<1:30:08,  3.25it/s]
 95%|█████████▌| 353886/371472 [6:00:07<1:25:42,  3.42it/s]
 95%|█████████▌| 353887/371472 [6:00:07<1:23:52,  3.49it/s]
 95%|█████████▌| 353888/371472 [6:00:08<1:23:23,  3.51it/s]
 95%|█████████▌| 353889/371472 [6:00:08<1:21:44,  3.58it/s]
 95%|█████████▌| 353890/371472 [6:00:08<1:20:45,  3.63it/s]
 95%|█████████▌| 353891/371472 [6:00:08<1:24:55,  3.45it/s]
 95%|█████████▌| 353892/371472 [6:00:09<1:29:50,  3.26it/s]
 95%|█████████▌| 353893/371472 [6:00:09<1:35:51,  3.06it/s]
 95%|█████████▌| 353894/371472 [6:00:09<1:27:39,  3.34it/s]
 95%|█████████▌| 353895/371472 [6:00:10<1:28:23,  3.31it/s]
 95%|█████████▌| 353896/371472 [6:00:10<1:24:08,  3.48it/s]
 95%|█████████▌| 353897/371472 [6:00:10<1:21:09,  3.61it/s]
 95%|█████████▌| 353898/371472 [6:00:10<1:20:47,  3.63it/s]
 95%|█████████▌| 353899/371472 [6:00:11<1:20:59,  3.62it/s]
 95%|█████████▌| 353900/371472 [6:00:11<1:25:03,  3.44it/s]
                                                           
{'loss': 2.5319, 'learning_rate': 1.425962636557564e-07, 'epoch': 15.24}
+
 95%|█████████▌| 353900/371472 [6:00:11<1:25:03,  3.44it/s]
 95%|█████████▌| 353901/371472 [6:00:11<1:22:53,  3.53it/s]
 95%|█████████▌| 353902/371472 [6:00:12<1:22:37,  3.54it/s]
 95%|█████████▌| 353903/371472 [6:00:12<1:28:47,  3.30it/s]
 95%|█████████▌| 353904/371472 [6:00:12<1:24:50,  3.45it/s]
 95%|█████████▌| 353905/371472 [6:00:12<1:22:18,  3.56it/s]
 95%|█████████▌| 353906/371472 [6:00:13<1:21:19,  3.60it/s]
 95%|█████████▌| 353907/371472 [6:00:13<1:18:50,  3.71it/s]
 95%|█████████▌| 353908/371472 [6:00:13<1:21:58,  3.57it/s]
 95%|█████████▌| 353909/371472 [6:00:14<1:18:05,  3.75it/s]
 95%|█████████▌| 353910/371472 [6:00:14<1:17:31,  3.78it/s]
 95%|█████████▌| 353911/371472 [6:00:14<1:16:39,  3.82it/s]
 95%|█████████▌| 353912/371472 [6:00:14<1:17:10,  3.79it/s]
 95%|█████████▌| 353913/371472 [6:00:15<1:18:06,  3.75it/s]
 95%|█████████▌| 353914/371472 [6:00:15<1:17:28,  3.78it/s]
 95%|█████████▌| 353915/371472 [6:00:15<1:17:08,  3.79it/s]
 95%|█████████▌| 353916/371472 [6:00:15<1:19:27,  3.68it/s]
 95%|█████████▌| 353917/371472 [6:00:16<1:25:02,  3.44it/s]
 95%|█████████▌| 353918/371472 [6:00:16<1:22:44,  3.54it/s]
 95%|█████████▌| 353919/371472 [6:00:16<1:21:28,  3.59it/s]
 95%|█████████▌| 353920/371472 [6:00:17<1:23:34,  3.50it/s]
                                                           
{'loss': 2.6817, 'learning_rate': 1.4254778168027756e-07, 'epoch': 15.24}
+
 95%|█████████▌| 353920/371472 [6:00:17<1:23:34,  3.50it/s]
 95%|█████████▌| 353921/371472 [6:00:17<1:20:54,  3.62it/s]
 95%|█████████▌| 353922/371472 [6:00:17<1:25:39,  3.42it/s]
 95%|█████████▌| 353923/371472 [6:00:17<1:28:25,  3.31it/s]
 95%|█████████▌| 353924/371472 [6:00:18<1:25:28,  3.42it/s]
 95%|█████████▌| 353925/371472 [6:00:18<1:32:07,  3.17it/s]
 95%|█████████▌| 353926/371472 [6:00:18<1:28:09,  3.32it/s]
 95%|█████████▌| 353927/371472 [6:00:19<1:28:25,  3.31it/s]
 95%|█████████▌| 353928/371472 [6:00:19<1:24:23,  3.46it/s]
 95%|█████████▌| 353929/371472 [6:00:19<1:21:45,  3.58it/s]
 95%|█████████▌| 353930/371472 [6:00:19<1:19:33,  3.67it/s]
 95%|█████████▌| 353931/371472 [6:00:20<1:23:25,  3.50it/s]
 95%|█████████▌| 353932/371472 [6:00:20<1:26:44,  3.37it/s]
 95%|█████████▌| 353933/371472 [6:00:20<1:25:50,  3.41it/s]
 95%|█████████▌| 353934/371472 [6:00:21<1:24:41,  3.45it/s]
 95%|█████████▌| 353935/371472 [6:00:21<1:27:21,  3.35it/s]
 95%|█████████▌| 353936/371472 [6:00:21<1:25:31,  3.42it/s]
 95%|█████████▌| 353937/371472 [6:00:22<1:21:36,  3.58it/s]
 95%|█████████▌| 353938/371472 [6:00:22<1:20:12,  3.64it/s]
 95%|█████████▌| 353939/371472 [6:00:22<1:28:00,  3.32it/s]
 95%|█████████▌| 353940/371472 [6:00:22<1:24:40,  3.45it/s]
                                                           
{'loss': 2.6248, 'learning_rate': 1.424992997047986e-07, 'epoch': 15.24}
+
 95%|█████████▌| 353940/371472 [6:00:22<1:24:40,  3.45it/s]
 95%|█████████▌| 353941/371472 [6:00:23<1:24:10,  3.47it/s]
 95%|█████████▌| 353942/371472 [6:00:23<1:22:34,  3.54it/s]
 95%|█████████▌| 353943/371472 [6:00:23<1:21:22,  3.59it/s]
 95%|█████████▌| 353944/371472 [6:00:23<1:19:03,  3.70it/s]
 95%|█████████▌| 353945/371472 [6:00:24<1:18:22,  3.73it/s]
 95%|█████████▌| 353946/371472 [6:00:24<1:18:57,  3.70it/s]
 95%|█████████▌| 353947/371472 [6:00:24<1:18:02,  3.74it/s]
 95%|█████████▌| 353948/371472 [6:00:25<1:17:49,  3.75it/s]
 95%|█████████▌| 353949/371472 [6:00:25<1:19:14,  3.69it/s]
 95%|█████████▌| 353950/371472 [6:00:25<1:18:33,  3.72it/s]
 95%|█████████▌| 353951/371472 [6:00:25<1:18:50,  3.70it/s]
 95%|█████████▌| 353952/371472 [6:00:26<1:17:53,  3.75it/s]
 95%|█████████▌| 353953/371472 [6:00:26<1:17:59,  3.74it/s]
 95%|█████████▌| 353954/371472 [6:00:26<1:19:17,  3.68it/s]
 95%|█████████▌| 353955/371472 [6:00:26<1:17:08,  3.78it/s]
 95%|█████████▌| 353956/371472 [6:00:27<1:22:59,  3.52it/s]
 95%|███���█████▌| 353957/371472 [6:00:27<1:26:02,  3.39it/s]
 95%|█████████▌| 353958/371472 [6:00:27<1:24:54,  3.44it/s]
 95%|█████████▌| 353959/371472 [6:00:28<1:23:25,  3.50it/s]
 95%|█████████▌| 353960/371472 [6:00:28<1:21:01,  3.60it/s]
                                                           
{'loss': 2.6229, 'learning_rate': 1.4245081772931978e-07, 'epoch': 15.25}
+
 95%|█████████▌| 353960/371472 [6:00:28<1:21:01,  3.60it/s]
 95%|█████████▌| 353961/371472 [6:00:28<1:20:19,  3.63it/s]
 95%|█████████▌| 353962/371472 [6:00:28<1:24:15,  3.46it/s]
 95%|█████████▌| 353963/371472 [6:00:29<1:27:02,  3.35it/s]
 95%|█████████▌| 353964/371472 [6:00:29<1:25:55,  3.40it/s]
 95%|█████████▌| 353965/371472 [6:00:29<1:24:21,  3.46it/s]
 95%|█████████▌| 353966/371472 [6:00:30<1:21:47,  3.57it/s]
 95%|█████████▌| 353967/371472 [6:00:30<1:19:49,  3.66it/s]
 95%|█████████▌| 353968/371472 [6:00:30<1:19:01,  3.69it/s]
 95%|█████████▌| 353969/371472 [6:00:30<1:23:56,  3.48it/s]
 95%|█████████▌| 353970/371472 [6:00:31<1:24:55,  3.43it/s]
 95%|█████████▌| 353971/371472 [6:00:31<1:21:32,  3.58it/s]
 95%|█████████▌| 353972/371472 [6:00:31<1:19:47,  3.66it/s]
 95%|█████████▌| 353973/371472 [6:00:32<1:20:38,  3.62it/s]
 95%|█████████▌| 353974/371472 [6:00:32<1:20:52,  3.61it/s]
 95%|█████████▌| 353975/371472 [6:00:32<1:21:35,  3.57it/s]
 95%|█████████▌| 353976/371472 [6:00:32<1:21:45,  3.57it/s]
 95%|█████████▌| 353977/371472 [6:00:33<1:19:56,  3.65it/s]
 95%|█████████▌| 353978/371472 [6:00:33<1:22:12,  3.55it/s]
 95%|█████████▌| 353979/371472 [6:00:33<1:24:52,  3.44it/s]
 95%|█████████▌| 353980/371472 [6:00:34<1:21:01,  3.60it/s]
                                                           
{'loss': 2.5831, 'learning_rate': 1.4240233575384083e-07, 'epoch': 15.25}
+
 95%|█████████▌| 353980/371472 [6:00:34<1:21:01,  3.60it/s]
 95%|█████████▌| 353981/371472 [6:00:34<1:20:56,  3.60it/s]
 95%|█████████▌| 353982/371472 [6:00:34<1:22:54,  3.52it/s]
 95%|█████████▌| 353983/371472 [6:00:34<1:21:08,  3.59it/s]
 95%|█████████▌| 353984/371472 [6:00:35<1:22:05,  3.55it/s]
 95%|█████████▌| 353985/371472 [6:00:35<1:22:05,  3.55it/s]
 95%|█████████▌| 353986/371472 [6:00:35<1:23:08,  3.51it/s]
 95%|█████████▌| 353987/371472 [6:00:36<1:24:57,  3.43it/s]
 95%|█████████▌| 353988/371472 [6:00:36<1:22:04,  3.55it/s]
 95%|█████████▌| 353989/371472 [6:00:36<1:20:37,  3.61it/s]
 95%|█████████▌| 353990/371472 [6:00:36<1:20:27,  3.62it/s]
 95%|█████████▌| 353991/371472 [6:00:37<1:18:41,  3.70it/s]
 95%|█████████▌| 353992/371472 [6:00:37<1:17:01,  3.78it/s]
 95%|█████████▌| 353993/371472 [6:00:37<1:19:30,  3.66it/s]
 95%|█████████▌| 353994/371472 [6:00:37<1:24:57,  3.43it/s]
 95%|█████████▌| 353995/371472 [6:00:38<1:23:35,  3.48it/s]
 95%|█████████▌| 353996/371472 [6:00:38<1:24:03,  3.46it/s]
 95%|█████████▌| 353997/371472 [6:00:38<1:22:59,  3.51it/s]
 95%|█████████▌| 353998/371472 [6:00:39<1:24:13,  3.46it/s]
 95%|█████████▌| 353999/371472 [6:00:39<1:22:08,  3.55it/s]
 95%|█████████▌| 354000/371472 [6:00:39<1:21:44,  3.56it/s]
                                                           
{'loss': 2.7255, 'learning_rate': 1.4235385377836198e-07, 'epoch': 15.25}
+
 95%|█████████▌| 354000/371472 [6:00:39<1:21:44,  3.56it/s]
 95%|█████████▌| 354001/371472 [6:00:39<1:20:55,  3.60it/s]
 95%|█████████▌| 354002/371472 [6:00:40<1:19:22,  3.67it/s]
 95%|█████████▌| 354003/371472 [6:00:40<1:17:24,  3.76it/s]
 95%|█████████▌| 354004/371472 [6:00:40<1:31:39,  3.18it/s]
 95%|█████████▌| 354005/371472 [6:00:41<1:28:37,  3.28it/s]
 95%|█████████▌| 354006/371472 [6:00:41<1:25:14,  3.42it/s]
 95%|█████████▌| 354007/371472 [6:00:41<1:21:50,  3.56it/s]
 95%|█████████▌| 354008/371472 [6:00:41<1:22:01,  3.55it/s]
 95%|█████████▌| 354009/371472 [6:00:42<1:26:15,  3.37it/s]
 95%|█████████▌| 354010/371472 [6:00:42<1:25:14,  3.41it/s]
 95%|█████████▌| 354011/371472 [6:00:42<1:26:58,  3.35it/s]
 95%|█████████▌| 354012/371472 [6:00:43<1:22:48,  3.51it/s]
 95%|��████████▌| 354013/371472 [6:00:43<1:23:36,  3.48it/s]
 95%|█████████▌| 354014/371472 [6:00:43<1:21:18,  3.58it/s]
 95%|█████████▌| 354015/371472 [6:00:43<1:21:09,  3.59it/s]
 95%|█████████▌| 354016/371472 [6:00:44<1:22:59,  3.51it/s]
 95%|█████████▌| 354017/371472 [6:00:44<1:24:33,  3.44it/s]
 95%|█████████▌| 354018/371472 [6:00:44<1:20:41,  3.61it/s]
 95%|█████████▌| 354019/371472 [6:00:45<1:21:45,  3.56it/s]
 95%|█████████▌| 354020/371472 [6:00:45<1:21:03,  3.59it/s]
                                                           
{'loss': 2.5514, 'learning_rate': 1.4230537180288305e-07, 'epoch': 15.25}
+
 95%|█████████▌| 354020/371472 [6:00:45<1:21:03,  3.59it/s]
 95%|█████████▌| 354021/371472 [6:00:45<1:19:27,  3.66it/s]
 95%|█████████▌| 354022/371472 [6:00:45<1:16:48,  3.79it/s]
 95%|█████████▌| 354023/371472 [6:00:46<1:19:28,  3.66it/s]
 95%|█████████▌| 354024/371472 [6:00:46<1:18:07,  3.72it/s]
 95%|█████████▌| 354025/371472 [6:00:46<1:19:10,  3.67it/s]
 95%|█████████▌| 354026/371472 [6:00:46<1:17:53,  3.73it/s]
 95%|█████████▌| 354027/371472 [6:00:47<1:19:46,  3.64it/s]
 95%|█████████▌| 354028/371472 [6:00:47<1:23:17,  3.49it/s]
 95%|█████████▌| 354029/371472 [6:00:47<1:26:23,  3.37it/s]
 95%|█████████▌| 354030/371472 [6:00:48<1:24:33,  3.44it/s]
 95%|█████████▌| 354031/371472 [6:00:48<1:23:25,  3.48it/s]
 95%|█████████▌| 354032/371472 [6:00:48<1:20:35,  3.61it/s]
 95%|█████████▌| 354033/371472 [6:00:49<1:22:36,  3.52it/s]
 95%|█████████▌| 354034/371472 [6:00:49<1:21:32,  3.56it/s]
 95%|█████████▌| 354035/371472 [6:00:49<1:20:46,  3.60it/s]
 95%|█████████▌| 354036/371472 [6:00:49<1:24:03,  3.46it/s]
 95%|█████████▌| 354037/371472 [6:00:50<1:22:15,  3.53it/s]
 95%|█████████▌| 354038/371472 [6:00:50<1:21:53,  3.55it/s]
 95%|█████████▌| 354039/371472 [6:00:50<1:22:44,  3.51it/s]
 95%|█████████▌| 354040/371472 [6:00:50<1:21:58,  3.54it/s]
                                                           
{'loss': 2.7804, 'learning_rate': 1.422568898274042e-07, 'epoch': 15.25}
+
 95%|█████████▌| 354040/371472 [6:00:50<1:21:58,  3.54it/s]
 95%|█████████▌| 354041/371472 [6:00:51<1:21:39,  3.56it/s]
 95%|█████████▌| 354042/371472 [6:00:51<1:21:09,  3.58it/s]
 95%|█████████▌| 354043/371472 [6:00:51<1:19:36,  3.65it/s]
 95%|█████████▌| 354044/371472 [6:00:52<1:23:20,  3.49it/s]
 95%|█████████▌| 354045/371472 [6:00:52<1:19:54,  3.64it/s]
 95%|█████████▌| 354046/371472 [6:00:52<1:18:59,  3.68it/s]
 95%|█████████▌| 354047/371472 [6:00:52<1:23:45,  3.47it/s]
 95%|█████████▌| 354048/371472 [6:00:53<1:26:21,  3.36it/s]
 95%|█████████▌| 354049/371472 [6:00:53<1:25:36,  3.39it/s]
 95%|█████████▌| 354050/371472 [6:00:53<1:33:40,  3.10it/s]
 95%|█████████▌| 354051/371472 [6:00:54<1:31:18,  3.18it/s]
 95%|█████████▌| 354052/371472 [6:00:54<1:29:12,  3.25it/s]
 95%|█████████▌| 354053/371472 [6:00:54<1:23:15,  3.49it/s]
 95%|█████████▌| 354054/371472 [6:00:55<1:21:24,  3.57it/s]
 95%|█████████▌| 354055/371472 [6:00:55<1:19:20,  3.66it/s]
 95%|█████████▌| 354056/371472 [6:00:55<1:20:05,  3.62it/s]
 95%|█████████▌| 354057/371472 [6:00:55<1:19:21,  3.66it/s]
 95%|█████████▌| 354058/371472 [6:00:56<1:21:50,  3.55it/s]
 95%|█████████▌| 354059/371472 [6:00:56<1:18:41,  3.69it/s]
 95%|█████████▌| 354060/371472 [6:00:56<1:21:25,  3.56it/s]
                                                           
{'loss': 2.6442, 'learning_rate': 1.4220840785192527e-07, 'epoch': 15.25}
+
 95%|█████████▌| 354060/371472 [6:00:56<1:21:25,  3.56it/s]
 95%|█████████▌| 354061/371472 [6:00:57<1:27:57,  3.30it/s]
 95%|█████████▌| 354062/371472 [6:00:57<1:27:52,  3.30it/s]
 95%|█████████▌| 354063/371472 [6:00:57<1:30:08,  3.22it/s]
 95%|█████████▌| 354064/371472 [6:00:57<1:26:22,  3.36it/s]
 95%|█████████▌| 354065/371472 [6:00:58<1:23:48,  3.46it/s]
 95%|█████████▌| 354066/371472 [6:00:58<1:26:37,  3.35it/s]
 95%|█████████▌| 354067/371472 [6:00:58<1:27:18,  3.32it/s]
 95%|█████████▌| 354068/371472 [6:00:59<1:27:07,  3.33it/s]
 95%|█████████▌| 354069/371472 [6:00:59<1:24:58,  3.41it/s]
 95%|█████████▌| 354070/371472 [6:00:59<1:21:09,  3.57it/s]
 95%|█████████▌| 354071/371472 [6:00:59<1:20:05,  3.62it/s]
 95%|█████████▌| 354072/371472 [6:01:00<1:21:15,  3.57it/s]
 95%|█████████▌| 354073/371472 [6:01:00<1:20:26,  3.61it/s]
 95%|█████████▌| 354074/371472 [6:01:00<1:18:27,  3.70it/s]
 95%|█████████▌| 354075/371472 [6:01:01<1:18:49,  3.68it/s]
 95%|█████████▌| 354076/371472 [6:01:01<1:20:50,  3.59it/s]
 95%|█████████▌| 354077/371472 [6:01:01<1:19:01,  3.67it/s]
 95%|█████████▌| 354078/371472 [6:01:01<1:16:33,  3.79it/s]
 95%|█████████▌| 354079/371472 [6:01:02<1:20:02,  3.62it/s]
 95%|█████████▌| 354080/371472 [6:01:02<1:20:30,  3.60it/s]
                                                           
{'loss': 2.4991, 'learning_rate': 1.4215992587644642e-07, 'epoch': 15.25}
+
 95%|█████████▌| 354080/371472 [6:01:02<1:20:30,  3.60it/s]
 95%|█████████▌| 354081/371472 [6:01:02<1:18:02,  3.71it/s]
 95%|█████████▌| 354082/371472 [6:01:02<1:21:55,  3.54it/s]
 95%|█████████▌| 354083/371472 [6:01:03<1:21:22,  3.56it/s]
 95%|█████████▌| 354084/371472 [6:01:03<1:25:49,  3.38it/s]
 95%|█████████▌| 354085/371472 [6:01:03<1:28:14,  3.28it/s]
 95%|█████████▌| 354086/371472 [6:01:04<1:29:07,  3.25it/s]
 95%|█████████▌| 354087/371472 [6:01:04<1:26:49,  3.34it/s]
 95%|█████████▌| 354088/371472 [6:01:04<1:29:19,  3.24it/s]
 95%|█████████▌| 354089/371472 [6:01:05<1:25:42,  3.38it/s]
 95%|█████████▌| 354090/371472 [6:01:05<1:26:51,  3.34it/s]
 95%|█████████▌| 354091/371472 [6:01:05<1:23:37,  3.46it/s]
 95%|█████████▌| 354092/371472 [6:01:06<1:39:12,  2.92it/s]
 95%|█████████▌| 354093/371472 [6:01:06<1:36:48,  2.99it/s]
 95%|█████████▌| 354094/371472 [6:01:06<1:32:42,  3.12it/s]
 95%|█████████▌| 354095/371472 [6:01:07<1:26:39,  3.34it/s]
 95%|█████████▌| 354096/371472 [6:01:07<1:22:33,  3.51it/s]
 95%|█████████▌| 354097/371472 [6:01:07<1:26:41,  3.34it/s]
 95%|█████████▌| 354098/371472 [6:01:07<1:25:43,  3.38it/s]
 95%|█████████▌| 354099/371472 [6:01:08<1:24:56,  3.41it/s]
 95%|█████████▌| 354100/371472 [6:01:08<1:24:33,  3.42it/s]
                                                           
{'loss': 2.5911, 'learning_rate': 1.421114439009675e-07, 'epoch': 15.25}
+
 95%|█████████▌| 354100/371472 [6:01:08<1:24:33,  3.42it/s]
 95%|█████████▌| 354101/371472 [6:01:08<1:27:22,  3.31it/s]
 95%|█████████▌| 354102/371472 [6:01:09<1:23:20,  3.47it/s]
 95%|█████████▌| 354103/371472 [6:01:09<1:23:07,  3.48it/s]
 95%|█████████▌| 354104/371472 [6:01:09<1:21:25,  3.55it/s]
 95%|█████████▌| 354105/371472 [6:01:09<1:26:07,  3.36it/s]
 95%|█████████▌| 354106/371472 [6:01:10<1:24:57,  3.41it/s]
 95%|█████████▌| 354107/371472 [6:01:10<1:24:12,  3.44it/s]
 95%|█████████▌| 354108/371472 [6:01:10<1:21:29,  3.55it/s]
 95%|█████████▌| 354109/371472 [6:01:11<1:21:55,  3.53it/s]
 95%|█████████▌| 354110/371472 [6:01:11<1:22:28,  3.51it/s]
 95%|█████████▌| 354111/371472 [6:01:11<1:21:43,  3.54it/s]
 95%|█████████▌| 354112/371472 [6:01:11<1:19:58,  3.62it/s]
 95%|█████████▌| 354113/371472 [6:01:12<1:19:53,  3.62it/s]
 95%|█████████▌| 354114/371472 [6:01:12<1:18:57,  3.66it/s]
 95%|█████████▌| 354115/371472 [6:01:12<1:19:09,  3.65it/s]
 95%|█████████▌| 354116/371472 [6:01:12<1:21:16,  3.56it/s]
 95%|█████████▌| 354117/371472 [6:01:13<1:25:09,  3.40it/s]
 95%|█████████▌| 354118/371472 [6:01:13<1:23:40,  3.46it/s]
 95%|█████████▌| 354119/371472 [6:01:13<1:29:20,  3.24it/s]
 95%|█████████▌| 354120/371472 [6:01:14<1:31:12,  3.17it/s]
                                                           
{'loss': 2.5611, 'learning_rate': 1.4206296192548864e-07, 'epoch': 15.25}
+
 95%|█████████▌| 354120/371472 [6:01:14<1:31:12,  3.17it/s]
 95%|█████████▌| 354121/371472 [6:01:14<1:26:28,  3.34it/s]
 95%|█████████▌| 354122/371472 [6:01:14<1:22:49,  3.49it/s]
 95%|█████████▌| 354123/371472 [6:01:15<1:19:42,  3.63it/s]
 95%|█████████▌| 354124/371472 [6:01:15<1:23:09,  3.48it/s]
 95%|█████████▌| 354125/371472 [6:01:15<1:22:06,  3.52it/s]
 95%|█████████▌| 354126/371472 [6:01:15<1:26:17,  3.35it/s]
 95%|█████████▌| 354127/371472 [6:01:16<1:26:48,  3.33it/s]
 95%|█████████▌| 354128/371472 [6:01:16<1:25:04,  3.40it/s]
 95%|█████████▌| 354129/371472 [6:01:16<1:29:29,  3.23it/s]
 95%|█████████▌| 354130/371472 [6:01:17<1:25:32,  3.38it/s]
 95%|█████████▌| 354131/371472 [6:01:17<1:22:50,  3.49it/s]
 95%|█████████▌| 354132/371472 [6:01:17<1:22:56,  3.48it/s]
 95%|█████████▌| 354133/371472 [6:01:17<1:21:45,  3.53it/s]
 95%|█████████▌| 354134/371472 [6:01:18<1:18:48,  3.67it/s]
 95%|█████████▌| 354135/371472 [6:01:18<1:17:05,  3.75it/s]
 95%|█████████▌| 354136/371472 [6:01:18<1:18:31,  3.68it/s]
 95%|█████████▌| 354137/371472 [6:01:19<1:18:14,  3.69it/s]
 95%|█████████▌| 354138/371472 [6:01:19<1:20:37,  3.58it/s]
 95%|█████████▌| 354139/371472 [6:01:19<1:19:10,  3.65it/s]
 95%|█████████▌| 354140/371472 [6:01:19<1:21:52,  3.53it/s]
                                                           
{'loss': 2.5594, 'learning_rate': 1.4201447995000966e-07, 'epoch': 15.25}
+
 95%|█████████▌| 354140/371472 [6:01:19<1:21:52,  3.53it/s]
 95%|█████████▌| 354141/371472 [6:01:20<1:24:58,  3.40it/s]
 95%|█████████▌| 354142/371472 [6:01:20<1:22:47,  3.49it/s]
 95%|█████████▌| 354143/371472 [6:01:20<1:21:30,  3.54it/s]
 95%|█████████▌| 354144/371472 [6:01:21<1:19:14,  3.64it/s]
 95%|█████████▌| 354145/371472 [6:01:21<1:18:47,  3.67it/s]
 95%|█████████▌| 354146/371472 [6:01:21<1:18:44,  3.67it/s]
 95%|█████████▌| 354147/371472 [6:01:21<1:18:09,  3.69it/s]
 95%|█████████▌| 354148/371472 [6:01:22<1:16:06,  3.79it/s]
 95%|█████████▌| 354149/371472 [6:01:22<1:17:37,  3.72it/s]
 95%|█████████▌| 354150/371472 [6:01:22<1:19:27,  3.63it/s]
 95%|█████████▌| 354151/371472 [6:01:23<1:30:52,  3.18it/s]
 95%|█████████▌| 354152/371472 [6:01:23<1:30:01,  3.21it/s]
 95%|█████████▌| 354153/371472 [6:01:23<1:30:54,  3.18it/s]
 95%|█████████▌| 354154/371472 [6:01:23<1:25:39,  3.37it/s]
 95%|█████████▌| 354155/371472 [6:01:24<1:24:13,  3.43it/s]
 95%|█████████▌| 354156/371472 [6:01:24<1:22:02,  3.52it/s]
 95%|█████████▌| 354157/371472 [6:01:24<1:19:47,  3.62it/s]
 95%|█████████▌| 354158/371472 [6:01:24<1:17:02,  3.75it/s]
 95%|█████████▌| 354159/371472 [6:01:25<1:14:48,  3.86it/s]
 95%|█████████▌| 354160/371472 [6:01:25<1:16:13,  3.79it/s]
                                                           
{'loss': 2.5811, 'learning_rate': 1.4196599797453076e-07, 'epoch': 15.25}
+
 95%|█████████▌| 354160/371472 [6:01:25<1:16:13,  3.79it/s]
 95%|█████████▌| 354161/371472 [6:01:25<1:16:00,  3.80it/s]
 95%|█████████▌| 354162/371472 [6:01:26<1:20:50,  3.57it/s]
 95%|█████████▌| 354163/371472 [6:01:26<1:19:02,  3.65it/s]
 95%|█████████▌| 354164/371472 [6:01:26<1:18:52,  3.66it/s]
 95%|█████████▌| 354165/371472 [6:01:26<1:21:41,  3.53it/s]
 95%|█████████▌| 354166/371472 [6:01:27<1:25:12,  3.38it/s]
 95%|█████████▌| 354167/371472 [6:01:27<1:25:44,  3.36it/s]
 95%|█████████▌| 354168/371472 [6:01:27<1:23:41,  3.45it/s]
 95%|█████████▌| 354169/371472 [6:01:28<1:23:17,  3.46it/s]
 95%|█████████▌| 354170/371472 [6:01:28<1:28:21,  3.26it/s]
 95%|█████████▌| 354171/371472 [6:01:28<1:25:01,  3.39it/s]
 95%|█████████▌| 354172/371472 [6:01:28<1:23:35,  3.45it/s]
 95%|█████████▌| 354173/371472 [6:01:29<1:22:03,  3.51it/s]
 95%|█████████▌| 354174/371472 [6:01:29<1:22:58,  3.47it/s]
 95%|█████████▌| 354175/371472 [6:01:29<1:21:36,  3.53it/s]
 95%|█████████▌| 354176/371472 [6:01:30<1:20:28,  3.58it/s]
 95%|█████████▌| 354177/371472 [6:01:30<1:18:07,  3.69it/s]
 95%|█████████▌| 354178/371472 [6:01:30<1:16:22,  3.77it/s]
 95%|█████████▌| 354179/371472 [6:01:30<1:20:35,  3.58it/s]
 95%|█████████▌| 354180/371472 [6:01:31<1:21:16,  3.55it/s]
                                                           
{'loss': 2.7166, 'learning_rate': 1.419175159990519e-07, 'epoch': 15.26}
+
 95%|█████████▌| 354180/371472 [6:01:31<1:21:16,  3.55it/s]
 95%|█████████▌| 354181/371472 [6:01:31<1:23:13,  3.46it/s]
 95%|█████████▌| 354182/371472 [6:01:31<1:29:59,  3.20it/s]
 95%|█████████▌| 354183/371472 [6:01:32<1:26:40,  3.32it/s]
 95%|█████████▌| 354184/371472 [6:01:32<1:26:28,  3.33it/s]
 95%|█████████▌| 354185/371472 [6:01:32<1:23:37,  3.45it/s]
 95%|█████████▌| 354186/371472 [6:01:33<1:24:58,  3.39it/s]
 95%|█████████▌| 354187/371472 [6:01:33<1:30:05,  3.20it/s]
 95%|█████████▌| 354188/371472 [6:01:33<1:26:16,  3.34it/s]
 95%|█████████▌| 354189/371472 [6:01:33<1:23:55,  3.43it/s]
 95%|█████████▌| 354190/371472 [6:01:34<1:26:59,  3.31it/s]
 95%|█████████▌| 354191/371472 [6:01:34<1:24:27,  3.41it/s]
 95%|█████████▌| 354192/371472 [6:01:34<1:24:52,  3.39it/s]
 95%|█████████▌| 354193/371472 [6:01:35<1:27:58,  3.27it/s]
 95%|█████████▌| 354194/371472 [6:01:35<1:28:26,  3.26it/s]
 95%|█████████▌| 354195/371472 [6:01:35<1:25:03,  3.39it/s]
 95%|█████████▌| 354196/371472 [6:01:35<1:22:17,  3.50it/s]
 95%|█████████▌| 354197/371472 [6:01:36<1:22:30,  3.49it/s]
 95%|█████████▌| 354198/371472 [6:01:36<1:25:15,  3.38it/s]
 95%|█████████▌| 354199/371472 [6:01:36<1:22:34,  3.49it/s]
 95%|█████████▌| 354200/371472 [6:01:37<1:21:11,  3.55it/s]
                                                           
{'loss': 2.7085, 'learning_rate': 1.4186903402357296e-07, 'epoch': 15.26}
+
 95%|█████████▌| 354200/371472 [6:01:37<1:21:11,  3.55it/s]
 95%|█████████▌| 354201/371472 [6:01:37<1:19:02,  3.64it/s]
 95%|█████████▌| 354202/371472 [6:01:37<1:23:29,  3.45it/s]
 95%|█████████▌| 354203/371472 [6:01:37<1:22:48,  3.48it/s]
 95%|█████████▌| 354204/371472 [6:01:38<1:21:05,  3.55it/s]
 95%|█████████▌| 354205/371472 [6:01:38<1:18:00,  3.69it/s]
 95%|█████████▌| 354206/371472 [6:01:38<1:17:03,  3.73it/s]
 95%|█████████▌| 354207/371472 [6:01:39<1:18:44,  3.65it/s]
 95%|█████████▌| 354208/371472 [6:01:39<1:21:47,  3.52it/s]
 95%|█████████▌| 354209/371472 [6:01:39<1:22:57,  3.47it/s]
 95%|█████████▌| 354210/371472 [6:01:39<1:19:46,  3.61it/s]
 95%|█████████▌| 354211/371472 [6:01:40<1:17:47,  3.70it/s]
 95%|█████████▌| 354212/371472 [6:01:40<1:17:51,  3.69it/s]
 95%|█████████▌| 354213/371472 [6:01:40<1:17:33,  3.71it/s]
 95%|█████████▌| 354214/371472 [6:01:41<1:24:37,  3.40it/s]
 95%|█████████▌| 354215/371472 [6:01:41<1:22:59,  3.47it/s]
 95%|█████████▌| 354216/371472 [6:01:41<1:26:45,  3.31it/s]
 95%|█████████▌| 354217/371472 [6:01:41<1:24:01,  3.42it/s]
 95%|█████████▌| 354218/371472 [6:01:42<1:23:39,  3.44it/s]
 95%|█████████▌| 354219/371472 [6:01:42<1:21:15,  3.54it/s]
 95%|█████████▌| 354220/371472 [6:01:42<1:29:29,  3.21it/s]
                                                           
{'loss': 2.4786, 'learning_rate': 1.4182055204809413e-07, 'epoch': 15.26}
+
 95%|█████████▌| 354220/371472 [6:01:42<1:29:29,  3.21it/s]
 95%|█████████▌| 354221/371472 [6:01:43<1:25:45,  3.35it/s]
 95%|█████████▌| 354222/371472 [6:01:43<1:25:13,  3.37it/s]
 95%|█████████▌| 354223/371472 [6:01:43<1:21:08,  3.54it/s]
 95%|█████████▌| 354224/371472 [6:01:43<1:21:08,  3.54it/s]
 95%|█████████▌| 354225/371472 [6:01:44<1:19:34,  3.61it/s]
 95%|█████████▌| 354226/371472 [6:01:44<1:22:13,  3.50it/s]
 95%|█████████▌| 354227/371472 [6:01:44<1:20:06,  3.59it/s]
 95%|█████████▌| 354228/371472 [6:01:45<1:18:16,  3.67it/s]
 95%|█████████▌| 354229/371472 [6:01:45<1:15:20,  3.81it/s]
 95%|█████████▌| 354230/371472 [6:01:45<1:19:22,  3.62it/s]
 95%|█████████▌| 354231/371472 [6:01:45<1:18:42,  3.65it/s]
 95%|█████████▌| 354232/371472 [6:01:46<1:17:23,  3.71it/s]
 95%|█████████▌| 354233/371472 [6:01:46<1:23:18,  3.45it/s]
 95%|█████████▌| 354234/371472 [6:01:46<1:26:43,  3.31it/s]
 95%|█████████▌| 354235/371472 [6:01:47<1:22:23,  3.49it/s]
 95%|█████████▌| 354236/371472 [6:01:47<1:23:04,  3.46it/s]
 95%|█████████▌| 354237/371472 [6:01:47<1:23:40,  3.43it/s]
 95%|█████████▌| 354238/371472 [6:01:47<1:20:38,  3.56it/s]
 95%|█████████��| 354239/371472 [6:01:48<1:18:20,  3.67it/s]
 95%|█████████▌| 354240/371472 [6:01:48<1:19:00,  3.63it/s]
                                                           
{'loss': 2.6354, 'learning_rate': 1.4177207007261518e-07, 'epoch': 15.26}
+
 95%|█████████▌| 354240/371472 [6:01:48<1:19:00,  3.63it/s]
 95%|█████████▌| 354241/371472 [6:01:48<1:18:24,  3.66it/s]
 95%|█████████▌| 354242/371472 [6:01:48<1:18:49,  3.64it/s]
 95%|█████████▌| 354243/371472 [6:01:49<1:20:49,  3.55it/s]
 95%|█████████▌| 354244/371472 [6:01:49<1:19:36,  3.61it/s]
 95%|█████████▌| 354245/371472 [6:01:49<1:19:39,  3.60it/s]
 95%|█████████▌| 354246/371472 [6:01:50<1:30:03,  3.19it/s]
 95%|█████████▌| 354247/371472 [6:01:50<1:27:34,  3.28it/s]
 95%|█████████▌| 354248/371472 [6:01:50<1:24:20,  3.40it/s]
 95%|█████████▌| 354249/371472 [6:01:51<1:20:54,  3.55it/s]
 95%|█████████▌| 354250/371472 [6:01:51<1:37:31,  2.94it/s]
 95%|█████████▌| 354251/371472 [6:01:51<1:30:13,  3.18it/s]
 95%|█████████▌| 354252/371472 [6:01:52<1:28:28,  3.24it/s]
 95%|█████████▌| 354253/371472 [6:01:52<1:24:02,  3.41it/s]
 95%|█████████▌| 354254/371472 [6:01:52<1:21:54,  3.50it/s]
 95%|█████████▌| 354255/371472 [6:01:52<1:23:56,  3.42it/s]
 95%|█████████▌| 354256/371472 [6:01:53<1:25:40,  3.35it/s]
 95%|█████████▌| 354257/371472 [6:01:53<1:23:31,  3.44it/s]
 95%|█████████▌| 354258/371472 [6:01:53<1:27:22,  3.28it/s]
 95%|█████████▌| 354259/371472 [6:01:54<1:26:32,  3.32it/s]
 95%|█████████▌| 354260/371472 [6:01:54<1:24:54,  3.38it/s]
                                                           
{'loss': 2.5064, 'learning_rate': 1.4172358809713633e-07, 'epoch': 15.26}
+
 95%|█████████▌| 354260/371472 [6:01:54<1:24:54,  3.38it/s]
 95%|█████████▌| 354261/371472 [6:01:54<1:21:29,  3.52it/s]
 95%|█████████▌| 354262/371472 [6:01:54<1:21:44,  3.51it/s]
 95%|█████████▌| 354263/371472 [6:01:55<1:22:16,  3.49it/s]
 95%|█████████▌| 354264/371472 [6:01:55<1:33:55,  3.05it/s]
 95%|█████████▌| 354265/371472 [6:01:55<1:28:05,  3.26it/s]
 95%|█████████▌| 354266/371472 [6:01:56<1:24:24,  3.40it/s]
 95%|█████████▌| 354267/371472 [6:01:56<1:25:23,  3.36it/s]
 95%|█████████▌| 354268/371472 [6:01:56<1:24:44,  3.38it/s]
 95%|█████████▌| 354269/371472 [6:01:57<1:29:21,  3.21it/s]
 95%|█████████▌| 354270/371472 [6:01:57<1:26:14,  3.32it/s]
 95%|█████████▌| 354271/371472 [6:01:57<1:22:19,  3.48it/s]
 95%|█████████▌| 354272/371472 [6:01:57<1:22:52,  3.46it/s]
 95%|█████████▌| 354273/371472 [6:01:58<1:22:11,  3.49it/s]
 95%|█████████▌| 354274/371472 [6:01:58<1:19:43,  3.60it/s]
 95%|█████████▌| 354275/371472 [6:01:58<1:20:38,  3.55it/s]
 95%|█████████▌| 354276/371472 [6:01:59<1:20:19,  3.57it/s]
 95%|█████████▌| 354277/371472 [6:01:59<1:28:18,  3.25it/s]
 95%|█████████▌| 354278/371472 [6:01:59<1:27:43,  3.27it/s]
 95%|█████████▌| 354279/371472 [6:01:59<1:24:49,  3.38it/s]
 95%|█████████▌| 354280/371472 [6:02:00<1:20:32,  3.56it/s]
                                                           
{'loss': 2.5734, 'learning_rate': 1.416751061216574e-07, 'epoch': 15.26}
+
 95%|█████████▌| 354280/371472 [6:02:00<1:20:32,  3.56it/s]
 95%|█████████▌| 354281/371472 [6:02:00<1:21:36,  3.51it/s]
 95%|█████████▌| 354282/371472 [6:02:00<1:23:08,  3.45it/s]
 95%|█████████▌| 354283/371472 [6:02:01<1:27:02,  3.29it/s]
 95%|█████████▌| 354284/371472 [6:02:01<1:25:48,  3.34it/s]
 95%|█████████▌| 354285/371472 [6:02:01<1:23:33,  3.43it/s]
 95%|█████████▌| 354286/371472 [6:02:02<1:23:17,  3.44it/s]
 95%|█████████▌| 354287/371472 [6:02:02<1:21:15,  3.52it/s]
 95%|█████████▌| 354288/371472 [6:02:02<1:21:19,  3.52it/s]
 95%|█████████▌| 354289/371472 [6:02:02<1:28:32,  3.23it/s]
 95%|█████████▌| 354290/371472 [6:02:03<1:27:15,  3.28it/s]
 95%|█████████▌| 354291/371472 [6:02:03<1:27:49,  3.26it/s]
 95%|█████████▌| 354292/371472 [6:02:04<1:41:42,  2.82it/s]
 95%|█████████▌| 354293/371472 [6:02:04<1:39:18,  2.88it/s]
 95%|█████████▌| 354294/371472 [6:02:04<1:32:42,  3.09it/s]
 95%|███████���█▌| 354295/371472 [6:02:04<1:28:37,  3.23it/s]
 95%|█████████▌| 354296/371472 [6:02:05<1:24:15,  3.40it/s]
 95%|█████████▌| 354297/371472 [6:02:05<1:22:09,  3.48it/s]
 95%|█████████▌| 354298/371472 [6:02:05<1:22:41,  3.46it/s]
 95%|█████████▌| 354299/371472 [6:02:06<1:26:44,  3.30it/s]
 95%|█████████▌| 354300/371472 [6:02:06<1:22:21,  3.48it/s]
                                                           
{'loss': 2.5081, 'learning_rate': 1.4162662414617855e-07, 'epoch': 15.26}
+
 95%|█████████▌| 354300/371472 [6:02:06<1:22:21,  3.48it/s]
 95%|█████████▌| 354301/371472 [6:02:06<1:21:36,  3.51it/s]
 95%|█████████▌| 354302/371472 [6:02:06<1:19:08,  3.62it/s]
 95%|█████████▌| 354303/371472 [6:02:07<1:19:34,  3.60it/s]
 95%|█████████▌| 354304/371472 [6:02:07<1:22:23,  3.47it/s]
 95%|█████████▌| 354305/371472 [6:02:07<1:22:07,  3.48it/s]
 95%|█████████▌| 354306/371472 [6:02:07<1:22:08,  3.48it/s]
 95%|█████████▌| 354307/371472 [6:02:08<1:22:04,  3.49it/s]
 95%|█████████▌| 354308/371472 [6:02:08<1:21:19,  3.52it/s]
 95%|█████████▌| 354309/371472 [6:02:08<1:18:12,  3.66it/s]
 95%|█████████▌| 354310/371472 [6:02:09<1:16:54,  3.72it/s]
 95%|█████████▌| 354311/371472 [6:02:09<1:17:02,  3.71it/s]
 95%|█████████▌| 354312/371472 [6:02:09<1:20:40,  3.54it/s]
 95%|█████████▌| 354313/371472 [6:02:09<1:22:25,  3.47it/s]
 95%|█████████▌| 354314/371472 [6:02:10<1:22:45,  3.46it/s]
 95%|█████████▌| 354315/371472 [6:02:10<1:19:17,  3.61it/s]
 95%|█████████▌| 354316/371472 [6:02:10<1:18:55,  3.62it/s]
 95%|█████████▌| 354317/371472 [6:02:11<1:17:53,  3.67it/s]
 95%|█████████▌| 354318/371472 [6:02:11<1:16:32,  3.74it/s]
 95%|█████████▌| 354319/371472 [6:02:11<1:23:11,  3.44it/s]
 95%|█████████▌| 354320/371472 [6:02:11<1:24:04,  3.40it/s]
                                                           
{'loss': 2.604, 'learning_rate': 1.415781421706996e-07, 'epoch': 15.26}
+
 95%|█████████▌| 354320/371472 [6:02:11<1:24:04,  3.40it/s]
 95%|█████████▌| 354321/371472 [6:02:12<1:24:43,  3.37it/s]
 95%|█████████▌| 354322/371472 [6:02:12<1:22:24,  3.47it/s]
 95%|█████████▌| 354323/371472 [6:02:12<1:23:18,  3.43it/s]
 95%|█████████▌| 354324/371472 [6:02:13<1:20:32,  3.55it/s]
 95%|█████████▌| 354325/371472 [6:02:13<1:17:34,  3.68it/s]
 95%|█████████▌| 354326/371472 [6:02:13<1:16:51,  3.72it/s]
 95%|█████████▌| 354327/371472 [6:02:13<1:17:24,  3.69it/s]
 95%|█████████▌| 354328/371472 [6:02:14<1:18:00,  3.66it/s]
 95%|█████████▌| 354329/371472 [6:02:14<1:16:13,  3.75it/s]
 95%|█████████▌| 354330/371472 [6:02:14<1:17:55,  3.67it/s]
 95%|█████████▌| 354331/371472 [6:02:14<1:18:42,  3.63it/s]
 95%|█████████▌| 354332/371472 [6:02:15<1:17:22,  3.69it/s]
 95%|█████████▌| 354333/371472 [6:02:15<1:18:11,  3.65it/s]
 95%|█████████▌| 354334/371472 [6:02:15<1:17:16,  3.70it/s]
 95%|█████████▌| 354335/371472 [6:02:16<1:17:20,  3.69it/s]
 95%|█████████▌| 354336/371472 [6:02:16<1:14:23,  3.84it/s]
 95%|█████████▌| 354337/371472 [6:02:16<1:25:54,  3.32it/s]
 95%|█████████▌| 354338/371472 [6:02:16<1:21:30,  3.50it/s]
 95%|█████████▌| 354339/371472 [6:02:17<1:21:25,  3.51it/s]
 95%|█████████▌| 354340/371472 [6:02:17<1:19:17,  3.60it/s]
                                                           
{'loss': 2.4917, 'learning_rate': 1.4152966019522077e-07, 'epoch': 15.26}
+
 95%|█████████▌| 354340/371472 [6:02:17<1:19:17,  3.60it/s]
 95%|█████████▌| 354341/371472 [6:02:17<1:18:22,  3.64it/s]
 95%|█████████▌| 354342/371472 [6:02:17<1:18:03,  3.66it/s]
 95%|█████████▌| 354343/371472 [6:02:18<1:21:34,  3.50it/s]
 95%|█████████▌| 354344/371472 [6:02:18<1:17:48,  3.67it/s]
 95%|█████████▌| 354345/371472 [6:02:18<1:16:42,  3.72it/s]
 95%|█████████▌| 354346/371472 [6:02:19<1:23:54,  3.40it/s]
 95%|█████████▌| 354347/371472 [6:02:19<1:23:31,  3.42it/s]
 95%|█████████▌| 354348/371472 [6:02:19<1:20:15,  3.56it/s]
 95%|█████████▌| 354349/371472 [6:02:19<1:21:39,  3.49it/s]
 95%|█████████▌| 354350/371472 [6:02:20<1:27:19,  3.27it/s]
 95%|█████���███▌| 354351/371472 [6:02:20<1:35:13,  3.00it/s]
 95%|█████████▌| 354352/371472 [6:02:21<1:29:21,  3.19it/s]
 95%|█████████▌| 354353/371472 [6:02:21<1:24:26,  3.38it/s]
 95%|█████████▌| 354354/371472 [6:02:21<1:22:00,  3.48it/s]
 95%|█████████▌| 354355/371472 [6:02:21<1:20:48,  3.53it/s]
 95%|█████████▌| 354356/371472 [6:02:22<1:18:23,  3.64it/s]
 95%|█████████▌| 354357/371472 [6:02:22<1:20:39,  3.54it/s]
 95%|█████████▌| 354358/371472 [6:02:22<1:17:16,  3.69it/s]
 95%|█████████▌| 354359/371472 [6:02:22<1:18:27,  3.64it/s]
 95%|█████████▌| 354360/371472 [6:02:23<1:17:15,  3.69it/s]
                                                           
{'loss': 2.6963, 'learning_rate': 1.4148117821974182e-07, 'epoch': 15.26}
+
 95%|█████████▌| 354360/371472 [6:02:23<1:17:15,  3.69it/s]
 95%|█████████▌| 354361/371472 [6:02:23<1:17:15,  3.69it/s]
 95%|█████████▌| 354362/371472 [6:02:23<1:16:29,  3.73it/s]
 95%|█████████▌| 354363/371472 [6:02:23<1:15:24,  3.78it/s]
 95%|█████████▌| 354364/371472 [6:02:24<1:22:49,  3.44it/s]
 95%|█████████▌| 354365/371472 [6:02:24<1:21:21,  3.50it/s]
 95%|█████████▌| 354366/371472 [6:02:24<1:18:34,  3.63it/s]
 95%|█████████▌| 354367/371472 [6:02:25<1:21:29,  3.50it/s]
 95%|█████████▌| 354368/371472 [6:02:25<1:25:01,  3.35it/s]
 95%|█████████▌| 354369/371472 [6:02:25<1:22:05,  3.47it/s]
 95%|█████████▌| 354370/371472 [6:02:26<1:22:28,  3.46it/s]
 95%|█████████▌| 354371/371472 [6:02:26<1:27:02,  3.27it/s]
 95%|█████████▌| 354372/371472 [6:02:26<1:23:44,  3.40it/s]
 95%|█████████▌| 354373/371472 [6:02:26<1:21:21,  3.50it/s]
 95%|█████████▌| 354374/371472 [6:02:27<1:22:13,  3.47it/s]
 95%|█████████▌| 354375/371472 [6:02:27<1:20:07,  3.56it/s]
 95%|█████████▌| 354376/371472 [6:02:27<1:22:00,  3.47it/s]
 95%|█████████▌| 354377/371472 [6:02:28<1:23:13,  3.42it/s]
 95%|█████████▌| 354378/371472 [6:02:28<1:24:43,  3.36it/s]
 95%|█████████▌| 354379/371472 [6:02:28<1:28:51,  3.21it/s]
 95%|█████████▌| 354380/371472 [6:02:28<1:26:52,  3.28it/s]
                                                           
{'loss': 2.4483, 'learning_rate': 1.4143269624426297e-07, 'epoch': 15.26}
+
 95%|█████████▌| 354380/371472 [6:02:28<1:26:52,  3.28it/s]
 95%|█████████▌| 354381/371472 [6:02:29<1:23:35,  3.41it/s]
 95%|█████████▌| 354382/371472 [6:02:29<1:18:28,  3.63it/s]
 95%|█████████▌| 354383/371472 [6:02:29<1:16:44,  3.71it/s]
 95%|█████████▌| 354384/371472 [6:02:30<1:15:36,  3.77it/s]
 95%|█████████▌| 354385/371472 [6:02:30<1:18:12,  3.64it/s]
 95%|█████████▌| 354386/371472 [6:02:30<1:18:20,  3.64it/s]
 95%|█████████▌| 354387/371472 [6:02:30<1:17:35,  3.67it/s]
 95%|█████████▌| 354388/371472 [6:02:31<1:16:50,  3.71it/s]
 95%|█████████▌| 354389/371472 [6:02:31<1:17:31,  3.67it/s]
 95%|█████████▌| 354390/371472 [6:02:31<1:15:32,  3.77it/s]
 95%|█████████▌| 354391/371472 [6:02:31<1:18:09,  3.64it/s]
 95%|█████████▌| 354392/371472 [6:02:32<1:15:13,  3.78it/s]
 95%|█████████▌| 354393/371472 [6:02:32<1:15:57,  3.75it/s]
 95%|█████████▌| 354394/371472 [6:02:32<1:13:21,  3.88it/s]
 95%|█████████▌| 354395/371472 [6:02:32<1:13:39,  3.86it/s]
 95%|█████████▌| 354396/371472 [6:02:33<1:14:34,  3.82it/s]
 95%|█████████▌| 354397/371472 [6:02:33<1:18:28,  3.63it/s]
 95%|█████████▌| 354398/371472 [6:02:33<1:16:34,  3.72it/s]
 95%|█████████▌| 354399/371472 [6:02:34<1:19:08,  3.60it/s]
 95%|█████████▌| 354400/371472 [6:02:34<1:18:11,  3.64it/s]
                                                           
{'loss': 2.5005, 'learning_rate': 1.4138421426878404e-07, 'epoch': 15.26}
+
 95%|█████████▌| 354400/371472 [6:02:34<1:18:11,  3.64it/s]
 95%|█████████▌| 354401/371472 [6:02:34<1:20:46,  3.52it/s]
 95%|█████████▌| 354402/371472 [6:02:34<1:20:57,  3.51it/s]
 95%|█████████▌| 354403/371472 [6:02:35<1:23:42,  3.40it/s]
 95%|█████████▌| 354404/371472 [6:02:35<1:23:15,  3.42it/s]
 95%|█████████▌| 354405/371472 [6:02:35<1:21:45,  3.48it/s]
 95%|█████████▌| 354406/371472 [6:02:36<1:23:52,  3.39it/s]
 95%|██��██████▌| 354407/371472 [6:02:36<1:23:11,  3.42it/s]
 95%|█████████▌| 354408/371472 [6:02:36<1:22:52,  3.43it/s]
 95%|█████████▌| 354409/371472 [6:02:37<1:30:37,  3.14it/s]
 95%|█████████▌| 354410/371472 [6:02:37<1:33:37,  3.04it/s]
 95%|█████████▌| 354411/371472 [6:02:37<1:29:51,  3.16it/s]
 95%|█████████▌| 354412/371472 [6:02:37<1:26:04,  3.30it/s]
 95%|█████████▌| 354413/371472 [6:02:38<1:25:39,  3.32it/s]
 95%|█████████▌| 354414/371472 [6:02:38<1:21:09,  3.50it/s]
 95%|█████████▌| 354415/371472 [6:02:38<1:22:33,  3.44it/s]
 95%|█████████▌| 354416/371472 [6:02:39<1:21:39,  3.48it/s]
 95%|█████████▌| 354417/371472 [6:02:39<1:21:52,  3.47it/s]
 95%|█████████▌| 354418/371472 [6:02:39<1:19:27,  3.58it/s]
 95%|█████████▌| 354419/371472 [6:02:39<1:18:39,  3.61it/s]
 95%|█████████▌| 354420/371472 [6:02:40<1:23:54,  3.39it/s]
                                                           
{'loss': 2.6328, 'learning_rate': 1.413357322933052e-07, 'epoch': 15.27}
+
 95%|█████████▌| 354420/371472 [6:02:40<1:23:54,  3.39it/s]
 95%|█████████▌| 354421/371472 [6:02:40<1:23:13,  3.41it/s]
 95%|█████████▌| 354422/371472 [6:02:40<1:19:41,  3.57it/s]
 95%|█████████▌| 354423/371472 [6:02:41<1:17:49,  3.65it/s]
 95%|█████████▌| 354424/371472 [6:02:41<1:17:17,  3.68it/s]
 95%|█████████▌| 354425/371472 [6:02:41<1:19:02,  3.59it/s]
 95%|█████████▌| 354426/371472 [6:02:41<1:18:33,  3.62it/s]
 95%|█████████▌| 354427/371472 [6:02:42<1:16:10,  3.73it/s]
 95%|█████████▌| 354428/371472 [6:02:42<1:15:51,  3.74it/s]
 95%|█████████▌| 354429/371472 [6:02:42<1:16:48,  3.70it/s]
 95%|█████████▌| 354430/371472 [6:02:43<1:20:49,  3.51it/s]
 95%|█████████▌| 354431/371472 [6:02:43<1:21:41,  3.48it/s]
 95%|█████████▌| 354432/371472 [6:02:43<1:20:49,  3.51it/s]
 95%|█████████▌| 354433/371472 [6:02:43<1:20:25,  3.53it/s]
 95%|█████████▌| 354434/371472 [6:02:44<1:19:51,  3.56it/s]
 95%|█████████▌| 354435/371472 [6:02:44<1:25:17,  3.33it/s]
 95%|█████████▌| 354436/371472 [6:02:44<1:20:53,  3.51it/s]
 95%|█████████▌| 354437/371472 [6:02:45<1:20:23,  3.53it/s]
 95%|█████████▌| 354438/371472 [6:02:45<1:19:36,  3.57it/s]
 95%|█████████▌| 354439/371472 [6:02:45<1:22:57,  3.42it/s]
 95%|█████████▌| 354440/371472 [6:02:45<1:20:32,  3.52it/s]
                                                           
{'loss': 2.5912, 'learning_rate': 1.4128725031782626e-07, 'epoch': 15.27}
+
 95%|█████████▌| 354440/371472 [6:02:45<1:20:32,  3.52it/s]
 95%|█████████▌| 354441/371472 [6:02:46<1:20:58,  3.51it/s]
 95%|█████████▌| 354442/371472 [6:02:46<1:18:33,  3.61it/s]
 95%|█████████▌| 354443/371472 [6:02:46<1:17:40,  3.65it/s]
 95%|█████████▌| 354444/371472 [6:02:46<1:18:01,  3.64it/s]
 95%|█████████▌| 354445/371472 [6:02:47<1:18:34,  3.61it/s]
 95%|█████████▌| 354446/371472 [6:02:47<1:18:19,  3.62it/s]
 95%|█████████▌| 354447/371472 [6:02:47<1:19:13,  3.58it/s]
 95%|█████████▌| 354448/371472 [6:02:48<1:17:03,  3.68it/s]
 95%|█████████▌| 354449/371472 [6:02:48<1:25:51,  3.30it/s]
 95%|█████████▌| 354450/371472 [6:02:48<1:21:30,  3.48it/s]
 95%|█████████▌| 354451/371472 [6:02:48<1:20:44,  3.51it/s]
 95%|█████████▌| 354452/371472 [6:02:49<1:18:51,  3.60it/s]
 95%|█████████▌| 354453/371472 [6:02:49<1:21:53,  3.46it/s]
 95%|█████████▌| 354454/371472 [6:02:49<1:19:37,  3.56it/s]
 95%|█████████▌| 354455/371472 [6:02:50<1:21:12,  3.49it/s]
 95%|█████████▌| 354456/371472 [6:02:50<1:19:47,  3.55it/s]
 95%|█████████▌| 354457/371472 [6:02:50<1:24:35,  3.35it/s]
 95%|█████████▌| 354458/371472 [6:02:51<1:25:06,  3.33it/s]
 95%|█████████▌| 354459/371472 [6:02:51<1:25:17,  3.32it/s]
 95%|█████████▌| 354460/371472 [6:02:51<1:21:59,  3.46it/s]
                                                           
{'loss': 2.5157, 'learning_rate': 1.4123876834234741e-07, 'epoch': 15.27}
+
 95%|█████████▌| 354460/371472 [6:02:51<1:21:59,  3.46it/s]
 95%|█████████▌| 354461/371472 [6:02:51<1:19:57,  3.55it/s]
 95%|█████████▌| 354462/371472 [6:02:52<1:16:50,  3.69it/s]
 95%|���████████▌| 354463/371472 [6:02:52<1:15:24,  3.76it/s]
 95%|█████████▌| 354464/371472 [6:02:52<1:13:03,  3.88it/s]
 95%|█████████▌| 354465/371472 [6:02:52<1:16:01,  3.73it/s]
 95%|█████████▌| 354466/371472 [6:02:53<1:17:03,  3.68it/s]
 95%|█████████▌| 354467/371472 [6:02:53<1:20:42,  3.51it/s]
 95%|█████████▌| 354468/371472 [6:02:53<1:25:47,  3.30it/s]
 95%|█████████▌| 354469/371472 [6:02:54<1:22:53,  3.42it/s]
 95%|█████████▌| 354470/371472 [6:02:54<1:21:14,  3.49it/s]
 95%|█████████▌| 354471/371472 [6:02:54<1:24:33,  3.35it/s]
 95%|█████████▌| 354472/371472 [6:02:55<1:26:09,  3.29it/s]
 95%|█████████▌| 354473/371472 [6:02:55<1:30:56,  3.12it/s]
 95%|█████████▌| 354474/371472 [6:02:55<1:29:36,  3.16it/s]
 95%|█████████▌| 354475/371472 [6:02:55<1:25:47,  3.30it/s]
 95%|█████████▌| 354476/371472 [6:02:56<1:23:15,  3.40it/s]
 95%|█████████▌| 354477/371472 [6:02:56<1:27:37,  3.23it/s]
 95%|█████████▌| 354478/371472 [6:02:56<1:23:28,  3.39it/s]
 95%|█████████▌| 354479/371472 [6:02:57<1:22:51,  3.42it/s]
 95%|█████████▌| 354480/371472 [6:02:57<1:20:20,  3.52it/s]
                                                           
{'loss': 2.5835, 'learning_rate': 1.411902863668685e-07, 'epoch': 15.27}
+
 95%|█████████▌| 354480/371472 [6:02:57<1:20:20,  3.52it/s]
 95%|█████████▌| 354481/371472 [6:02:57<1:19:25,  3.57it/s]
 95%|█████████▌| 354482/371472 [6:02:57<1:17:59,  3.63it/s]
 95%|█████████▌| 354483/371472 [6:02:58<1:16:29,  3.70it/s]
 95%|█████████▌| 354484/371472 [6:02:58<1:15:21,  3.76it/s]
 95%|█████████▌| 354485/371472 [6:02:58<1:18:16,  3.62it/s]
 95%|█████████▌| 354486/371472 [6:02:58<1:18:08,  3.62it/s]
 95%|█████████▌| 354487/371472 [6:02:59<1:19:11,  3.57it/s]
 95%|█████████▌| 354488/371472 [6:02:59<1:17:14,  3.66it/s]
 95%|█████████▌| 354489/371472 [6:02:59<1:15:57,  3.73it/s]
 95%|█████████▌| 354490/371472 [6:03:00<1:19:50,  3.54it/s]
 95%|█████████▌| 354491/371472 [6:03:00<1:23:29,  3.39it/s]
 95%|█████████▌| 354492/371472 [6:03:00<1:20:14,  3.53it/s]
 95%|█████████▌| 354493/371472 [6:03:01<1:23:05,  3.41it/s]
 95%|█████████▌| 354494/371472 [6:03:01<1:19:44,  3.55it/s]
 95%|█████████▌| 354495/371472 [6:03:01<1:32:18,  3.07it/s]
 95%|█████████▌| 354496/371472 [6:03:02<1:30:11,  3.14it/s]
 95%|█████████▌| 354497/371472 [6:03:02<1:25:12,  3.32it/s]
 95%|█████████▌| 354498/371472 [6:03:02<1:25:57,  3.29it/s]
 95%|█████████▌| 354499/371472 [6:03:02<1:22:38,  3.42it/s]
 95%|█████████▌| 354500/371472 [6:03:03<1:29:15,  3.17it/s]
                                                           
{'loss': 2.5653, 'learning_rate': 1.4114180439138964e-07, 'epoch': 15.27}
+
 95%|█████████▌| 354500/371472 [6:03:03<1:29:15,  3.17it/s]
 95%|█████████▌| 354501/371472 [6:03:03<1:24:51,  3.33it/s]
 95%|█████████▌| 354502/371472 [6:03:03<1:20:05,  3.53it/s]
 95%|█████████▌| 354503/371472 [6:03:04<1:24:56,  3.33it/s]
 95%|█████████▌| 354504/371472 [6:03:04<1:27:48,  3.22it/s]
 95%|█████████▌| 354505/371472 [6:03:04<1:23:53,  3.37it/s]
 95%|█████████▌| 354506/371472 [6:03:04<1:21:52,  3.45it/s]
 95%|█████████▌| 354507/371472 [6:03:05<1:19:15,  3.57it/s]
 95%|█████████▌| 354508/371472 [6:03:05<1:18:38,  3.60it/s]
 95%|█████████▌| 354509/371472 [6:03:05<1:18:13,  3.61it/s]
 95%|█████████▌| 354510/371472 [6:03:06<1:18:08,  3.62it/s]
 95%|█████████▌| 354511/371472 [6:03:06<1:18:19,  3.61it/s]
 95%|█████████▌| 354512/371472 [6:03:06<1:18:21,  3.61it/s]
 95%|█████████▌| 354513/371472 [6:03:06<1:18:04,  3.62it/s]
 95%|█████████▌| 354514/371472 [6:03:07<1:21:57,  3.45it/s]
 95%|█████████▌| 354515/371472 [6:03:07<1:26:13,  3.28it/s]
 95%|█████████▌| 354516/371472 [6:03:07<1:21:40,  3.46it/s]
 95%|█████████▌| 354517/371472 [6:03:08<1:28:57,  3.18it/s]
 95%|█████████▌| 354518/371472 [6:03:08<1:25:02,  3.32it/s]
 95%|█████████▌| 354519/371472 [6:03:08<1:24:48,  3.33it/s]
 95%|█████████▌| 354520/371472 [6:03:08<1:24:04,  3.36it/s]
                                                           
{'loss': 2.6409, 'learning_rate': 1.4109332241591068e-07, 'epoch': 15.27}
+
 95%|█████████▌| 354520/371472 [6:03:08<1:24:04,  3.36it/s]
 95%|█████████▌| 354521/371472 [6:03:09<1:25:21,  3.31it/s]
 95%|█████████▌| 354522/371472 [6:03:09<1:26:51,  3.25it/s]
 95%|█████████▌| 354523/371472 [6:03:09<1:22:02,  3.44it/s]
 95%|█████████▌| 354524/371472 [6:03:10<1:26:19,  3.27it/s]
 95%|█████████▌| 354525/371472 [6:03:10<1:29:56,  3.14it/s]
 95%|█████████▌| 354526/371472 [6:03:10<1:29:06,  3.17it/s]
 95%|█████████▌| 354527/371472 [6:03:11<1:29:44,  3.15it/s]
 95%|█████████▌| 354528/371472 [6:03:11<1:31:33,  3.08it/s]
 95%|█████████▌| 354529/371472 [6:03:11<1:28:42,  3.18it/s]
 95%|█████████▌| 354530/371472 [6:03:12<1:27:43,  3.22it/s]
 95%|█████████▌| 354531/371472 [6:03:12<1:24:29,  3.34it/s]
 95%|█████████▌| 354532/371472 [6:03:12<1:23:58,  3.36it/s]
 95%|█████████▌| 354533/371472 [6:03:12<1:21:31,  3.46it/s]
 95%|█████████▌| 354534/371472 [6:03:13<1:19:21,  3.56it/s]
 95%|█████████▌| 354535/371472 [6:03:13<1:18:09,  3.61it/s]
 95%|█████████▌| 354536/371472 [6:03:13<1:16:57,  3.67it/s]
 95%|█████████▌| 354537/371472 [6:03:14<1:20:57,  3.49it/s]
 95%|█████████▌| 354538/371472 [6:03:14<1:17:46,  3.63it/s]
 95%|█████████▌| 354539/371472 [6:03:14<1:16:16,  3.70it/s]
 95%|█████████▌| 354540/371472 [6:03:14<1:16:41,  3.68it/s]
                                                           
{'loss': 2.3744, 'learning_rate': 1.4104484044043186e-07, 'epoch': 15.27}
+
 95%|█████████▌| 354540/371472 [6:03:14<1:16:41,  3.68it/s]
 95%|█████████▌| 354541/371472 [6:03:15<1:18:16,  3.61it/s]
 95%|█████████▌| 354542/371472 [6:03:15<1:15:47,  3.72it/s]
 95%|█████████▌| 354543/371472 [6:03:15<1:16:38,  3.68it/s]
 95%|█████████▌| 354544/371472 [6:03:15<1:16:30,  3.69it/s]
 95%|█████████▌| 354545/371472 [6:03:16<1:18:44,  3.58it/s]
 95%|█████████▌| 354546/371472 [6:03:16<1:23:39,  3.37it/s]
 95%|█████████▌| 354547/371472 [6:03:16<1:21:04,  3.48it/s]
 95%|█████████▌| 354548/371472 [6:03:17<1:23:44,  3.37it/s]
 95%|█████████▌| 354549/371472 [6:03:17<1:21:29,  3.46it/s]
 95%|█████████▌| 354550/371472 [6:03:17<1:25:56,  3.28it/s]
 95%|█████████▌| 354551/371472 [6:03:18<1:21:55,  3.44it/s]
 95%|█████████▌| 354552/371472 [6:03:18<1:21:20,  3.47it/s]
 95%|█████████▌| 354553/371472 [6:03:18<1:31:00,  3.10it/s]
 95%|█████████▌| 354554/371472 [6:03:19<1:29:38,  3.15it/s]
 95%|█████████▌| 354555/371472 [6:03:19<1:24:36,  3.33it/s]
 95%|█████████▌| 354556/371472 [6:03:19<1:20:16,  3.51it/s]
 95%|█████████▌| 354557/371472 [6:03:19<1:24:01,  3.36it/s]
 95%|█████████▌| 354558/371472 [6:03:20<1:19:56,  3.53it/s]
 95%|█████████▌| 354559/371472 [6:03:20<1:18:06,  3.61it/s]
 95%|█████████▌| 354560/371472 [6:03:20<1:17:31,  3.64it/s]
                                                           
{'loss': 2.6112, 'learning_rate': 1.409963584649529e-07, 'epoch': 15.27}
+
 95%|█████████▌| 354560/371472 [6:03:20<1:17:31,  3.64it/s]
 95%|█████████▌| 354561/371472 [6:03:20<1:18:43,  3.58it/s]
 95%|█████████▌| 354562/371472 [6:03:21<1:16:55,  3.66it/s]
 95%|█████████▌| 354563/371472 [6:03:21<1:22:08,  3.43it/s]
 95%|█████████▌| 354564/371472 [6:03:21<1:17:53,  3.62it/s]
 95%|█████████▌| 354565/371472 [6:03:22<1:18:23,  3.59it/s]
 95%|█████████▌| 354566/371472 [6:03:22<1:18:37,  3.58it/s]
 95%|█████████▌| 354567/371472 [6:03:22<1:17:54,  3.62it/s]
 95%|█████████▌| 354568/371472 [6:03:22<1:21:44,  3.45it/s]
 95%|█████████▌| 354569/371472 [6:03:23<1:19:16,  3.55it/s]
 95%|█████████▌| 354570/371472 [6:03:23<1:18:49,  3.57it/s]
 95%|█████████▌| 354571/371472 [6:03:23<1:17:42,  3.62it/s]
 95%|█████████▌| 354572/371472 [6:03:23<1:16:36,  3.68it/s]
 95%|█████████▌| 354573/371472 [6:03:24<1:17:18,  3.64it/s]
 95%|█████████▌| 354574/371472 [6:03:24<1:14:22,  3.79it/s]
 95%|█████████▌| 354575/371472 [6:03:24<1:14:23,  3.79it/s]
 95%|█████████▌| 354576/371472 [6:03:25<1:15:34,  3.73it/s]
 95%|█████████▌| 354577/371472 [6:03:25<1:25:00,  3.31it/s]
 95%|█████████▌| 354578/371472 [6:03:25<1:26:04,  3.27it/s]
 95%|█████████▌| 354579/371472 [6:03:26<1:23:01,  3.39it/s]
 95%|█████████▌| 354580/371472 [6:03:26<1:23:53,  3.36it/s]
                                                           
{'loss': 2.6178, 'learning_rate': 1.4094787648947406e-07, 'epoch': 15.27}
+
 95%|█████████▌| 354580/371472 [6:03:26<1:23:53,  3.36it/s]
 95%|█████████▌| 354581/371472 [6:03:26<1:23:41,  3.36it/s]
 95%|█████████▌| 354582/371472 [6:03:26<1:21:27,  3.46it/s]
 95%|█████████▌| 354583/371472 [6:03:27<1:23:53,  3.36it/s]
 95%|█████████▌| 354584/371472 [6:03:27<1:22:50,  3.40it/s]
 95%|█████████▌| 354585/371472 [6:03:27<1:36:02,  2.93it/s]
 95%|█████████▌| 354586/371472 [6:03:28<1:28:16,  3.19it/s]
 95%|█████████▌| 354587/371472 [6:03:28<1:24:55,  3.31it/s]
 95%|█████████▌| 354588/371472 [6:03:28<1:23:47,  3.36it/s]
 95%|█████████▌| 354589/371472 [6:03:29<1:24:34,  3.33it/s]
 95%|█████████▌| 354590/371472 [6:03:29<1:20:18,  3.50it/s]
 95%|█████████▌| 354591/371472 [6:03:29<1:20:00,  3.52it/s]
 95%|█████████▌| 354592/371472 [6:03:29<1:20:11,  3.51it/s]
 95%|█████████▌| 354593/371472 [6:03:30<1:19:05,  3.56it/s]
 95%|█████████▌| 354594/371472 [6:03:30<1:16:31,  3.68it/s]
 95%|█████████▌| 354595/371472 [6:03:30<1:20:42,  3.49it/s]
 95%|█████████▌| 354596/371472 [6:03:31<1:20:17,  3.50it/s]
 95%|█████████▌| 354597/371472 [6:03:31<1:20:11,  3.51it/s]
 95%|█████████▌| 354598/371472 [6:03:31<1:20:39,  3.49it/s]
 95%|█████████▌| 354599/371472 [6:03:31<1:24:10,  3.34it/s]
 95%|█████████▌| 354600/371472 [6:03:32<1:21:49,  3.44it/s]
                                                           
{'loss': 2.5548, 'learning_rate': 1.4089939451399513e-07, 'epoch': 15.27}
+
 95%|█████████▌| 354600/371472 [6:03:32<1:21:49,  3.44it/s]
 95%|█████████▌| 354601/371472 [6:03:32<1:19:40,  3.53it/s]
 95%|█████████▌| 354602/371472 [6:03:32<1:24:04,  3.34it/s]
 95%|█████████▌| 354603/371472 [6:03:33<1:22:37,  3.40it/s]
 95%|█████████▌| 354604/371472 [6:03:33<1:21:37,  3.44it/s]
 95%|█████████▌| 354605/371472 [6:03:33<1:21:04,  3.47it/s]
 95%|█████████▌| 354606/371472 [6:03:33<1:19:44,  3.52it/s]
 95%|█████████▌| 354607/371472 [6:03:34<1:18:58,  3.56it/s]
 95%|█████████▌| 354608/371472 [6:03:34<1:16:33,  3.67it/s]
 95%|█████████▌| 354609/371472 [6:03:34<1:22:35,  3.40it/s]
 95%|█████████▌| 354610/371472 [6:03:35<1:19:31,  3.53it/s]
 95%|█████████▌| 354611/371472 [6:03:35<1:16:59,  3.65it/s]
 95%|█████████▌| 354612/371472 [6:03:35<1:19:47,  3.52it/s]
 95%|█████████▌| 354613/371472 [6:03:35<1:18:30,  3.58it/s]
 95%|█████████▌| 354614/371472 [6:03:36<1:18:41,  3.57it/s]
 95%|█████████▌| 354615/371472 [6:03:36<1:21:05,  3.46it/s]
 95%|█████████▌| 354616/371472 [6:03:36<1:21:13,  3.46it/s]
 95%|█████████▌| 354617/371472 [6:03:37<1:32:51,  3.03it/s]
 95%|█████████▌| 354618/371472 [6:03:37<1:29:43,  3.13it/s]
 95%|█████████▌| 354619/371472 [6:03:37<1:27:16,  3.22it/s]
 95%|█████████▌| 354620/371472 [6:03:38<1:25:55,  3.27it/s]
                                                           
{'loss': 2.5435, 'learning_rate': 1.4085091253851628e-07, 'epoch': 15.27}
+
 95%|█████████▌| 354620/371472 [6:03:38<1:25:55,  3.27it/s]
 95%|█████████▌| 354621/371472 [6:03:38<1:26:38,  3.24it/s]
 95%|█████████▌| 354622/371472 [6:03:38<1:23:26,  3.37it/s]
 95%|█████████▌| 354623/371472 [6:03:38<1:22:21,  3.41it/s]
 95%|█████████▌| 354624/371472 [6:03:39<1:19:26,  3.53it/s]
 95%|█████████▌| 354625/371472 [6:03:39<1:17:33,  3.62it/s]
 95%|█████████▌| 354626/371472 [6:03:39<1:16:29,  3.67it/s]
 95%|█████████▌| 354627/371472 [6:03:39<1:16:40,  3.66it/s]
 95%|█████████▌| 354628/371472 [6:03:40<1:14:16,  3.78it/s]
 95%|█████████▌| 354629/371472 [6:03:40<1:14:19,  3.78it/s]
 95%|█████████▌| 354630/371472 [6:03:40<1:15:12,  3.73it/s]
 95%|█████████▌| 354631/371472 [6:03:41<1:19:06,  3.55it/s]
 95%|█████████▌| 354632/371472 [6:03:41<1:16:49,  3.65it/s]
 95%|█████████▌| 354633/371472 [6:03:41<1:17:46,  3.61it/s]
 95%|█████████▌| 354634/371472 [6:03:41<1:19:36,  3.53it/s]
 95%|█████████▌| 354635/371472 [6:03:42<1:31:29,  3.07it/s]
 95%|█████████▌| 354636/371472 [6:03:42<1:27:41,  3.20it/s]
 95%|█████████▌| 354637/371472 [6:03:42<1:28:41,  3.16it/s]
 95%|█████████▌| 354638/371472 [6:03:43<1:26:31,  3.24it/s]
 95%|█████████▌| 354639/371472 [6:03:43<1:21:31,  3.44it/s]
 95%|█████████▌| 354640/371472 [6:03:43<1:22:11,  3.41it/s]
                                                           
{'loss': 2.6694, 'learning_rate': 1.4080243056303732e-07, 'epoch': 15.28}
+
 95%|█████████▌| 354640/371472 [6:03:43<1:22:11,  3.41it/s]
 95%|█████████▌| 354641/371472 [6:03:44<1:25:00,  3.30it/s]
 95%|█████████▌| 354642/371472 [6:03:44<1:25:05,  3.30it/s]
 95%|█████████▌| 354643/371472 [6:03:44<1:22:50,  3.39it/s]
 95%|█████████▌| 354644/371472 [6:03:44<1:19:13,  3.54it/s]
 95%|█████████▌| 354645/371472 [6:03:45<1:18:20,  3.58it/s]
 95%|█████████▌| 354646/371472 [6:03:45<1:18:35,  3.57it/s]
 95%|█████████▌| 354647/371472 [6:03:45<1:18:05,  3.59it/s]
 95%|█████████▌| 354648/371472 [6:03:46<1:18:07,  3.59it/s]
 95%|█████████▌| 354649/371472 [6:03:46<1:20:05,  3.50it/s]
 95%|█████████▌| 354650/371472 [6:03:46<1:19:08,  3.54it/s]
 95%|█████████▌| 354651/371472 [6:03:46<1:24:02,  3.34it/s]
 95%|█████████▌| 354652/371472 [6:03:47<1:24:01,  3.34it/s]
 95%|█████████▌| 354653/371472 [6:03:47<1:26:40,  3.23it/s]
 95%|█████████▌| 354654/371472 [6:03:47<1:22:13,  3.41it/s]
 95%|█████████▌| 354655/371472 [6:03:48<1:27:29,  3.20it/s]
 95%|█████████▌| 354656/371472 [6:03:48<1:24:00,  3.34it/s]
 95%|█████████▌| 354657/371472 [6:03:48<1:24:29,  3.32it/s]
 95%|█████████▌| 354658/371472 [6:03:49<1:20:04,  3.50it/s]
 95%|█████████▌| 354659/371472 [6:03:49<1:19:24,  3.53it/s]
 95%|█████████▌| 354660/371472 [6:03:49<1:21:16,  3.45it/s]
                                                           
{'loss': 2.3769, 'learning_rate': 1.407539485875585e-07, 'epoch': 15.28}
+
 95%|█████████▌| 354660/371472 [6:03:49<1:21:16,  3.45it/s]
 95%|█████████▌| 354661/371472 [6:03:49<1:23:18,  3.36it/s]
 95%|█████████▌| 354662/371472 [6:03:50<1:22:49,  3.38it/s]
 95%|█████████▌| 354663/371472 [6:03:50<1:18:55,  3.55it/s]
 95%|█████████▌| 354664/371472 [6:03:50<1:19:45,  3.51it/s]
 95%|█████████▌| 354665/371472 [6:03:51<1:21:55,  3.42it/s]
 95%|█████████▌| 354666/371472 [6:03:51<1:21:36,  3.43it/s]
 95%|█████████▌| 354667/371472 [6:03:51<1:25:24,  3.28it/s]
 95%|█████████▌| 354668/371472 [6:03:51<1:25:08,  3.29it/s]
 95%|█████████▌| 354669/371472 [6:03:52<1:22:45,  3.38it/s]
 95%|█████████▌| 354670/371472 [6:03:52<1:19:40,  3.51it/s]
 95%|█████████▌| 354671/371472 [6:03:52<1:20:09,  3.49it/s]
 95%|█████████▌| 354672/371472 [6:03:53<1:17:30,  3.61it/s]
 95%|█████████▌| 354673/371472 [6:03:53<1:16:15,  3.67it/s]
 95%|█████████▌| 354674/371472 [6:03:53<1:14:32,  3.76it/s]
 95%|█████████▌| 354675/371472 [6:03:53<1:14:58,  3.73it/s]
 95%|█████████▌| 354676/371472 [6:03:54<1:19:41,  3.51it/s]
 95%|█████████▌| 354677/371472 [6:03:54<1:21:53,  3.42it/s]
 95%|█████████▌| 354678/371472 [6:03:54<1:22:59,  3.37it/s]
 95%|█████████▌| 354679/371472 [6:03:55<1:22:43,  3.38it/s]
 95%|█████████▌| 354680/371472 [6:03:55<1:21:12,  3.45it/s]
                                                           
{'loss': 2.6113, 'learning_rate': 1.4070546661207954e-07, 'epoch': 15.28}
+
 95%|█████████▌| 354680/371472 [6:03:55<1:21:12,  3.45it/s]
 95%|█████████▌| 354681/371472 [6:03:55<1:24:48,  3.30it/s]
 95%|█████████▌| 354682/371472 [6:03:55<1:23:31,  3.35it/s]
 95%|█████████▌| 354683/371472 [6:03:56<1:24:40,  3.30it/s]
 95%|█████████▌| 354684/371472 [6:03:56<1:24:47,  3.30it/s]
 95%|█████████▌| 354685/371472 [6:03:56<1:21:22,  3.44it/s]
 95%|█████████▌| 354686/371472 [6:03:57<1:19:28,  3.52it/s]
 95%|█████████▌| 354687/371472 [6:03:57<1:30:11,  3.10it/s]
 95%|█████████▌| 354688/371472 [6:03:57<1:25:45,  3.26it/s]
 95%|█████████���| 354689/371472 [6:03:58<1:28:18,  3.17it/s]
 95%|█████████▌| 354690/371472 [6:03:58<1:25:04,  3.29it/s]
 95%|█████████▌| 354691/371472 [6:03:58<1:27:37,  3.19it/s]
 95%|█████████▌| 354692/371472 [6:03:59<1:23:52,  3.33it/s]
 95%|█████████▌| 354693/371472 [6:03:59<1:22:23,  3.39it/s]
 95%|█████████▌| 354694/371472 [6:03:59<1:20:56,  3.45it/s]
 95%|█████████▌| 354695/371472 [6:03:59<1:21:12,  3.44it/s]
 95%|█████████▌| 354696/371472 [6:04:00<1:18:43,  3.55it/s]
 95%|█████████▌| 354697/371472 [6:04:00<1:18:10,  3.58it/s]
 95%|█████████▌| 354698/371472 [6:04:00<1:16:34,  3.65it/s]
 95%|█████████▌| 354699/371472 [6:04:01<1:23:01,  3.37it/s]
 95%|█████████▌| 354700/371472 [6:04:01<1:21:54,  3.41it/s]
                                                           
{'loss': 2.435, 'learning_rate': 1.406569846366006e-07, 'epoch': 15.28}
+
 95%|█████████▌| 354700/371472 [6:04:01<1:21:54,  3.41it/s]
 95%|█████████▌| 354701/371472 [6:04:01<1:23:30,  3.35it/s]
 95%|█████████▌| 354702/371472 [6:04:01<1:20:41,  3.46it/s]
 95%|█████████▌| 354703/371472 [6:04:02<1:19:05,  3.53it/s]
 95%|█████████▌| 354704/371472 [6:04:02<1:16:01,  3.68it/s]
 95%|█████████▌| 354705/371472 [6:04:02<1:15:45,  3.69it/s]
 95%|█████████▌| 354706/371472 [6:04:03<1:20:41,  3.46it/s]
 95%|█████████▌| 354707/371472 [6:04:03<1:19:53,  3.50it/s]
 95%|█████████▌| 354708/371472 [6:04:03<1:18:32,  3.56it/s]
 95%|█████████▌| 354709/371472 [6:04:03<1:21:49,  3.41it/s]
 95%|█████████▌| 354710/371472 [6:04:04<1:20:24,  3.47it/s]
 95%|█████████▌| 354711/371472 [6:04:04<1:26:20,  3.24it/s]
 95%|█████████▌| 354712/371472 [6:04:04<1:22:14,  3.40it/s]
 95%|█████████▌| 354713/371472 [6:04:05<1:19:09,  3.53it/s]
 95%|█████████▌| 354714/371472 [6:04:05<1:18:37,  3.55it/s]
 95%|█████████▌| 354715/371472 [6:04:05<1:16:51,  3.63it/s]
 95%|█████████▌| 354716/371472 [6:04:05<1:23:32,  3.34it/s]
 95%|█████████▌| 354717/371472 [6:04:06<1:21:35,  3.42it/s]
 95%|█████████▌| 354718/371472 [6:04:06<1:23:06,  3.36it/s]
 95%|█████████▌| 354719/371472 [6:04:06<1:23:47,  3.33it/s]
 95%|█████████▌| 354720/371472 [6:04:07<1:22:03,  3.40it/s]
                                                           
{'loss': 2.563, 'learning_rate': 1.4060850266112177e-07, 'epoch': 15.28}
+
 95%|█████████▌| 354720/371472 [6:04:07<1:22:03,  3.40it/s]
 95%|█████████▌| 354721/371472 [6:04:07<1:19:21,  3.52it/s]
 95%|█████████▌| 354722/371472 [6:04:07<1:19:16,  3.52it/s]
 95%|█████████▌| 354723/371472 [6:04:07<1:17:32,  3.60it/s]
 95%|█████████▌| 354724/371472 [6:04:08<1:19:35,  3.51it/s]
 95%|█████████▌| 354725/371472 [6:04:08<1:22:29,  3.38it/s]
 95%|█████████▌| 354726/371472 [6:04:08<1:20:37,  3.46it/s]
 95%|█████████▌| 354727/371472 [6:04:09<1:26:28,  3.23it/s]
 95%|█████████▌| 354728/371472 [6:04:09<1:22:12,  3.39it/s]
 95%|█████████▌| 354729/371472 [6:04:09<1:24:51,  3.29it/s]
 95%|█████████▌| 354730/371472 [6:04:10<1:23:34,  3.34it/s]
 95%|█████████▌| 354731/371472 [6:04:10<1:22:19,  3.39it/s]
 95%|█████████▌| 354732/371472 [6:04:10<1:20:47,  3.45it/s]
 95%|█████████▌| 354733/371472 [6:04:10<1:20:45,  3.45it/s]
 95%|█████████▌| 354734/371472 [6:04:11<1:18:21,  3.56it/s]
 95%|█████████▌| 354735/371472 [6:04:11<1:18:25,  3.56it/s]
 95%|█████████▌| 354736/371472 [6:04:11<1:18:06,  3.57it/s]
 95%|█████████▌| 354737/371472 [6:04:11<1:15:35,  3.69it/s]
 95%|█████████▌| 354738/371472 [6:04:12<1:17:14,  3.61it/s]
 95%|█████████▌| 354739/371472 [6:04:12<1:16:42,  3.64it/s]
 95%|█████████▌| 354740/371472 [6:04:12<1:22:50,  3.37it/s]
                                                           
{'loss': 2.6459, 'learning_rate': 1.405600206856428e-07, 'epoch': 15.28}
+
 95%|█████████▌| 354740/371472 [6:04:12<1:22:50,  3.37it/s]
 95%|█████████▌| 354741/371472 [6:04:13<1:24:31,  3.30it/s]
 95%|█████████▌| 354742/371472 [6:04:13<1:24:17,  3.31it/s]
 95%|█████████▌| 354743/371472 [6:04:13<1:29:15,  3.12it/s]
 95%|█████████▌| 354744/371472 [6:04:14<1:29:49,  3.10it/s]
 95%|█████████▌| 354745/371472 [6:04:14<1:28:39,  3.14it/s]
 95%|█████████▌| 354746/371472 [6:04:14<1:31:22,  3.05it/s]
 95%|█████████▌| 354747/371472 [6:04:15<1:33:02,  3.00it/s]
 95%|█████████▌| 354748/371472 [6:04:15<1:31:54,  3.03it/s]
 95%|█████████▌| 354749/371472 [6:04:15<1:29:19,  3.12it/s]
 95%|█████████▌| 354750/371472 [6:04:16<1:27:02,  3.20it/s]
 95%|█████████▌| 354751/371472 [6:04:16<1:30:04,  3.09it/s]
 95%|█████████▌| 354752/371472 [6:04:16<1:28:31,  3.15it/s]
 95%|█████████▌| 354753/371472 [6:04:17<1:24:09,  3.31it/s]
 95%|█████████▌| 354754/371472 [6:04:17<1:22:00,  3.40it/s]
 95%|█████████▌| 354755/371472 [6:04:17<1:18:23,  3.55it/s]
 96%|█████████▌| 354756/371472 [6:04:17<1:19:05,  3.52it/s]
 96%|█████████▌| 354757/371472 [6:04:18<1:24:36,  3.29it/s]
 96%|█████████▌| 354758/371472 [6:04:18<1:20:19,  3.47it/s]
 96%|█████████▌| 354759/371472 [6:04:18<1:20:55,  3.44it/s]
 96%|█████████▌| 354760/371472 [6:04:19<1:21:27,  3.42it/s]
                                                           
{'loss': 2.5291, 'learning_rate': 1.4051153871016396e-07, 'epoch': 15.28}
+
 96%|█████████▌| 354760/371472 [6:04:19<1:21:27,  3.42it/s]
 96%|█████████▌| 354761/371472 [6:04:19<1:22:49,  3.36it/s]
 96%|█████████▌| 354762/371472 [6:04:19<1:19:18,  3.51it/s]
 96%|█████████▌| 354763/371472 [6:04:19<1:24:10,  3.31it/s]
 96%|█████████▌| 354764/371472 [6:04:20<1:22:24,  3.38it/s]
 96%|█████████▌| 354765/371472 [6:04:20<1:19:55,  3.48it/s]
 96%|█████████▌| 354766/371472 [6:04:20<1:30:53,  3.06it/s]
 96%|█████████▌| 354767/371472 [6:04:21<1:26:14,  3.23it/s]
 96%|█████████▌| 354768/371472 [6:04:21<1:23:24,  3.34it/s]
 96%|█████████▌| 354769/371472 [6:04:21<1:23:30,  3.33it/s]
 96%|█████████▌| 354770/371472 [6:04:22<1:26:01,  3.24it/s]
 96%|█████████▌| 354771/371472 [6:04:22<1:21:55,  3.40it/s]
 96%|█████████▌| 354772/371472 [6:04:22<1:19:33,  3.50it/s]
 96%|█████████▌| 354773/371472 [6:04:22<1:20:31,  3.46it/s]
 96%|█████████▌| 354774/371472 [6:04:23<1:19:35,  3.50it/s]
 96%|█████████▌| 354775/371472 [6:04:23<1:28:10,  3.16it/s]
 96%|█████████▌| 354776/371472 [6:04:23<1:25:16,  3.26it/s]
 96%|█████████▌| 354777/371472 [6:04:24<1:26:26,  3.22it/s]
 96%|█████████▌| 354778/371472 [6:04:24<1:26:56,  3.20it/s]
 96%|█████████▌| 354779/371472 [6:04:24<1:24:54,  3.28it/s]
 96%|█████████▌| 354780/371472 [6:04:25<1:21:58,  3.39it/s]
                                                           
{'loss': 2.6228, 'learning_rate': 1.4046305673468503e-07, 'epoch': 15.28}
+
 96%|█████████▌| 354780/371472 [6:04:25<1:21:58,  3.39it/s]
 96%|█████████▌| 354781/371472 [6:04:25<1:19:37,  3.49it/s]
 96%|█████████▌| 354782/371472 [6:04:25<1:20:36,  3.45it/s]
 96%|█████████▌| 354783/371472 [6:04:26<1:31:18,  3.05it/s]
 96%|█████████▌| 354784/371472 [6:04:26<1:24:32,  3.29it/s]
 96%|█████████▌| 354785/371472 [6:04:26<1:27:54,  3.16it/s]
 96%|█████████▌| 354786/371472 [6:04:26<1:23:08,  3.34it/s]
 96%|█████████▌| 354787/371472 [6:04:27<1:19:59,  3.48it/s]
 96%|█████████▌| 354788/371472 [6:04:27<1:18:55,  3.52it/s]
 96%|█████████▌| 354789/371472 [6:04:27<1:22:40,  3.36it/s]
 96%|█████████▌| 354790/371472 [6:04:28<1:20:58,  3.43it/s]
 96%|█████████▌| 354791/371472 [6:04:28<1:18:57,  3.52it/s]
 96%|█████████▌| 354792/371472 [6:04:28<1:20:27,  3.46it/s]
 96%|█████████▌| 354793/371472 [6:04:28<1:18:04,  3.56it/s]
 96%|█████████▌| 354794/371472 [6:04:29<1:18:37,  3.54it/s]
 96%|█████████▌| 354795/371472 [6:04:29<1:17:13,  3.60it/s]
 96%|█████████▌| 354796/371472 [6:04:29<1:21:01,  3.43it/s]
 96%|█████████▌| 354797/371472 [6:04:29<1:18:45,  3.53it/s]
 96%|█████████▌| 354798/371472 [6:04:30<1:19:06,  3.51it/s]
 96%|█████████▌| 354799/371472 [6:04:30<1:17:32,  3.58it/s]
 96%|█████████▌| 354800/371472 [6:04:30<1:15:34,  3.68it/s]
                                                           
{'loss': 2.5245, 'learning_rate': 1.4041457475920619e-07, 'epoch': 15.28}
+
 96%|█████████▌| 354800/371472 [6:04:30<1:15:34,  3.68it/s]
 96%|█████���███▌| 354801/371472 [6:04:31<1:13:59,  3.76it/s]
 96%|█████████▌| 354802/371472 [6:04:31<1:14:32,  3.73it/s]
 96%|█████████▌| 354803/371472 [6:04:31<1:19:48,  3.48it/s]
 96%|█████████▌| 354804/371472 [6:04:31<1:18:37,  3.53it/s]
 96%|█████████▌| 354805/371472 [6:04:32<1:21:39,  3.40it/s]
 96%|█████████▌| 354806/371472 [6:04:32<1:20:39,  3.44it/s]
 96%|█████████▌| 354807/371472 [6:04:32<1:19:06,  3.51it/s]
 96%|█████████▌| 354808/371472 [6:04:33<1:17:27,  3.59it/s]
 96%|█████████▌| 354809/371472 [6:04:33<1:20:11,  3.46it/s]
 96%|█████████▌| 354810/371472 [6:04:33<1:17:11,  3.60it/s]
 96%|█████████▌| 354811/371472 [6:04:33<1:19:34,  3.49it/s]
 96%|█████████▌| 354812/371472 [6:04:34<1:20:31,  3.45it/s]
 96%|█████████▌| 354813/371472 [6:04:34<1:23:09,  3.34it/s]
 96%|█████████▌| 354814/371472 [6:04:34<1:20:26,  3.45it/s]
 96%|█████████▌| 354815/371472 [6:04:35<1:23:03,  3.34it/s]
 96%|█████████▌| 354816/371472 [6:04:35<1:24:43,  3.28it/s]
 96%|█████████▌| 354817/371472 [6:04:35<1:21:25,  3.41it/s]
 96%|█████████▌| 354818/371472 [6:04:36<1:20:20,  3.45it/s]
 96%|█████████▌| 354819/371472 [6:04:36<1:19:29,  3.49it/s]
 96%|█████████▌| 354820/371472 [6:04:36<1:17:25,  3.58it/s]
                                                           
{'loss': 2.5262, 'learning_rate': 1.4036609278372726e-07, 'epoch': 15.28}
+
 96%|█████████▌| 354820/371472 [6:04:36<1:17:25,  3.58it/s]
 96%|█████████▌| 354821/371472 [6:04:36<1:17:11,  3.59it/s]
 96%|█████████▌| 354822/371472 [6:04:37<1:16:49,  3.61it/s]
 96%|█████████▌| 354823/371472 [6:04:37<1:18:30,  3.53it/s]
 96%|█████████▌| 354824/371472 [6:04:37<1:15:27,  3.68it/s]
 96%|█████████▌| 354825/371472 [6:04:37<1:12:46,  3.81it/s]
 96%|█████████▌| 354826/371472 [6:04:38<1:17:18,  3.59it/s]
 96%|█████████▌| 354827/371472 [6:04:38<1:21:06,  3.42it/s]
 96%|█████████▌| 354828/371472 [6:04:38<1:19:00,  3.51it/s]
 96%|█████████▌| 354829/371472 [6:04:39<1:21:08,  3.42it/s]
 96%|█████████▌| 354830/371472 [6:04:39<1:17:24,  3.58it/s]
 96%|█████████▌| 354831/371472 [6:04:39<1:15:25,  3.68it/s]
 96%|█████████▌| 354832/371472 [6:04:39<1:12:24,  3.83it/s]
 96%|█████████▌| 354833/371472 [6:04:40<1:17:12,  3.59it/s]
 96%|█████████▌| 354834/371472 [6:04:40<1:21:41,  3.39it/s]
 96%|█████████▌| 354835/371472 [6:04:40<1:19:59,  3.47it/s]
 96%|█████████▌| 354836/371472 [6:04:41<1:18:31,  3.53it/s]
 96%|█████████▌| 354837/371472 [6:04:41<1:20:07,  3.46it/s]
 96%|█████████▌| 354838/371472 [6:04:41<1:20:58,  3.42it/s]
 96%|█████████▌| 354839/371472 [6:04:41<1:21:10,  3.42it/s]
 96%|█████████▌| 354840/371472 [6:04:42<1:24:45,  3.27it/s]
                                                           
{'loss': 2.6586, 'learning_rate': 1.403176108082484e-07, 'epoch': 15.28}
+
 96%|█████████▌| 354840/371472 [6:04:42<1:24:45,  3.27it/s]
 96%|█████████▌| 354841/371472 [6:04:42<1:21:21,  3.41it/s]
 96%|█████████▌| 354842/371472 [6:04:42<1:20:48,  3.43it/s]
 96%|█████████▌| 354843/371472 [6:04:43<1:18:21,  3.54it/s]
 96%|█████████▌| 354844/371472 [6:04:43<1:16:44,  3.61it/s]
 96%|█████████▌| 354845/371472 [6:04:43<1:23:10,  3.33it/s]
 96%|█████████▌| 354846/371472 [6:04:43<1:20:02,  3.46it/s]
 96%|█████████▌| 354847/371472 [6:04:44<1:18:44,  3.52it/s]
 96%|█████████▌| 354848/371472 [6:04:44<1:17:38,  3.57it/s]
 96%|█████████▌| 354849/371472 [6:04:44<1:16:38,  3.62it/s]
 96%|█████████▌| 354850/371472 [6:04:45<1:16:58,  3.60it/s]
 96%|█████████▌| 354851/371472 [6:04:45<1:16:25,  3.63it/s]
 96%|█████████▌| 354852/371472 [6:04:45<1:16:42,  3.61it/s]
 96%|█████████▌| 354853/371472 [6:04:46<1:26:12,  3.21it/s]
 96%|█████████▌| 354854/371472 [6:04:46<1:21:04,  3.42it/s]
 96%|█████████▌| 354855/371472 [6:04:46<1:20:36,  3.44it/s]
 96%|█████████▌| 354856/371472 [6:04:46<1:18:54,  3.51it/s]
 96%|█████████▌| 354857/371472 [6:04:47<1:17:25,  3.58it/s]
 96%|█████████▌| 354858/371472 [6:04:47<1:17:55,  3.55it/s]
 96%|█████████▌| 354859/371472 [6:04:47<1:13:48,  3.75it/s]
 96%|█████████▌| 354860/371472 [6:04:47<1:15:41,  3.66it/s]
                                                           
{'loss': 2.521, 'learning_rate': 1.4026912883276948e-07, 'epoch': 15.28}
+
 96%|█████████▌| 354860/371472 [6:04:47<1:15:41,  3.66it/s]
 96%|█████████▌| 354861/371472 [6:04:48<1:20:26,  3.44it/s]
 96%|█████████▌| 354862/371472 [6:04:48<1:18:54,  3.51it/s]
 96%|█████████▌| 354863/371472 [6:04:48<1:20:37,  3.43it/s]
 96%|█████████▌| 354864/371472 [6:04:49<1:21:58,  3.38it/s]
 96%|█████████▌| 354865/371472 [6:04:49<1:29:05,  3.11it/s]
 96%|█████████▌| 354866/371472 [6:04:49<1:23:43,  3.31it/s]
 96%|█████████▌| 354867/371472 [6:04:50<1:22:04,  3.37it/s]
 96%|█████████▌| 354868/371472 [6:04:50<1:26:47,  3.19it/s]
 96%|█████████▌| 354869/371472 [6:04:50<1:22:15,  3.36it/s]
 96%|█████████▌| 354870/371472 [6:04:50<1:20:31,  3.44it/s]
 96%|█████████▌| 354871/371472 [6:04:51<1:21:54,  3.38it/s]
 96%|█████████▌| 354872/371472 [6:04:51<1:19:47,  3.47it/s]
 96%|█████████▌| 354873/371472 [6:04:51<1:17:42,  3.56it/s]
 96%|█████████▌| 354874/371472 [6:04:52<1:17:54,  3.55it/s]
 96%|█████████▌| 354875/371472 [6:04:52<1:19:10,  3.49it/s]
 96%|█████████▌| 354876/371472 [6:04:52<1:24:52,  3.26it/s]
 96%|█████████▌| 354877/371472 [6:04:52<1:21:30,  3.39it/s]
 96%|█████████▌| 354878/371472 [6:04:53<1:19:42,  3.47it/s]
 96%|█████████▌| 354879/371472 [6:04:53<1:21:02,  3.41it/s]
 96%|█████████▌| 354880/371472 [6:04:53<1:17:04,  3.59it/s]
                                                           
{'loss': 2.6081, 'learning_rate': 1.4022064685729063e-07, 'epoch': 15.29}
+
 96%|█████████▌| 354880/371472 [6:04:53<1:17:04,  3.59it/s]
 96%|█████████▌| 354881/371472 [6:04:54<1:24:42,  3.26it/s]
 96%|█████████▌| 354882/371472 [6:04:54<1:20:03,  3.45it/s]
 96%|█████████▌| 354883/371472 [6:04:54<1:18:16,  3.53it/s]
 96%|█████████▌| 354884/371472 [6:04:55<1:25:52,  3.22it/s]
 96%|█████████▌| 354885/371472 [6:04:55<1:21:33,  3.39it/s]
 96%|█████████▌| 354886/371472 [6:04:55<1:22:14,  3.36it/s]
 96%|█████████▌| 354887/371472 [6:04:55<1:21:52,  3.38it/s]
 96%|█████████▌| 354888/371472 [6:04:56<1:22:50,  3.34it/s]
 96%|█████████▌| 354889/371472 [6:04:56<1:19:05,  3.49it/s]
 96%|█████████▌| 354890/371472 [6:04:56<1:19:33,  3.47it/s]
 96%|█████████▌| 354891/371472 [6:04:57<1:20:17,  3.44it/s]
 96%|█████████▌| 354892/371472 [6:04:57<1:16:30,  3.61it/s]
 96%|█████████▌| 354893/371472 [6:04:57<1:14:50,  3.69it/s]
 96%|█████████▌| 354894/371472 [6:04:57<1:17:17,  3.57it/s]
 96%|█████████▌| 354895/371472 [6:04:58<1:14:39,  3.70it/s]
 96%|█████████▌| 354896/371472 [6:04:58<1:21:45,  3.38it/s]
 96%|█████████▌| 354897/371472 [6:04:58<1:17:45,  3.55it/s]
 96%|█████████▌| 354898/371472 [6:04:58<1:15:24,  3.66it/s]
 96%|█████████▌| 354899/371472 [6:04:59<1:14:28,  3.71it/s]
 96%|█████████▌| 354900/371472 [6:04:59<1:16:40,  3.60it/s]
                                                           
{'loss': 2.6854, 'learning_rate': 1.4017216488181168e-07, 'epoch': 15.29}
+
 96%|█████████▌| 354900/371472 [6:04:59<1:16:40,  3.60it/s]
 96%|█████████▌| 354901/371472 [6:04:59<1:16:03,  3.63it/s]
 96%|█████████▌| 354902/371472 [6:05:00<1:13:47,  3.74it/s]
 96%|█████████▌| 354903/371472 [6:05:00<1:13:32,  3.75it/s]
 96%|█████████▌| 354904/371472 [6:05:00<1:11:30,  3.86it/s]
 96%|█████████▌| 354905/371472 [6:05:00<1:10:45,  3.90it/s]
 96%|█████████▌| 354906/371472 [6:05:01<1:11:35,  3.86it/s]
 96%|█████████▌| 354907/371472 [6:05:01<1:10:38,  3.91it/s]
 96%|█████████▌| 354908/371472 [6:05:01<1:13:48,  3.74it/s]
 96%|█████████▌| 354909/371472 [6:05:01<1:11:30,  3.86it/s]
 96%|█████████▌| 354910/371472 [6:05:02<1:14:29,  3.71it/s]
 96%|█████████▌| 354911/371472 [6:05:02<1:15:57,  3.63it/s]
 96%|█████████▌| 354912/371472 [6:05:02<1:21:12,  3.40it/s]
 96%|█████████▌| 354913/371472 [6:05:03<1:24:44,  3.26it/s]
 96%|█████████▌| 354914/371472 [6:05:03<1:21:04,  3.40it/s]
 96%|█████████▌| 354915/371472 [6:05:03<1:19:04,  3.49it/s]
 96%|█████████▌| 354916/371472 [6:05:03<1:15:29,  3.66it/s]
 96%|█████████▌| 354917/371472 [6:05:04<1:15:14,  3.67it/s]
 96%|█████████▌| 354918/371472 [6:05:04<1:15:19,  3.66it/s]
 96%|█████████▌| 354919/371472 [6:05:04<1:15:33,  3.65it/s]
 96%|█████████▌| 354920/371472 [6:05:05<1:18:26,  3.52it/s]
                                                           
{'loss': 2.4633, 'learning_rate': 1.4012368290633285e-07, 'epoch': 15.29}
+
 96%|█████████▌| 354920/371472 [6:05:05<1:18:26,  3.52it/s]
 96%|█████████▌| 354921/371472 [6:05:05<1:18:54,  3.50it/s]
 96%|█████████▌| 354922/371472 [6:05:05<1:16:29,  3.61it/s]
 96%|█████████▌| 354923/371472 [6:05:05<1:22:06,  3.36it/s]
 96%|█████████▌| 354924/371472 [6:05:06<1:20:24,  3.43it/s]
 96%|█████████▌| 354925/371472 [6:05:06<1:17:48,  3.54it/s]
 96%|█████████▌| 354926/371472 [6:05:06<1:18:40,  3.51it/s]
 96%|█████████▌| 354927/371472 [6:05:07<1:19:24,  3.47it/s]
 96%|█████████▌| 354928/371472 [6:05:07<1:19:28,  3.47it/s]
 96%|█████████▌| 354929/371472 [6:05:07<1:19:32,  3.47it/s]
 96%|█████████▌| 354930/371472 [6:05:07<1:16:22,  3.61it/s]
 96%|█████████▌| 354931/371472 [6:05:08<1:16:02,  3.63it/s]
 96%|█████████▌| 354932/371472 [6:05:08<1:17:19,  3.57it/s]
 96%|█████████▌| 354933/371472 [6:05:08<1:16:04,  3.62it/s]
 96%|█████████▌| 354934/371472 [6:05:08<1:16:28,  3.60it/s]
 96%|█████████▌| 354935/371472 [6:05:09<1:18:43,  3.50it/s]
 96%|█████████▌| 354936/371472 [6:05:09<1:17:16,  3.57it/s]
 96%|█████████▌| 354937/371472 [6:05:09<1:17:20,  3.56it/s]
 96%|█████████▌| 354938/371472 [6:05:10<1:15:59,  3.63it/s]
 96%|█████████▌| 354939/371472 [6:05:10<1:19:05,  3.48it/s]
 96%|█████████▌| 354940/371472 [6:05:10<1:17:15,  3.57it/s]
                                                           
{'loss': 2.5005, 'learning_rate': 1.400752009308539e-07, 'epoch': 15.29}
+
 96%|█████████▌| 354940/371472 [6:05:10<1:17:15,  3.57it/s]
 96%|█████████▌| 354941/371472 [6:05:10<1:14:48,  3.68it/s]
 96%|█████████▌| 354942/371472 [6:05:11<1:15:43,  3.64it/s]
 96%|█████████▌| 354943/371472 [6:05:11<1:16:53,  3.58it/s]
 96%|█████████▌| 354944/371472 [6:05:11<1:22:58,  3.32it/s]
 96%|█████████▌| 354945/371472 [6:05:12<1:24:13,  3.27it/s]
 96%|█████████▌| 354946/371472 [6:05:12<1:21:14,  3.39it/s]
 96%|█████████▌| 354947/371472 [6:05:12<1:16:46,  3.59it/s]
 96%|█████████▌| 354948/371472 [6:05:13<1:26:29,  3.18it/s]
 96%|█████████▌| 354949/371472 [6:05:13<1:20:56,  3.40it/s]
 96%|█████████▌| 354950/371472 [6:05:13<1:25:25,  3.22it/s]
 96%|█████████▌| 354951/371472 [6:05:13<1:24:43,  3.25it/s]
 96%|█████████▌| 354952/371472 [6:05:14<1:25:17,  3.23it/s]
 96%|█████████▌| 354953/371472 [6:05:14<1:22:18,  3.35it/s]
 96%|█████████▌| 354954/371472 [6:05:14<1:21:11,  3.39it/s]
 96%|█████████▌| 354955/371472 [6:05:15<1:17:42,  3.54it/s]
 96%|█████████▌| 354956/371472 [6:05:15<1:21:40,  3.37it/s]
 96%|█████████▌| 354957/371472 [6:05:15<1:17:10,  3.57it/s]
 96%|█████████▌| 354958/371472 [6:05:15<1:15:34,  3.64it/s]
 96%|█████████▌| 354959/371472 [6:05:16<1:15:30,  3.65it/s]
 96%|█████████▌| 354960/371472 [6:05:16<1:23:31,  3.29it/s]
                                                           
{'loss': 2.5811, 'learning_rate': 1.4002671895537505e-07, 'epoch': 15.29}
+
 96%|█████████▌| 354960/371472 [6:05:16<1:23:31,  3.29it/s]
 96%|█████████▌| 354961/371472 [6:05:16<1:19:51,  3.45it/s]
 96%|█████████▌| 354962/371472 [6:05:17<1:16:23,  3.60it/s]
 96%|█████████▌| 354963/371472 [6:05:17<1:14:29,  3.69it/s]
 96%|█████████▌| 354964/371472 [6:05:17<1:22:47,  3.32it/s]
 96%|█████████▌| 354965/371472 [6:05:18<1:24:46,  3.25it/s]
 96%|█████████▌| 354966/371472 [6:05:18<1:19:38,  3.45it/s]
 96%|█████████▌| 354967/371472 [6:05:18<1:16:52,  3.58it/s]
 96%|█████████▌| 354968/371472 [6:05:18<1:27:28,  3.14it/s]
 96%|█████████▌| 354969/371472 [6:05:19<1:24:36,  3.25it/s]
 96%|█████████▌| 354970/371472 [6:05:19<1:20:33,  3.41it/s]
 96%|█████████▌| 354971/371472 [6:05:19<1:17:06,  3.57it/s]
 96%|█████████▌| 354972/371472 [6:05:19<1:16:47,  3.58it/s]
 96%|█████████▌| 354973/371472 [6:05:20<1:14:55,  3.67it/s]
 96%|█████████▌| 354974/371472 [6:05:20<1:13:47,  3.73it/s]
 96%|█████████▌| 354975/371472 [6:05:20<1:13:00,  3.77it/s]
 96%|█████████▌| 354976/371472 [6:05:21<1:13:31,  3.74it/s]
 96%|█████████▌| 354977/371472 [6:05:21<1:19:50,  3.44it/s]
 96%|█████████▌| 354978/371472 [6:05:21<1:23:10,  3.31it/s]
 96%|█████████▌| 354979/371472 [6:05:22<1:22:29,  3.33it/s]
 96%|█████████▌| 354980/371472 [6:05:22<1:18:59,  3.48it/s]
                                                           
{'loss': 2.7288, 'learning_rate': 1.3997823697989612e-07, 'epoch': 15.29}
+
 96%|█████████▌| 354980/371472 [6:05:22<1:18:59,  3.48it/s]
 96%|█████████▌| 354981/371472 [6:05:22<1:23:22,  3.30it/s]
 96%|█████████▌| 354982/371472 [6:05:22<1:20:22,  3.42it/s]
 96%|█████████▌| 354983/371472 [6:05:23<1:18:31,  3.50it/s]
 96%|█████████▌| 354984/371472 [6:05:23<1:17:12,  3.56it/s]
 96%|█████████▌| 354985/371472 [6:05:23<1:15:43,  3.63it/s]
 96%|█████████▌| 354986/371472 [6:05:23<1:14:05,  3.71it/s]
 96%|█████████▌| 354987/371472 [6:05:24<1:14:11,  3.70it/s]
 96%|█████████▌| 354988/371472 [6:05:24<1:14:29,  3.69it/s]
 96%|█████████▌| 354989/371472 [6:05:24<1:15:41,  3.63it/s]
 96%|█████████▌| 354990/371472 [6:05:25<1:18:31,  3.50it/s]
 96%|█████████▌| 354991/371472 [6:05:25<1:21:19,  3.38it/s]
 96%|█████████▌| 354992/371472 [6:05:25<1:24:31,  3.25it/s]
 96%|█████████▌| 354993/371472 [6:05:26<1:25:00,  3.23it/s]
 96%|█████████▌| 354994/371472 [6:05:26<1:23:45,  3.28it/s]
 96%|█████████▌| 354995/371472 [6:05:26<1:22:46,  3.32it/s]
 96%|█████████▌| 354996/371472 [6:05:26<1:23:29,  3.29it/s]
 96%|█████████▌| 354997/371472 [6:05:27<1:18:50,  3.48it/s]
 96%|█████████▌| 354998/371472 [6:05:27<1:26:10,  3.19it/s]
 96%|█████████▌| 354999/371472 [6:05:27<1:24:36,  3.24it/s]
 96%|█████████▌| 355000/371472 [6:05:28<1:29:37,  3.06it/s]
                                                           
{'loss': 2.4586, 'learning_rate': 1.3992975500441727e-07, 'epoch': 15.29}
+
 96%|█████████▌| 355000/371472 [6:05:28<1:29:37,  3.06it/s]
 96%|█████████▌| 355001/371472 [6:05:28<1:30:16,  3.04it/s]
 96%|█████████▌| 355002/371472 [6:05:28<1:25:23,  3.21it/s]
 96%|█████████▌| 355003/371472 [6:05:29<1:35:18,  2.88it/s]
 96%|█████████▌| 355004/371472 [6:05:29<1:27:41,  3.13it/s]
 96%|█████████▌| 355005/371472 [6:05:29<1:22:22,  3.33it/s]
 96%|█████████▌| 355006/371472 [6:05:30<1:19:58,  3.43it/s]
 96%|█████████▌| 355007/371472 [6:05:30<1:16:16,  3.60it/s]
 96%|█████████▌| 355008/371472 [6:05:30<1:16:56,  3.57it/s]
 96%|█████████▌| 355009/371472 [6:05:30<1:14:32,  3.68it/s]
 96%|█████████▌| 355010/371472 [6:05:31<1:15:01,  3.66it/s]
 96%|█████████▌| 355011/371472 [6:05:31<1:16:13,  3.60it/s]
 96%|█████████▌| 355012/371472 [6:05:31<1:14:48,  3.67it/s]
 96%|█████████▌| 355013/371472 [6:05:32<1:26:40,  3.17it/s]
 96%|█████████▌| 355014/371472 [6:05:32<1:25:31,  3.21it/s]
 96%|█████████▌| 355015/371472 [6:05:32<1:20:44,  3.40it/s]
 96%|█████████▌| 355016/371472 [6:05:32<1:18:13,  3.51it/s]
 96%|█████████▌| 355017/371472 [6:05:33<1:19:04,  3.47it/s]
 96%|█████████▌| 355018/371472 [6:05:33<1:18:46,  3.48it/s]
 96%|█████████▌| 355019/371472 [6:05:33<1:15:13,  3.65it/s]
 96%|█████████▌| 355020/371472 [6:05:33<1:15:12,  3.65it/s]
                                                           
{'loss': 2.6385, 'learning_rate': 1.3988127302893832e-07, 'epoch': 15.29}
+
 96%|█████████▌| 355020/371472 [6:05:33<1:15:12,  3.65it/s]
 96%|█████████▌| 355021/371472 [6:05:34<1:14:27,  3.68it/s]
 96%|█████████▌| 355022/371472 [6:05:34<1:14:03,  3.70it/s]
 96%|█████████▌| 355023/371472 [6:05:34<1:15:38,  3.62it/s]
 96%|█████████▌| 355024/371472 [6:05:35<1:13:11,  3.75it/s]
 96%|█████████▌| 355025/371472 [6:05:35<1:15:37,  3.62it/s]
 96%|█████████▌| 355026/371472 [6:05:35<1:16:43,  3.57it/s]
 96%|█████████▌| 355027/371472 [6:05:35<1:16:40,  3.57it/s]
 96%|█████████▌| 355028/371472 [6:05:36<1:18:03,  3.51it/s]
 96%|█████████▌| 355029/371472 [6:05:36<1:16:10,  3.60it/s]
 96%|█████████▌| 355030/371472 [6:05:36<1:18:17,  3.50it/s]
 96%|█████████▌| 355031/371472 [6:05:37<1:19:41,  3.44it/s]
 96%|█████████▌| 355032/371472 [6:05:37<1:23:15,  3.29it/s]
 96%|█████████▌| 355033/371472 [6:05:37<1:24:09,  3.26it/s]
 96%|█████████▌| 355034/371472 [6:05:38<1:24:08,  3.26it/s]
 96%|█████████▌| 355035/371472 [6:05:38<1:20:28,  3.40it/s]
 96%|█████████▌| 355036/371472 [6:05:38<1:21:59,  3.34it/s]
 96%|█████████▌| 355037/371472 [6:05:38<1:18:35,  3.49it/s]
 96%|█████████▌| 355038/371472 [6:05:39<1:18:07,  3.51it/s]
 96%|█████████▌| 355039/371472 [6:05:39<1:18:09,  3.50it/s]
 96%|█████████▌| 355040/371472 [6:05:39<1:15:42,  3.62it/s]
                                                           
{'loss': 2.5244, 'learning_rate': 1.398327910534595e-07, 'epoch': 15.29}
+
 96%|█████████▌| 355040/371472 [6:05:39<1:15:42,  3.62it/s]
 96%|█████████▌| 355041/371472 [6:05:39<1:14:32,  3.67it/s]
 96%|█████████▌| 355042/371472 [6:05:40<1:16:55,  3.56it/s]
 96%|█████████▌| 355043/371472 [6:05:40<1:16:19,  3.59it/s]
 96%|█████████▌| 355044/371472 [6:05:40<1:17:02,  3.55it/s]
 96%|█████████▌| 355045/371472 [6:05:41<1:13:11,  3.74it/s]
 96%|█████████▌| 355046/371472 [6:05:41<1:13:05,  3.75it/s]
 96%|█████████▌| 355047/371472 [6:05:41<1:10:52,  3.86it/s]
 96%|█████████▌| 355048/371472 [6:05:41<1:09:36,  3.93it/s]
 96%|█████████▌| 355049/371472 [6:05:42<1:15:03,  3.65it/s]
 96%|█████████▌| 355050/371472 [6:05:42<1:20:39,  3.39it/s]
 96%|█████████▌| 355051/371472 [6:05:42<1:19:36,  3.44it/s]
 96%|█████████▌| 355052/371472 [6:05:43<1:20:46,  3.39it/s]
 96%|█████████▌| 355053/371472 [6:05:43<1:22:32,  3.32it/s]
 96%|█████████▌| 355054/371472 [6:05:43<1:20:17,  3.41it/s]
 96%|█████████▌| 355055/371472 [6:05:43<1:19:49,  3.43it/s]
 96%|█████████▌| 355056/371472 [6:05:44<1:18:09,  3.50it/s]
 96%|█████████▌| 355057/371472 [6:05:44<1:22:21,  3.32it/s]
 96%|█████████▌| 355058/371472 [6:05:44<1:17:50,  3.51it/s]
 96%|█████████▌| 355059/371472 [6:05:45<1:17:22,  3.54it/s]
 96%|█████████▌| 355060/371472 [6:05:45<1:14:27,  3.67it/s]
                                                           
{'loss': 2.6683, 'learning_rate': 1.3978430907798054e-07, 'epoch': 15.29}
+
 96%|█████████▌| 355060/371472 [6:05:45<1:14:27,  3.67it/s]
 96%|█████████▌| 355061/371472 [6:05:45<1:20:35,  3.39it/s]
 96%|█████████▌| 355062/371472 [6:05:45<1:17:00,  3.55it/s]
 96%|█████████▌| 355063/371472 [6:05:46<1:13:53,  3.70it/s]
 96%|█████████▌| 355064/371472 [6:05:46<1:15:35,  3.62it/s]
 96%|█████████▌| 355065/371472 [6:05:46<1:14:34,  3.67it/s]
 96%|█████████▌| 355066/371472 [6:05:46<1:13:24,  3.73it/s]
 96%|█████████▌| 355067/371472 [6:05:47<1:17:04,  3.55it/s]
 96%|█████████▌| 355068/371472 [6:05:47<1:17:33,  3.53it/s]
 96%|█████████▌| 355069/371472 [6:05:47<1:18:24,  3.49it/s]
 96%|█████████▌| 355070/371472 [6:05:48<1:15:31,  3.62it/s]
 96%|█████████▌| 355071/371472 [6:05:48<1:18:42,  3.47it/s]
 96%|█████████▌| 355072/371472 [6:05:48<1:16:38,  3.57it/s]
 96%|█████████▌| 355073/371472 [6:05:48<1:14:36,  3.66it/s]
 96%|█████████▌| 355074/371472 [6:05:49<1:16:06,  3.59it/s]
 96%|█████████▌| 355075/371472 [6:05:49<1:15:30,  3.62it/s]
 96%|█████████▌| 355076/371472 [6:05:49<1:18:44,  3.47it/s]
 96%|█████████▌| 355077/371472 [6:05:50<1:16:21,  3.58it/s]
 96%|█████████▌| 355078/371472 [6:05:50<1:13:55,  3.70it/s]
 96%|█████████▌| 355079/371472 [6:05:50<1:16:15,  3.58it/s]
 96%|█████████▌| 355080/371472 [6:05:50<1:17:32,  3.52it/s]
                                                           
{'loss': 2.5025, 'learning_rate': 1.397358271025017e-07, 'epoch': 15.29}
+
 96%|█████████▌| 355080/371472 [6:05:50<1:17:32,  3.52it/s]
 96%|█████████▌| 355081/371472 [6:05:51<1:16:52,  3.55it/s]
 96%|█████████▌| 355082/371472 [6:05:51<1:19:54,  3.42it/s]
 96%|█████████▌| 355083/371472 [6:05:51<1:25:42,  3.19it/s]
 96%|█████████▌| 355084/371472 [6:05:52<1:32:59,  2.94it/s]
 96%|█████████▌| 355085/371472 [6:05:52<1:27:44,  3.11it/s]
 96%|█████████▌| 355086/371472 [6:05:52<1:24:26,  3.23it/s]
 96%|█████████▌| 355087/371472 [6:05:53<1:20:47,  3.38it/s]
 96%|█████████▌| 355088/371472 [6:05:53<1:24:17,  3.24it/s]
 96%|█████████▌| 355089/371472 [6:05:53<1:21:39,  3.34it/s]
 96%|█████████▌| 355090/371472 [6:05:54<1:19:18,  3.44it/s]
 96%|█████████▌| 355091/371472 [6:05:54<1:16:54,  3.55it/s]
 96%|█████████▌| 355092/371472 [6:05:54<1:29:13,  3.06it/s]
 96%|█████████▌| 355093/371472 [6:05:54<1:23:16,  3.28it/s]
 96%|█████████▌| 355094/371472 [6:05:55<1:17:43,  3.51it/s]
 96%|█████████▌| 355095/371472 [6:05:55<1:15:31,  3.61it/s]
 96%|█████████▌| 355096/371472 [6:05:55<1:14:02,  3.69it/s]
 96%|█████████▌| 355097/371472 [6:05:56<1:16:23,  3.57it/s]
 96%|█████████▌| 355098/371472 [6:05:56<1:14:52,  3.64it/s]
 96%|█████████▌| 355099/371472 [6:05:56<1:14:20,  3.67it/s]
 96%|█████████▌| 355100/371472 [6:05:56<1:19:13,  3.44it/s]
                                                           
{'loss': 2.5527, 'learning_rate': 1.3968734512702276e-07, 'epoch': 15.29}
+
 96%|█████████▌| 355100/371472 [6:05:56<1:19:13,  3.44it/s]
 96%|█████████▌| 355101/371472 [6:05:57<1:19:01,  3.45it/s]
 96%|█████████▌| 355102/371472 [6:05:57<1:17:19,  3.53it/s]
 96%|█████████▌| 355103/371472 [6:05:57<1:14:32,  3.66it/s]
 96%|█████████▌| 355104/371472 [6:05:58<1:19:28,  3.43it/s]
 96%|█████████▌| 355105/371472 [6:05:58<1:19:13,  3.44it/s]
 96%|█████████▌| 355106/371472 [6:05:58<1:17:15,  3.53it/s]
 96%|█████████▌| 355107/371472 [6:05:58<1:19:07,  3.45it/s]
 96%|█████████▌| 355108/371472 [6:05:59<1:14:34,  3.66it/s]
 96%|█████████▌| 355109/371472 [6:05:59<1:17:08,  3.54it/s]
 96%|█████████▌| 355110/371472 [6:05:59<1:14:23,  3.67it/s]
 96%|█████████▌| 355111/371472 [6:05:59<1:15:08,  3.63it/s]
 96%|█████████▌| 355112/371472 [6:06:00<1:17:53,  3.50it/s]
 96%|█████████▌| 355113/371472 [6:06:00<1:22:31,  3.30it/s]
 96%|█████████▌| 355114/371472 [6:06:00<1:25:36,  3.18it/s]
 96%|█████████▌| 355115/371472 [6:06:01<1:21:45,  3.33it/s]
 96%|█████████▌| 355116/371472 [6:06:01<1:18:23,  3.48it/s]
 96%|█████████▌| 355117/371472 [6:06:01<1:20:18,  3.39it/s]
 96%|█████████▌| 355118/371472 [6:06:02<1:16:27,  3.56it/s]
 96%|█████████▌| 355119/371472 [6:06:02<1:13:51,  3.69it/s]
 96%|█████████▌| 355120/371472 [6:06:02<1:12:38,  3.75it/s]
                                                           
{'loss': 2.5698, 'learning_rate': 1.396388631515439e-07, 'epoch': 15.3}
+
 96%|█████████▌| 355120/371472 [6:06:02<1:12:38,  3.75it/s]
 96%|█████████▌| 355121/371472 [6:06:02<1:13:09,  3.73it/s]
 96%|█████████▌| 355122/371472 [6:06:03<1:18:22,  3.48it/s]
 96%|█████████▌| 355123/371472 [6:06:03<1:16:38,  3.56it/s]
 96%|█████████▌| 355124/371472 [6:06:03<1:15:25,  3.61it/s]
 96%|█████████▌| 355125/371472 [6:06:03<1:13:57,  3.68it/s]
 96%|█████████▌| 355126/371472 [6:06:04<1:13:12,  3.72it/s]
 96%|█████████▌| 355127/371472 [6:06:04<1:13:50,  3.69it/s]
 96%|█████████▌| 355128/371472 [6:06:04<1:15:11,  3.62it/s]
 96%|█████████▌| 355129/371472 [6:06:05<1:19:29,  3.43it/s]
 96%|█████████▌| 355130/371472 [6:06:05<1:21:15,  3.35it/s]
 96%|█████████▌| 355131/371472 [6:06:05<1:21:39,  3.34it/s]
 96%|█████████▌| 355132/371472 [6:06:05<1:18:05,  3.49it/s]
 96%|█████████▌| 355133/371472 [6:06:06<1:15:22,  3.61it/s]
 96%|█████████▌| 355134/371472 [6:06:06<1:14:14,  3.67it/s]
 96%|█████████▌| 355135/371472 [6:06:06<1:16:33,  3.56it/s]
 96%|█████████▌| 355136/371472 [6:06:07<1:14:40,  3.65it/s]
 96%|█████████▌| 355137/371472 [6:06:07<1:20:00,  3.40it/s]
 96%|█████████▌| 355138/371472 [6:06:07<1:19:31,  3.42it/s]
 96%|█████████▌| 355139/371472 [6:06:07<1:21:55,  3.32it/s]
 96%|█████████▌| 355140/371472 [6:06:08<1:18:46,  3.46it/s]
                                                           
{'loss': 2.7704, 'learning_rate': 1.3959038117606498e-07, 'epoch': 15.3}
+
 96%|█████████▌| 355140/371472 [6:06:08<1:18:46,  3.46it/s]
 96%|█████████▌| 355141/371472 [6:06:08<1:17:29,  3.51it/s]
 96%|█████████▌| 355142/371472 [6:06:08<1:30:28,  3.01it/s]
 96%|█████████▌| 355143/371472 [6:06:09<1:25:32,  3.18it/s]
 96%|█████████▌| 355144/371472 [6:06:09<1:20:17,  3.39it/s]
 96%|█████████▌| 355145/371472 [6:06:09<1:19:03,  3.44it/s]
 96%|█████████▌| 355146/371472 [6:06:10<1:17:50,  3.50it/s]
 96%|█████████▌| 355147/371472 [6:06:10<1:17:43,  3.50it/s]
 96%|█████████▌| 355148/371472 [6:06:10<1:15:54,  3.58it/s]
 96%|█████████▌| 355149/371472 [6:06:10<1:17:23,  3.51it/s]
 96%|█████████▌| 355150/371472 [6:06:11<1:14:25,  3.66it/s]
 96%|█████████▌| 355151/371472 [6:06:11<1:19:25,  3.43it/s]
 96%|█████████▌| 355152/371472 [6:06:11<1:27:14,  3.12it/s]
 96%|█████████▌| 355153/371472 [6:06:12<1:22:40,  3.29it/s]
 96%|█████████▌| 355154/371472 [6:06:12<1:19:10,  3.43it/s]
 96%|█████████▌| 355155/371472 [6:06:12<1:16:33,  3.55it/s]
 96%|█████████▌| 355156/371472 [6:06:12<1:17:14,  3.52it/s]
 96%|█████████▌| 355157/371472 [6:06:13<1:17:56,  3.49it/s]
 96%|█████████▌| 355158/371472 [6:06:13<1:14:28,  3.65it/s]
 96%|█████████▌| 355159/371472 [6:06:13<1:14:01,  3.67it/s]
 96%|█████████▌| 355160/371472 [6:06:13<1:13:06,  3.72it/s]
                                                           
{'loss': 2.406, 'learning_rate': 1.3954189920058613e-07, 'epoch': 15.3}
+
 96%|█████████▌| 355160/371472 [6:06:13<1:13:06,  3.72it/s]
 96%|█████████▌| 355161/371472 [6:06:14<1:16:09,  3.57it/s]
 96%|█████████▌| 355162/371472 [6:06:14<1:17:02,  3.53it/s]
 96%|█████████▌| 355163/371472 [6:06:14<1:19:54,  3.40it/s]
 96%|█████████▌| 355164/371472 [6:06:15<1:18:45,  3.45it/s]
 96%|█████████▌| 355165/371472 [6:06:15<1:21:07,  3.35it/s]
 96%|█████████▌| 355166/371472 [6:06:15<1:18:48,  3.45it/s]
 96%|█████████▌| 355167/371472 [6:06:16<1:14:49,  3.63it/s]
 96%|█████████▌| 355168/371472 [6:06:16<1:17:34,  3.50it/s]
 96%|█████████▌| 355169/371472 [6:06:16<1:21:40,  3.33it/s]
 96%|█████████▌| 355170/371472 [6:06:16<1:18:41,  3.45it/s]
 96%|█████████▌| 355171/371472 [6:06:17<1:19:32,  3.42it/s]
 96%|█████████▌| 355172/371472 [6:06:17<1:16:44,  3.54it/s]
 96%|█████████▌| 355173/371472 [6:06:17<1:16:04,  3.57it/s]
 96%|█████████▌| 355174/371472 [6:06:18<1:17:16,  3.52it/s]
 96%|█████████▌| 355175/371472 [6:06:18<1:14:43,  3.63it/s]
 96%|█████████▌| 355176/371472 [6:06:18<1:14:43,  3.63it/s]
 96%|█████████▌| 355177/371472 [6:06:18<1:16:42,  3.54it/s]
 96%|█████████▌| 355178/371472 [6:06:19<1:18:49,  3.45it/s]
 96%|█████████▌| 355179/371472 [6:06:19<1:16:52,  3.53it/s]
 96%|█████████▌| 355180/371472 [6:06:19<1:16:39,  3.54it/s]
                                                           
{'loss': 2.8004, 'learning_rate': 1.394934172251072e-07, 'epoch': 15.3}
+
 96%|█████████▌| 355180/371472 [6:06:19<1:16:39,  3.54it/s]
 96%|█████████▌| 355181/371472 [6:06:20<1:17:21,  3.51it/s]
 96%|█████████▌| 355182/371472 [6:06:20<1:15:19,  3.60it/s]
 96%|█████████▌| 355183/371472 [6:06:20<1:13:58,  3.67it/s]
 96%|█████████▌| 355184/371472 [6:06:20<1:12:05,  3.77it/s]
 96%|█████████▌| 355185/371472 [6:06:21<1:13:34,  3.69it/s]
 96%|█████████▌| 355186/371472 [6:06:21<1:18:40,  3.45it/s]
 96%|█████████▌| 355187/371472 [6:06:21<1:21:39,  3.32it/s]
 96%|█████████▌| 355188/371472 [6:06:22<1:25:50,  3.16it/s]
 96%|█████████▌| 355189/371472 [6:06:22<1:23:38,  3.24it/s]
 96%|█████████▌| 355190/371472 [6:06:22<1:19:08,  3.43it/s]
 96%|█████████▌| 355191/371472 [6:06:22<1:16:13,  3.56it/s]
 96%|█████████▌| 355192/371472 [6:06:23<1:16:11,  3.56it/s]
 96%|█████████▌| 355193/371472 [6:06:23<1:16:59,  3.52it/s]
 96%|█████████▌| 355194/371472 [6:06:23<1:18:25,  3.46it/s]
 96%|█████████▌| 355195/371472 [6:06:24<1:15:44,  3.58it/s]
 96%|█████████▌| 355196/371472 [6:06:24<1:15:44,  3.58it/s]
 96%|█████████▌| 355197/371472 [6:06:24<1:19:36,  3.41it/s]
 96%|█���███████▌| 355198/371472 [6:06:24<1:16:24,  3.55it/s]
 96%|█████████▌| 355199/371472 [6:06:25<1:14:04,  3.66it/s]
 96%|█████████▌| 355200/371472 [6:06:25<1:12:30,  3.74it/s]
                                                           
{'loss': 2.5196, 'learning_rate': 1.3944493524962836e-07, 'epoch': 15.3}
+
 96%|█████████▌| 355200/371472 [6:06:25<1:12:30,  3.74it/s]
 96%|█████████▌| 355201/371472 [6:06:25<1:14:02,  3.66it/s]
 96%|█████████▌| 355202/371472 [6:06:25<1:13:30,  3.69it/s]
 96%|█████████▌| 355203/371472 [6:06:26<1:16:08,  3.56it/s]
 96%|█████████▌| 355204/371472 [6:06:26<1:15:39,  3.58it/s]
 96%|█████████▌| 355205/371472 [6:06:26<1:16:51,  3.53it/s]
 96%|█████████▌| 355206/371472 [6:06:27<1:20:43,  3.36it/s]
 96%|█████████▌| 355207/371472 [6:06:27<1:18:26,  3.46it/s]
 96%|█████████▌| 355208/371472 [6:06:27<1:18:54,  3.43it/s]
 96%|█████████▌| 355209/371472 [6:06:28<1:21:37,  3.32it/s]
 96%|█████████▌| 355210/371472 [6:06:28<1:20:52,  3.35it/s]
 96%|█████████▌| 355211/371472 [6:06:28<1:16:18,  3.55it/s]
 96%|█████████▌| 355212/371472 [6:06:28<1:16:11,  3.56it/s]
 96%|█████████▌| 355213/371472 [6:06:29<1:15:13,  3.60it/s]
 96%|█████████▌| 355214/371472 [6:06:29<1:16:59,  3.52it/s]
 96%|█████████▌| 355215/371472 [6:06:29<1:21:50,  3.31it/s]
 96%|█████████▌| 355216/371472 [6:06:30<1:27:01,  3.11it/s]
 96%|█████████▌| 355217/371472 [6:06:30<1:26:25,  3.13it/s]
 96%|█████████▌| 355218/371472 [6:06:30<1:24:10,  3.22it/s]
 96%|█████████▌| 355219/371472 [6:06:31<1:21:24,  3.33it/s]
 96%|█████████▌| 355220/371472 [6:06:31<1:20:33,  3.36it/s]
                                                           
{'loss': 2.665, 'learning_rate': 1.393964532741494e-07, 'epoch': 15.3}
+
 96%|█████████▌| 355220/371472 [6:06:31<1:20:33,  3.36it/s]
 96%|█████████▌| 355221/371472 [6:06:31<1:18:13,  3.46it/s]
 96%|█████████▌| 355222/371472 [6:06:31<1:16:43,  3.53it/s]
 96%|█████████▌| 355223/371472 [6:06:32<1:17:44,  3.48it/s]
 96%|█████████▌| 355224/371472 [6:06:32<1:18:37,  3.44it/s]
 96%|█████████▌| 355225/371472 [6:06:32<1:17:57,  3.47it/s]
 96%|█████████▌| 355226/371472 [6:06:33<1:22:50,  3.27it/s]
 96%|█████████▌| 355227/371472 [6:06:33<1:20:24,  3.37it/s]
 96%|█████████▌| 355228/371472 [6:06:33<1:17:53,  3.48it/s]
 96%|█████████▌| 355229/371472 [6:06:33<1:16:19,  3.55it/s]
 96%|█████████▌| 355230/371472 [6:06:34<1:21:02,  3.34it/s]
 96%|█████████▌| 355231/371472 [6:06:34<1:17:27,  3.49it/s]
 96%|█████████▌| 355232/371472 [6:06:34<1:18:45,  3.44it/s]
 96%|█████████▌| 355233/371472 [6:06:35<1:16:26,  3.54it/s]
 96%|█████████▌| 355234/371472 [6:06:35<1:16:50,  3.52it/s]
 96%|█████████▌| 355235/371472 [6:06:35<1:15:46,  3.57it/s]
 96%|█████████▌| 355236/371472 [6:06:35<1:19:16,  3.41it/s]
 96%|█████████▌| 355237/371472 [6:06:36<1:15:53,  3.57it/s]
 96%|█████████▌| 355238/371472 [6:06:36<1:17:16,  3.50it/s]
 96%|█████████▌| 355239/371472 [6:06:36<1:18:53,  3.43it/s]
 96%|█████████▌| 355240/371472 [6:06:37<1:20:20,  3.37it/s]
                                                           
{'loss': 2.6103, 'learning_rate': 1.3934797129867047e-07, 'epoch': 15.3}
+
 96%|█████████▌| 355240/371472 [6:06:37<1:20:20,  3.37it/s]
 96%|█████████▌| 355241/371472 [6:06:37<1:35:47,  2.82it/s]
 96%|█████████▌| 355242/371472 [6:06:37<1:28:16,  3.06it/s]
 96%|█████████▌| 355243/371472 [6:06:38<1:23:55,  3.22it/s]
 96%|█████████▌| 355244/371472 [6:06:38<1:26:30,  3.13it/s]
 96%|█████████▌| 355245/371472 [6:06:38<1:23:12,  3.25it/s]
 96%|█████████▌| 355246/371472 [6:06:39<1:21:18,  3.33it/s]
 96%|█████████▌| 355247/371472 [6:06:39<1:18:22,  3.45it/s]
 96%|█████████▌| 355248/371472 [6:06:39<1:23:05,  3.25it/s]
 96%|█████████▌| 355249/371472 [6:06:39<1:19:01,  3.42it/s]
 96%|█████████▌| 355250/371472 [6:06:40<1:20:05,  3.38it/s]
 96%|█████████▌| 355251/371472 [6:06:40<1:21:41,  3.31it/s]
 96%|█████████▌| 355252/371472 [6:06:40<1:17:27,  3.49it/s]
 96%|█████████▌| 355253/371472 [6:06:41<1:17:44,  3.48it/s]
 96%|���████████▌| 355254/371472 [6:06:41<1:17:32,  3.49it/s]
 96%|█████████▌| 355255/371472 [6:06:41<1:16:10,  3.55it/s]
 96%|█████████▌| 355256/371472 [6:06:41<1:18:24,  3.45it/s]
 96%|█████████▌| 355257/371472 [6:06:42<1:18:41,  3.43it/s]
 96%|█████████▌| 355258/371472 [6:06:42<1:15:30,  3.58it/s]
 96%|█████████▌| 355259/371472 [6:06:42<1:16:07,  3.55it/s]
 96%|█████████▌| 355260/371472 [6:06:42<1:15:03,  3.60it/s]
                                                           
{'loss': 2.587, 'learning_rate': 1.3929948932319162e-07, 'epoch': 15.3}
+
 96%|█████████▌| 355260/371472 [6:06:43<1:15:03,  3.60it/s]
 96%|█████████▌| 355261/371472 [6:06:43<1:14:56,  3.61it/s]
 96%|█████████▌| 355262/371472 [6:06:43<1:14:58,  3.60it/s]
 96%|█████████▌| 355263/371472 [6:06:43<1:18:37,  3.44it/s]
 96%|█████████▌| 355264/371472 [6:06:44<1:20:41,  3.35it/s]
 96%|█████████▌| 355265/371472 [6:06:44<1:24:13,  3.21it/s]
 96%|█████████▌| 355266/371472 [6:06:44<1:19:44,  3.39it/s]
 96%|█████████▌| 355267/371472 [6:06:45<1:19:20,  3.40it/s]
 96%|█████████▌| 355268/371472 [6:06:45<1:15:17,  3.59it/s]
 96%|█████████▌| 355269/371472 [6:06:45<1:12:56,  3.70it/s]
 96%|█████████▌| 355270/371472 [6:06:45<1:13:46,  3.66it/s]
 96%|█████████▌| 355271/371472 [6:06:46<1:12:38,  3.72it/s]
 96%|█████████▌| 355272/371472 [6:06:46<1:10:40,  3.82it/s]
 96%|█████████▌| 355273/371472 [6:06:46<1:15:32,  3.57it/s]
 96%|█████████▌| 355274/371472 [6:06:46<1:13:26,  3.68it/s]
 96%|█████████▌| 355275/371472 [6:06:47<1:14:48,  3.61it/s]
 96%|█████████▌| 355276/371472 [6:06:47<1:13:38,  3.67it/s]
 96%|█████████▌| 355277/371472 [6:06:47<1:20:18,  3.36it/s]
 96%|█████████▌| 355278/371472 [6:06:48<1:18:05,  3.46it/s]
 96%|█████████▌| 355279/371472 [6:06:48<1:15:12,  3.59it/s]
 96%|█████████▌| 355280/371472 [6:06:48<1:12:16,  3.73it/s]
                                                           
{'loss': 2.7581, 'learning_rate': 1.3925100734771267e-07, 'epoch': 15.3}
+
 96%|█████████▌| 355280/371472 [6:06:48<1:12:16,  3.73it/s]
 96%|█████████▌| 355281/371472 [6:06:48<1:18:46,  3.43it/s]
 96%|█████████▌| 355282/371472 [6:06:49<1:20:05,  3.37it/s]
 96%|█████████▌| 355283/371472 [6:06:49<1:17:07,  3.50it/s]
 96%|█████████▌| 355284/371472 [6:06:49<1:15:14,  3.59it/s]
 96%|█████████▌| 355285/371472 [6:06:50<1:16:32,  3.52it/s]
 96%|█████████▌| 355286/371472 [6:06:50<1:17:07,  3.50it/s]
 96%|█████████▌| 355287/371472 [6:06:50<1:14:25,  3.62it/s]
 96%|█████████▌| 355288/371472 [6:06:50<1:16:13,  3.54it/s]
 96%|█████████▌| 355289/371472 [6:06:51<1:16:49,  3.51it/s]
 96%|█████████▌| 355290/371472 [6:06:51<1:15:21,  3.58it/s]
 96%|█████████▌| 355291/371472 [6:06:51<1:13:09,  3.69it/s]
 96%|█████████▌| 355292/371472 [6:06:52<1:13:35,  3.66it/s]
 96%|█████████▌| 355293/371472 [6:06:52<1:14:26,  3.62it/s]
 96%|█████████▌| 355294/371472 [6:06:52<1:12:37,  3.71it/s]
 96%|█████████▌| 355295/371472 [6:06:52<1:18:51,  3.42it/s]
 96%|█████████▌| 355296/371472 [6:06:53<1:18:58,  3.41it/s]
 96%|█████████▌| 355297/371472 [6:06:53<1:15:58,  3.55it/s]
 96%|█████████▌| 355298/371472 [6:06:53<1:14:48,  3.60it/s]
 96%|█████████▌| 355299/371472 [6:06:53<1:15:08,  3.59it/s]
 96%|█████████▌| 355300/371472 [6:06:54<1:20:56,  3.33it/s]
                                                           
{'loss': 2.5031, 'learning_rate': 1.3920252537223385e-07, 'epoch': 15.3}
+
 96%|█████████▌| 355300/371472 [6:06:54<1:20:56,  3.33it/s]
 96%|█████████▌| 355301/371472 [6:06:54<1:16:31,  3.52it/s]
 96%|█████████▌| 355302/371472 [6:06:54<1:16:30,  3.52it/s]
 96%|█████████▌| 355303/371472 [6:06:55<1:15:52,  3.55it/s]
 96%|█████████▌| 355304/371472 [6:06:55<1:16:17,  3.53it/s]
 96%|█████████▌| 355305/371472 [6:06:55<1:14:05,  3.64it/s]
 96%|█████████▌| 355306/371472 [6:06:55<1:15:06,  3.59it/s]
 96%|█████████▌| 355307/371472 [6:06:56<1:16:01,  3.54it/s]
 96%|█████████▌| 355308/371472 [6:06:56<1:14:52,  3.60it/s]
 96%|█████████▌| 355309/371472 [6:06:56<1:14:41,  3.61it/s]
 96%|█████████▌| 355310/371472 [6:06:57<1:15:43,  3.56it/s]
 96%|█████████▌| 355311/371472 [6:06:57<1:13:16,  3.68it/s]
 96%|█████████▌| 355312/371472 [6:06:57<1:14:21,  3.62it/s]
 96%|█████████▌| 355313/371472 [6:06:58<1:26:25,  3.12it/s]
 96%|█████████▌| 355314/371472 [6:06:58<1:22:10,  3.28it/s]
 96%|█████████▌| 355315/371472 [6:06:58<1:19:20,  3.39it/s]
 96%|█████████▌| 355316/371472 [6:06:58<1:18:19,  3.44it/s]
 96%|█████████▌| 355317/371472 [6:06:59<1:15:56,  3.55it/s]
 96%|█████████▌| 355318/371472 [6:06:59<1:17:06,  3.49it/s]
 96%|█████████▌| 355319/371472 [6:06:59<1:16:21,  3.53it/s]
 96%|█████████▌| 355320/371472 [6:06:59<1:14:33,  3.61it/s]
                                                           
{'loss': 2.8277, 'learning_rate': 1.3915404339675487e-07, 'epoch': 15.3}
+
 96%|█████████▌| 355320/371472 [6:06:59<1:14:33,  3.61it/s]
 96%|█████████▌| 355321/371472 [6:07:00<1:14:50,  3.60it/s]
 96%|█████████▌| 355322/371472 [6:07:00<1:13:04,  3.68it/s]
 96%|█████████▌| 355323/371472 [6:07:00<1:13:23,  3.67it/s]
 96%|█████████▌| 355324/371472 [6:07:01<1:12:56,  3.69it/s]
 96%|█████████▌| 355325/371472 [6:07:01<1:12:50,  3.69it/s]
 96%|█████████▌| 355326/371472 [6:07:01<1:22:39,  3.26it/s]
 96%|█████████▌| 355327/371472 [6:07:01<1:18:34,  3.42it/s]
 96%|█████████▌| 355328/371472 [6:07:02<1:20:11,  3.36it/s]
 96%|█████████▌| 355329/371472 [6:07:02<1:18:03,  3.45it/s]
 96%|█████████▌| 355330/371472 [6:07:02<1:16:39,  3.51it/s]
 96%|█████████▌| 355331/371472 [6:07:03<1:14:54,  3.59it/s]
 96%|█████████▌| 355332/371472 [6:07:03<1:14:46,  3.60it/s]
 96%|█████████▌| 355333/371472 [6:07:03<1:12:55,  3.69it/s]
 96%|█████████▌| 355334/371472 [6:07:03<1:12:43,  3.70it/s]
 96%|█████████▌| 355335/371472 [6:07:04<1:13:30,  3.66it/s]
 96%|█████████▌| 355336/371472 [6:07:04<1:14:04,  3.63it/s]
 96%|█████████▌| 355337/371472 [6:07:04<1:13:24,  3.66it/s]
 96%|█████████▌| 355338/371472 [6:07:05<1:13:06,  3.68it/s]
 96%|█████████▌| 355339/371472 [6:07:05<1:13:35,  3.65it/s]
 96%|█████████▌| 355340/371472 [6:07:05<1:13:06,  3.68it/s]
                                                           
{'loss': 2.5798, 'learning_rate': 1.3910556142127604e-07, 'epoch': 15.31}
+
 96%|█████████▌| 355340/371472 [6:07:05<1:13:06,  3.68it/s]
 96%|█████████▌| 355341/371472 [6:07:05<1:14:32,  3.61it/s]
 96%|█████████▌| 355342/371472 [6:07:06<1:14:16,  3.62it/s]
 96%|█████████▌| 355343/371472 [6:07:06<1:13:42,  3.65it/s]
 96%|█████████▌| 355344/371472 [6:07:06<1:19:06,  3.40it/s]
 96%|█████████▌| 355345/371472 [6:07:06<1:15:19,  3.57it/s]
 96%|█████████▌| 355346/371472 [6:07:07<1:14:58,  3.59it/s]
 96%|█████████▌| 355347/371472 [6:07:07<1:16:59,  3.49it/s]
 96%|█████████▌| 355348/371472 [6:07:07<1:17:51,  3.45it/s]
 96%|█████████▌| 355349/371472 [6:07:08<1:15:10,  3.57it/s]
 96%|█████████▌| 355350/371472 [6:07:08<1:13:06,  3.68it/s]
 96%|█████████▌| 355351/371472 [6:07:08<1:12:12,  3.72it/s]
 96%|█████████▌| 355352/371472 [6:07:08<1:10:55,  3.79it/s]
 96%|█████████▌| 355353/371472 [6:07:09<1:12:49,  3.69it/s]
 96%|█████████▌| 355354/371472 [6:07:09<1:16:02,  3.53it/s]
 96%|█████████▌| 355355/371472 [6:07:09<1:13:51,  3.64it/s]
 96%|█████████▌| 355356/371472 [6:07:10<1:16:16,  3.52it/s]
 96%|█████████▌| 355357/371472 [6:07:10<1:14:46,  3.59it/s]
 96%|█████████▌| 355358/371472 [6:07:10<1:20:17,  3.34it/s]
 96%|█████████▌| 355359/371472 [6:07:10<1:22:58,  3.24it/s]
 96%|█████████▌| 355360/371472 [6:07:11<1:20:12,  3.35it/s]
                                                           
{'loss': 2.566, 'learning_rate': 1.3905707944579711e-07, 'epoch': 15.31}
+
 96%|█████████▌| 355360/371472 [6:07:11<1:20:12,  3.35it/s]
 96%|█████████▌| 355361/371472 [6:07:11<1:18:52,  3.40it/s]
 96%|█████████▌| 355362/371472 [6:07:11<1:16:36,  3.51it/s]
 96%|█████████▌| 355363/371472 [6:07:12<1:18:57,  3.40it/s]
 96%|█████████▌| 355364/371472 [6:07:12<1:19:03,  3.40it/s]
 96%|█████████▌| 355365/371472 [6:07:12<1:18:30,  3.42it/s]
 96%|█████████▌| 355366/371472 [6:07:13<1:21:28,  3.29it/s]
 96%|█████████▌| 355367/371472 [6:07:13<1:20:15,  3.34it/s]
 96%|█████████▌| 355368/371472 [6:07:13<1:17:04,  3.48it/s]
 96%|█████████▌| 355369/371472 [6:07:13<1:15:42,  3.55it/s]
 96%|█████████▌| 355370/371472 [6:07:14<1:20:13,  3.35it/s]
 96%|█████████▌| 355371/371472 [6:07:14<1:16:17,  3.52it/s]
 96%|█████████▌| 355372/371472 [6:07:14<1:13:44,  3.64it/s]
 96%|█████████▌| 355373/371472 [6:07:14<1:13:15,  3.66it/s]
 96%|█████████▌| 355374/371472 [6:07:15<1:14:44,  3.59it/s]
 96%|█████████▌| 355375/371472 [6:07:15<1:17:36,  3.46it/s]
 96%|█████████▌| 355376/371472 [6:07:15<1:16:28,  3.51it/s]
 96%|█████████▌| 355377/371472 [6:07:16<1:15:44,  3.54it/s]
 96%|█████████▌| 355378/371472 [6:07:16<1:25:43,  3.13it/s]
 96%|█████████▌| 355379/371472 [6:07:16<1:21:42,  3.28it/s]
 96%|█████████▌| 355380/371472 [6:07:17<1:23:25,  3.21it/s]
                                                           
{'loss': 2.6812, 'learning_rate': 1.3900859747031826e-07, 'epoch': 15.31}
+
 96%|█████████▌| 355380/371472 [6:07:17<1:23:25,  3.21it/s]
 96%|█████████▌| 355381/371472 [6:07:17<1:18:56,  3.40it/s]
 96%|█████████▌| 355382/371472 [6:07:17<1:21:56,  3.27it/s]
 96%|█████████▌| 355383/371472 [6:07:17<1:18:44,  3.41it/s]
 96%|█████████▌| 355384/371472 [6:07:18<1:20:17,  3.34it/s]
 96%|█████████▌| 355385/371472 [6:07:18<1:19:04,  3.39it/s]
 96%|█████████▌| 355386/371472 [6:07:18<1:15:48,  3.54it/s]
 96%|█████████▌| 355387/371472 [6:07:19<1:16:56,  3.48it/s]
 96%|█████████▌| 355388/371472 [6:07:19<1:15:06,  3.57it/s]
 96%|█████████▌| 355389/371472 [6:07:19<1:13:23,  3.65it/s]
 96%|█████████▌| 355390/371472 [6:07:19<1:11:22,  3.76it/s]
 96%|█████████▌| 355391/371472 [6:07:20<1:13:05,  3.67it/s]
 96%|█████████▌| 355392/371472 [6:07:20<1:15:44,  3.54it/s]
 96%|█████████▌| 355393/371472 [6:07:20<1:17:15,  3.47it/s]
 96%|█████████▌| 355394/371472 [6:07:21<1:17:53,  3.44it/s]
 96%|█████████▌| 355395/371472 [6:07:21<1:23:05,  3.22it/s]
 96%|█████████▌| 355396/371472 [6:07:21<1:20:50,  3.31it/s]
 96%|█████████▌| 355397/371472 [6:07:21<1:16:59,  3.48it/s]
 96%|█████████▌| 355398/371472 [6:07:22<1:16:24,  3.51it/s]
 96%|█████████▌| 355399/371472 [6:07:22<1:15:29,  3.55it/s]
 96%|█████████▌| 355400/371472 [6:07:22<1:17:55,  3.44it/s]
                                                           
{'loss': 2.7132, 'learning_rate': 1.389601154948393e-07, 'epoch': 15.31}
+
 96%|█████████▌| 355400/371472 [6:07:22<1:17:55,  3.44it/s]
 96%|█████████▌| 355401/371472 [6:07:23<1:15:19,  3.56it/s]
 96%|█████████▌| 355402/371472 [6:07:23<1:12:58,  3.67it/s]
 96%|█████████▌| 355403/371472 [6:07:23<1:16:07,  3.52it/s]
 96%|█████████▌| 355404/371472 [6:07:24<1:22:32,  3.24it/s]
 96%|█████████▌| 355405/371472 [6:07:24<1:18:04,  3.43it/s]
 96%|█████████▌| 355406/371472 [6:07:24<1:26:20,  3.10it/s]
 96%|█████████▌| 355407/371472 [6:07:24<1:19:36,  3.36it/s]
 96%|█████████▌| 355408/371472 [6:07:25<1:18:43,  3.40it/s]
 96%|█████████▌| 355409/371472 [6:07:25<1:17:21,  3.46it/s]
 96%|█████████▌| 355410/371472 [6:07:25<1:14:20,  3.60it/s]
 96%|█████████▌| 355411/371472 [6:07:26<1:49:56,  2.43it/s]
 96%|█████████▌| 355412/371472 [6:07:26<1:40:54,  2.65it/s]
 96%|█████████▌| 355413/371472 [6:07:27<1:33:40,  2.86it/s]
 96%|█████████▌| 355414/371472 [6:07:27<1:28:55,  3.01it/s]
 96%|█████████▌| 355415/371472 [6:07:27<1:25:11,  3.14it/s]
 96%|█████████▌| 355416/371472 [6:07:27<1:22:39,  3.24it/s]
 96%|█████████▌| 355417/371472 [6:07:28<1:17:53,  3.44it/s]
 96%|█████████▌| 355418/371472 [6:07:28<1:15:28,  3.55it/s]
 96%|█████████▌| 355419/371472 [6:07:28<1:12:16,  3.70it/s]
 96%|█████████▌| 355420/371472 [6:07:28<1:15:15,  3.55it/s]
                                                           
{'loss': 2.5162, 'learning_rate': 1.389116335193605e-07, 'epoch': 15.31}
+
 96%|█████████▌| 355420/371472 [6:07:28<1:15:15,  3.55it/s]
 96%|█████████▌| 355421/371472 [6:07:29<1:18:29,  3.41it/s]
 96%|█████████▌| 355422/371472 [6:07:29<1:14:37,  3.58it/s]
 96%|█████████▌| 355423/371472 [6:07:29<1:14:09,  3.61it/s]
 96%|█████████▌| 355424/371472 [6:07:30<1:11:30,  3.74it/s]
 96%|█████████▌| 355425/371472 [6:07:30<1:13:15,  3.65it/s]
 96%|█████████▌| 355426/371472 [6:07:30<1:11:34,  3.74it/s]
 96%|█████████▌| 355427/371472 [6:07:30<1:11:49,  3.72it/s]
 96%|█████████▌| 355428/371472 [6:07:31<1:11:07,  3.76it/s]
 96%|█████████▌| 355429/371472 [6:07:31<1:12:32,  3.69it/s]
 96%|█████████▌| 355430/371472 [6:07:31<1:16:10,  3.51it/s]
 96%|█████████▌| 355431/371472 [6:07:31<1:15:52,  3.52it/s]
 96%|█████████▌| 355432/371472 [6:07:32<1:14:31,  3.59it/s]
 96%|█████████▌| 355433/371472 [6:07:32<1:16:47,  3.48it/s]
 96%|█████████▌| 355434/371472 [6:07:32<1:15:12,  3.55it/s]
 96%|█████████▌| 355435/371472 [6:07:33<1:13:11,  3.65it/s]
 96%|█████████▌| 355436/371472 [6:07:33<1:16:39,  3.49it/s]
 96%|█████████▌| 355437/371472 [6:07:33<1:16:46,  3.48it/s]
 96%|█████████▌| 355438/371472 [6:07:34<1:19:15,  3.37it/s]
 96%|█████████▌| 355439/371472 [6:07:34<1:18:14,  3.42it/s]
 96%|█████████▌| 355440/371472 [6:07:34<1:15:53,  3.52it/s]
                                                           
{'loss': 2.6849, 'learning_rate': 1.3886315154388153e-07, 'epoch': 15.31}
+
 96%|█████████▌| 355440/371472 [6:07:34<1:15:53,  3.52it/s]
 96%|█████████▌| 355441/371472 [6:07:34<1:17:29,  3.45it/s]
 96%|█████████▌| 355442/371472 [6:07:35<1:16:08,  3.51it/s]
 96%|█████████▌| 355443/371472 [6:07:35<1:23:19,  3.21it/s]
 96%|█████████▌| 355444/371472 [6:07:35<1:20:54,  3.30it/s]
 96%|█████████▌| 355445/371472 [6:07:36<1:17:13,  3.46it/s]
 96%|█████████▌| 355446/371472 [6:07:36<1:16:09,  3.51it/s]
 96%|█████████▌| 355447/371472 [6:07:36<1:19:03,  3.38it/s]
 96%|█████████▌| 355448/371472 [6:07:36<1:19:54,  3.34it/s]
 96%|█████████▌| 355449/371472 [6:07:37<1:17:38,  3.44it/s]
 96%|█████████▌| 355450/371472 [6:07:37<1:16:52,  3.47it/s]
 96%|█████████▌| 355451/371472 [6:07:37<1:18:05,  3.42it/s]
 96%|█████████▌| 355452/371472 [6:07:38<1:17:18,  3.45it/s]
 96%|█████████▌| 355453/371472 [6:07:38<1:15:34,  3.53it/s]
 96%|█████████▌| 355454/371472 [6:07:38<1:17:35,  3.44it/s]
 96%|█████████▌| 355455/371472 [6:07:38<1:14:53,  3.56it/s]
 96%|█████████▌| 355456/371472 [6:07:39<1:13:48,  3.62it/s]
 96%|█████████▌| 355457/371472 [6:07:39<1:27:38,  3.05it/s]
 96%|█████████▌| 355458/371472 [6:07:39<1:26:55,  3.07it/s]
 96%|█████████▌| 355459/371472 [6:07:40<1:33:38,  2.85it/s]
 96%|█████████▌| 355460/371472 [6:07:40<1:27:04,  3.06it/s]
                                                           
{'loss': 2.5024, 'learning_rate': 1.3881466956840268e-07, 'epoch': 15.31}
+
 96%|█████████▌| 355460/371472 [6:07:40<1:27:04,  3.06it/s]
 96%|█████████▌| 355461/371472 [6:07:40<1:25:31,  3.12it/s]
 96%|█████████▌| 355462/371472 [6:07:41<1:23:50,  3.18it/s]
 96%|█████████▌| 355463/371472 [6:07:41<1:24:08,  3.17it/s]
 96%|█████████▌| 355464/371472 [6:07:41<1:19:16,  3.37it/s]
 96%|█████████▌| 355465/371472 [6:07:42<1:16:44,  3.48it/s]
 96%|█████████▌| 355466/371472 [6:07:42<1:14:51,  3.56it/s]
 96%|█████████▌| 355467/371472 [6:07:42<1:11:48,  3.71it/s]
 96%|█████████▌| 355468/371472 [6:07:42<1:12:40,  3.67it/s]
 96%|█████████▌| 355469/371472 [6:07:43<1:13:14,  3.64it/s]
 96%|█████████▌| 355470/371472 [6:07:43<1:21:50,  3.26it/s]
 96%|█████████▌| 355471/371472 [6:07:43<1:25:31,  3.12it/s]
 96%|█████████▌| 355472/371472 [6:07:44<1:27:22,  3.05it/s]
 96%|█████████▌| 355473/371472 [6:07:44<1:24:35,  3.15it/s]
 96%|█████████▌| 355474/371472 [6:07:44<1:20:12,  3.32it/s]
 96%|█████████▌| 355475/371472 [6:07:45<1:16:38,  3.48it/s]
 96%|█████████▌| 355476/371472 [6:07:45<1:16:49,  3.47it/s]
 96%|█████████▌| 355477/371472 [6:07:45<1:16:09,  3.50it/s]
 96%|█████████▌| 355478/371472 [6:07:46<1:25:22,  3.12it/s]
 96%|█████████▌| 355479/371472 [6:07:46<1:25:21,  3.12it/s]
 96%|█████████▌| 355480/371472 [6:07:46<1:23:40,  3.19it/s]
                                                           
{'loss': 2.617, 'learning_rate': 1.3876618759292375e-07, 'epoch': 15.31}
+
 96%|█████████▌| 355480/371472 [6:07:46<1:23:40,  3.19it/s]
 96%|█████████▌| 355481/371472 [6:07:46<1:19:38,  3.35it/s]
 96%|█████████▌| 355482/371472 [6:07:47<1:19:52,  3.34it/s]
 96%|█████████▌| 355483/371472 [6:07:47<1:18:09,  3.41it/s]
 96%|█████████▌| 355484/371472 [6:07:47<1:16:18,  3.49it/s]
 96%|█████████▌| 355485/371472 [6:07:47<1:12:58,  3.65it/s]
 96%|█████████▌| 355486/371472 [6:07:48<1:14:42,  3.57it/s]
 96%|█████████▌| 355487/371472 [6:07:48<1:20:10,  3.32it/s]
 96%|█████████▌| 355488/371472 [6:07:48<1:17:06,  3.45it/s]
 96%|█████████▌| 355489/371472 [6:07:49<1:15:24,  3.53it/s]
 96%|█████████▌| 355490/371472 [6:07:49<1:15:38,  3.52it/s]
 96%|█████████▌| 355491/371472 [6:07:49<1:15:28,  3.53it/s]
 96%|█████████▌| 355492/371472 [6:07:49<1:13:32,  3.62it/s]
 96%|█████████▌| 355493/371472 [6:07:50<1:27:14,  3.05it/s]
 96%|█████████▌| 355494/371472 [6:07:50<1:22:18,  3.24it/s]
 96%|█████████▌| 355495/371472 [6:07:51<1:22:21,  3.23it/s]
 96%|█████████▌| 355496/371472 [6:07:51<1:23:16,  3.20it/s]
 96%|█████████▌| 355497/371472 [6:07:51<1:20:49,  3.29it/s]
 96%|█████████▌| 355498/371472 [6:07:51<1:18:40,  3.38it/s]
 96%|█████████▌| 355499/371472 [6:07:52<1:17:50,  3.42it/s]
 96%|█████████▌| 355500/371472 [6:07:52<1:16:19,  3.49it/s]
                                                           
{'loss': 2.6808, 'learning_rate': 1.3871770561744493e-07, 'epoch': 15.31}
+
 96%|█████████▌| 355500/371472 [6:07:52<1:16:19,  3.49it/s]
 96%|█████████▌| 355501/371472 [6:07:52<1:16:20,  3.49it/s]
 96%|█████████▌| 355502/371472 [6:07:53<1:18:27,  3.39it/s]
 96%|█████████▌| 355503/371472 [6:07:53<1:26:15,  3.09it/s]
 96%|█████████▌| 355504/371472 [6:07:53<1:24:04,  3.17it/s]
 96%|█████████▌| 355505/371472 [6:07:54<1:20:45,  3.30it/s]
 96%|█████████▌| 355506/371472 [6:07:54<1:22:51,  3.21it/s]
 96%|█████████▌| 355507/371472 [6:07:54<1:18:34,  3.39it/s]
 96%|█████████▌| 355508/371472 [6:07:54<1:17:23,  3.44it/s]
 96%|█████████▌| 355509/371472 [6:07:55<1:14:41,  3.56it/s]
 96%|█████████▌| 355510/371472 [6:07:55<1:11:56,  3.70it/s]
 96%|█████████▌| 355511/371472 [6:07:55<1:11:35,  3.72it/s]
 96%|█████████▌| 355512/371472 [6:07:55<1:10:57,  3.75it/s]
 96%|█████████▌| 355513/371472 [6:07:56<1:12:01,  3.69it/s]
 96%|█████████▌| 355514/371472 [6:07:56<1:10:54,  3.75it/s]
 96%|█████████▌| 355515/371472 [6:07:56<1:11:02,  3.74it/s]
 96%|█████████▌| 355516/371472 [6:07:56<1:09:57,  3.80it/s]
 96%|█████████▌| 355517/371472 [6:07:57<1:10:03,  3.80it/s]
 96%|█████████▌| 355518/371472 [6:07:57<1:16:04,  3.50it/s]
 96%|█████████▌| 355519/371472 [6:07:57<1:15:13,  3.53it/s]
 96%|█████████▌| 355520/371472 [6:07:58<1:12:15,  3.68it/s]
                                                           
{'loss': 2.5115, 'learning_rate': 1.3866922364196598e-07, 'epoch': 15.31}
+
 96%|█████████▌| 355520/371472 [6:07:58<1:12:15,  3.68it/s]
 96%|█████████▌| 355521/371472 [6:07:58<1:10:26,  3.77it/s]
 96%|█████████▌| 355522/371472 [6:07:58<1:10:33,  3.77it/s]
 96%|█████████▌| 355523/371472 [6:07:58<1:12:02,  3.69it/s]
 96%|█████████▌| 355524/371472 [6:07:59<1:11:10,  3.73it/s]
 96%|█████████▌| 355525/371472 [6:07:59<1:13:49,  3.60it/s]
 96%|█████████▌| 355526/371472 [6:07:59<1:14:58,  3.54it/s]
 96%|█████████▌| 355527/371472 [6:08:00<1:18:56,  3.37it/s]
 96%|█████████▌| 355528/371472 [6:08:00<1:17:07,  3.45it/s]
 96%|█████████▌| 355529/371472 [6:08:00<1:18:25,  3.39it/s]
 96%|█████████▌| 355530/371472 [6:08:00<1:15:41,  3.51it/s]
 96%|█████████▌| 355531/371472 [6:08:01<1:12:55,  3.64it/s]
 96%|█████████▌| 355532/371472 [6:08:01<1:11:01,  3.74it/s]
 96%|█████████▌| 355533/371472 [6:08:01<1:10:57,  3.74it/s]
 96%|█████████▌| 355534/371472 [6:08:01<1:11:34,  3.71it/s]
 96%|█████████▌| 355535/371472 [6:08:02<1:09:24,  3.83it/s]
 96%|████████��▌| 355536/371472 [6:08:02<1:10:17,  3.78it/s]
 96%|█████████▌| 355537/371472 [6:08:02<1:13:20,  3.62it/s]
 96%|█████████▌| 355538/371472 [6:08:03<1:16:16,  3.48it/s]
 96%|█████████▌| 355539/371472 [6:08:03<1:14:39,  3.56it/s]
 96%|█████████▌| 355540/371472 [6:08:03<1:17:37,  3.42it/s]
                                                           
{'loss': 2.445, 'learning_rate': 1.3862074166648713e-07, 'epoch': 15.31}
+
 96%|█████████▌| 355540/371472 [6:08:03<1:17:37,  3.42it/s]
 96%|█████████▌| 355541/371472 [6:08:03<1:15:02,  3.54it/s]
 96%|█████████▌| 355542/371472 [6:08:04<1:16:27,  3.47it/s]
 96%|█████████▌| 355543/371472 [6:08:04<1:20:19,  3.30it/s]
 96%|█████████▌| 355544/371472 [6:08:04<1:23:41,  3.17it/s]
 96%|█████████▌| 355545/371472 [6:08:05<1:20:18,  3.31it/s]
 96%|█████████▌| 355546/371472 [6:08:05<1:17:24,  3.43it/s]
 96%|█████████▌| 355547/371472 [6:08:05<1:15:23,  3.52it/s]
 96%|█████████▌| 355548/371472 [6:08:06<1:14:41,  3.55it/s]
 96%|█████████▌| 355549/371472 [6:08:06<1:14:25,  3.57it/s]
 96%|█████████▌| 355550/371472 [6:08:06<1:12:16,  3.67it/s]
 96%|█████████▌| 355551/371472 [6:08:06<1:11:38,  3.70it/s]
 96%|█████████▌| 355552/371472 [6:08:07<1:11:14,  3.72it/s]
 96%|█████████▌| 355553/371472 [6:08:07<1:12:25,  3.66it/s]
 96%|█████████▌| 355554/371472 [6:08:07<1:13:53,  3.59it/s]
 96%|█████████▌| 355555/371472 [6:08:07<1:13:24,  3.61it/s]
 96%|█████████▌| 355556/371472 [6:08:08<1:12:27,  3.66it/s]
 96%|█████████▌| 355557/371472 [6:08:08<1:17:05,  3.44it/s]
 96%|█████████▌| 355558/371472 [6:08:08<1:15:01,  3.54it/s]
 96%|█████████▌| 355559/371472 [6:08:09<1:18:18,  3.39it/s]
 96%|█████████▌| 355560/371472 [6:08:09<1:16:34,  3.46it/s]
                                                           
{'loss': 2.6425, 'learning_rate': 1.385722596910082e-07, 'epoch': 15.31}
+
 96%|█████████▌| 355560/371472 [6:08:09<1:16:34,  3.46it/s]
 96%|█████████▌| 355561/371472 [6:08:09<1:20:14,  3.30it/s]
 96%|█████████▌| 355562/371472 [6:08:09<1:18:01,  3.40it/s]
 96%|█████████▌| 355563/371472 [6:08:10<1:14:47,  3.55it/s]
 96%|█████████▌| 355564/371472 [6:08:10<1:20:18,  3.30it/s]
 96%|█████████▌| 355565/371472 [6:08:10<1:20:38,  3.29it/s]
 96%|█████████▌| 355566/371472 [6:08:11<1:20:23,  3.30it/s]
 96%|█████████▌| 355567/371472 [6:08:11<1:17:13,  3.43it/s]
 96%|█████████▌| 355568/371472 [6:08:11<1:16:04,  3.48it/s]
 96%|█████████▌| 355569/371472 [6:08:12<1:15:35,  3.51it/s]
 96%|█████████▌| 355570/371472 [6:08:12<1:18:33,  3.37it/s]
 96%|█████████▌| 355571/371472 [6:08:12<1:19:33,  3.33it/s]
 96%|█████████▌| 355572/371472 [6:08:12<1:18:07,  3.39it/s]
 96%|█████████▌| 355573/371472 [6:08:13<1:15:48,  3.50it/s]
 96%|█████████▌| 355574/371472 [6:08:13<1:18:18,  3.38it/s]
 96%|█████████▌| 355575/371472 [6:08:13<1:15:25,  3.51it/s]
 96%|█████████▌| 355576/371472 [6:08:14<1:13:50,  3.59it/s]
 96%|█████████▌| 355577/371472 [6:08:14<1:16:32,  3.46it/s]
 96%|█████████▌| 355578/371472 [6:08:14<1:13:35,  3.60it/s]
 96%|█████████▌| 355579/371472 [6:08:14<1:20:16,  3.30it/s]
 96%|█████████▌| 355580/371472 [6:08:15<1:17:44,  3.41it/s]
                                                           
{'loss': 2.7245, 'learning_rate': 1.3852377771552935e-07, 'epoch': 15.32}
+
 96%|█████████▌| 355580/371472 [6:08:15<1:17:44,  3.41it/s]
 96%|█████████▌| 355581/371472 [6:08:15<1:14:23,  3.56it/s]
 96%|█████████▌| 355582/371472 [6:08:15<1:12:25,  3.66it/s]
 96%|█████████▌| 355583/371472 [6:08:16<1:13:06,  3.62it/s]
 96%|█████████▌| 355584/371472 [6:08:16<1:12:18,  3.66it/s]
 96%|█████████▌| 355585/371472 [6:08:16<1:11:57,  3.68it/s]
 96%|█████████▌| 355586/371472 [6:08:16<1:11:42,  3.69it/s]
 96%|█████████▌| 355587/371472 [6:08:17<1:09:15,  3.82it/s]
 96%|█████████▌| 355588/371472 [6:08:17<1:09:45,  3.80it/s]
 96%|█████████▌| 355589/371472 [6:08:17<1:10:56,  3.73it/s]
 96%|█████████▌| 355590/371472 [6:08:17<1:17:44,  3.40it/s]
 96%|█████████▌| 355591/371472 [6:08:18<1:15:42,  3.50it/s]
 96%|██████��██▌| 355592/371472 [6:08:18<1:15:07,  3.52it/s]
 96%|█████████▌| 355593/371472 [6:08:18<1:13:39,  3.59it/s]
 96%|█████████▌| 355594/371472 [6:08:19<1:13:53,  3.58it/s]
 96%|█████████▌| 355595/371472 [6:08:19<1:18:04,  3.39it/s]
 96%|█████████▌| 355596/371472 [6:08:19<1:21:28,  3.25it/s]
 96%|█████████▌| 355597/371472 [6:08:20<1:21:20,  3.25it/s]
 96%|█████████▌| 355598/371472 [6:08:20<1:16:53,  3.44it/s]
 96%|█████████▌| 355599/371472 [6:08:20<1:20:50,  3.27it/s]
 96%|█████████▌| 355600/371472 [6:08:20<1:16:07,  3.47it/s]
                                                           
{'loss': 2.656, 'learning_rate': 1.384752957400504e-07, 'epoch': 15.32}
+
 96%|█████████▌| 355600/371472 [6:08:20<1:16:07,  3.47it/s]
 96%|█████████▌| 355601/371472 [6:08:21<1:14:24,  3.55it/s]
 96%|█████████▌| 355602/371472 [6:08:21<1:31:08,  2.90it/s]
 96%|█████████▌| 355603/371472 [6:08:21<1:28:04,  3.00it/s]
 96%|█████████▌| 355604/371472 [6:08:22<1:22:56,  3.19it/s]
 96%|█████████▌| 355605/371472 [6:08:22<1:19:15,  3.34it/s]
 96%|█████████▌| 355606/371472 [6:08:22<1:21:03,  3.26it/s]
 96%|█████████▌| 355607/371472 [6:08:23<1:17:28,  3.41it/s]
 96%|█████████▌| 355608/371472 [6:08:23<1:14:30,  3.55it/s]
 96%|█████████▌| 355609/371472 [6:08:23<1:13:43,  3.59it/s]
 96%|█████████▌| 355610/371472 [6:08:23<1:15:49,  3.49it/s]
 96%|█████████▌| 355611/371472 [6:08:24<1:17:14,  3.42it/s]
 96%|█████████▌| 355612/371472 [6:08:24<1:19:58,  3.30it/s]
 96%|█████████▌| 355613/371472 [6:08:24<1:19:14,  3.34it/s]
 96%|█████████▌| 355614/371472 [6:08:25<1:17:49,  3.40it/s]
 96%|█████████▌| 355615/371472 [6:08:25<1:17:25,  3.41it/s]
 96%|█████████▌| 355616/371472 [6:08:25<1:18:43,  3.36it/s]
 96%|█████████▌| 355617/371472 [6:08:26<1:18:55,  3.35it/s]
 96%|█████████▌| 355618/371472 [6:08:26<1:15:47,  3.49it/s]
 96%|█████████▌| 355619/371472 [6:08:26<1:15:29,  3.50it/s]
 96%|█████████▌| 355620/371472 [6:08:26<1:19:31,  3.32it/s]
                                                           
{'loss': 2.7099, 'learning_rate': 1.3842681376457157e-07, 'epoch': 15.32}
+
 96%|█████████▌| 355620/371472 [6:08:26<1:19:31,  3.32it/s]
 96%|█████████▌| 355621/371472 [6:08:27<1:18:29,  3.37it/s]
 96%|█████████▌| 355622/371472 [6:08:27<1:15:29,  3.50it/s]
 96%|█████████▌| 355623/371472 [6:08:27<1:15:54,  3.48it/s]
 96%|█████████▌| 355624/371472 [6:08:28<1:17:30,  3.41it/s]
 96%|█████████▌| 355625/371472 [6:08:28<1:17:32,  3.41it/s]
 96%|█████████▌| 355626/371472 [6:08:28<1:13:41,  3.58it/s]
 96%|█████████▌| 355627/371472 [6:08:28<1:16:24,  3.46it/s]
 96%|█████████▌| 355628/371472 [6:08:29<1:18:44,  3.35it/s]
 96%|█████████▌| 355629/371472 [6:08:29<1:19:13,  3.33it/s]
 96%|█████████▌| 355630/371472 [6:08:29<1:17:24,  3.41it/s]
 96%|█████████▌| 355631/371472 [6:08:30<1:18:01,  3.38it/s]
 96%|█████████▌| 355632/371472 [6:08:30<1:17:33,  3.40it/s]
 96%|█████████▌| 355633/371472 [6:08:30<1:15:27,  3.50it/s]
 96%|█████████▌| 355634/371472 [6:08:30<1:18:20,  3.37it/s]
 96%|█████████▌| 355635/371472 [6:08:31<1:19:16,  3.33it/s]
 96%|█████████▌| 355636/371472 [6:08:31<1:16:54,  3.43it/s]
 96%|█████████▌| 355637/371472 [6:08:31<1:18:28,  3.36it/s]
 96%|█████████▌| 355638/371472 [6:08:32<1:23:36,  3.16it/s]
 96%|█████████▌| 355639/371472 [6:08:32<1:20:19,  3.29it/s]
 96%|█████████▌| 355640/371472 [6:08:32<1:26:57,  3.03it/s]
                                                           
{'loss': 2.5826, 'learning_rate': 1.3837833178909262e-07, 'epoch': 15.32}
+
 96%|█████████▌| 355640/371472 [6:08:32<1:26:57,  3.03it/s]
 96%|█████████▌| 355641/371472 [6:08:33<1:21:08,  3.25it/s]
 96%|█████████▌| 355642/371472 [6:08:33<1:16:29,  3.45it/s]
 96%|█████████▌| 355643/371472 [6:08:33<1:15:59,  3.47it/s]
 96%|█████████▌| 355644/371472 [6:08:33<1:12:55,  3.62it/s]
 96%|█████████▌| 355645/371472 [6:08:34<1:14:30,  3.54it/s]
 96%|█████████▌| 355646/371472 [6:08:34<1:13:50,  3.57it/s]
 96%|█████████▌| 355647/371472 [6:08:34<1:20:24,  3.28it/s]
 96%|████��████▌| 355648/371472 [6:08:35<1:21:40,  3.23it/s]
 96%|█████████▌| 355649/371472 [6:08:35<1:22:01,  3.21it/s]
 96%|█████████▌| 355650/371472 [6:08:35<1:23:06,  3.17it/s]
 96%|█████████▌| 355651/371472 [6:08:36<1:18:39,  3.35it/s]
 96%|█████████▌| 355652/371472 [6:08:36<1:15:09,  3.51it/s]
 96%|█████████▌| 355653/371472 [6:08:36<1:12:12,  3.65it/s]
 96%|█████████▌| 355654/371472 [6:08:36<1:12:37,  3.63it/s]
 96%|█████████▌| 355655/371472 [6:08:37<1:11:04,  3.71it/s]
 96%|█████████▌| 355656/371472 [6:08:37<1:09:59,  3.77it/s]
 96%|█████████▌| 355657/371472 [6:08:37<1:12:47,  3.62it/s]
 96%|█████████▌| 355658/371472 [6:08:37<1:11:34,  3.68it/s]
 96%|█████████▌| 355659/371472 [6:08:38<1:13:25,  3.59it/s]
 96%|█████████▌| 355660/371472 [6:08:38<1:10:22,  3.74it/s]
                                                           
{'loss': 2.7634, 'learning_rate': 1.3832984981361377e-07, 'epoch': 15.32}
+
 96%|█████████▌| 355660/371472 [6:08:38<1:10:22,  3.74it/s]
 96%|█████████▌| 355661/371472 [6:08:38<1:18:03,  3.38it/s]
 96%|█████████▌| 355662/371472 [6:08:39<1:15:18,  3.50it/s]
 96%|█████████▌| 355663/371472 [6:08:39<1:14:55,  3.52it/s]
 96%|█████████▌| 355664/371472 [6:08:39<1:19:37,  3.31it/s]
 96%|█████████▌| 355665/371472 [6:08:40<1:19:04,  3.33it/s]
 96%|█████████▌| 355666/371472 [6:08:40<1:16:45,  3.43it/s]
 96%|█████████▌| 355667/371472 [6:08:40<1:15:21,  3.50it/s]
 96%|█████████▌| 355668/371472 [6:08:40<1:17:37,  3.39it/s]
 96%|█████████▌| 355669/371472 [6:08:41<1:17:38,  3.39it/s]
 96%|█████████▌| 355670/371472 [6:08:41<1:15:10,  3.50it/s]
 96%|█████████▌| 355671/371472 [6:08:41<1:14:59,  3.51it/s]
 96%|█████████▌| 355672/371472 [6:08:42<1:16:41,  3.43it/s]
 96%|█████████▌| 355673/371472 [6:08:42<1:13:40,  3.57it/s]
 96%|█████████▌| 355674/371472 [6:08:42<1:14:27,  3.54it/s]
 96%|█████████▌| 355675/371472 [6:08:42<1:24:44,  3.11it/s]
 96%|█████████▌| 355676/371472 [6:08:43<1:20:17,  3.28it/s]
 96%|█████████▌| 355677/371472 [6:08:43<1:23:53,  3.14it/s]
 96%|█████████▌| 355678/371472 [6:08:43<1:22:44,  3.18it/s]
 96%|█████████▌| 355679/371472 [6:08:44<1:18:50,  3.34it/s]
 96%|█████████▌| 355680/371472 [6:08:44<1:15:26,  3.49it/s]
                                                           
{'loss': 2.6542, 'learning_rate': 1.3828136783813484e-07, 'epoch': 15.32}
+
 96%|█████████▌| 355680/371472 [6:08:44<1:15:26,  3.49it/s]
 96%|█████████▌| 355681/371472 [6:08:44<1:14:31,  3.53it/s]
 96%|█████████▌| 355682/371472 [6:08:44<1:15:54,  3.47it/s]
 96%|█████████▌| 355683/371472 [6:08:45<1:19:42,  3.30it/s]
 96%|█████████▌| 355684/371472 [6:08:45<1:16:46,  3.43it/s]
 96%|█████████▌| 355685/371472 [6:08:45<1:15:44,  3.47it/s]
 96%|█████████▌| 355686/371472 [6:08:46<1:13:05,  3.60it/s]
 96%|█████████▌| 355687/371472 [6:08:46<1:11:28,  3.68it/s]
 96%|█████████▌| 355688/371472 [6:08:46<1:12:03,  3.65it/s]
 96%|█████████▌| 355689/371472 [6:08:46<1:10:43,  3.72it/s]
 96%|█████████▌| 355690/371472 [6:08:47<1:18:35,  3.35it/s]
 96%|█████████▌| 355691/371472 [6:08:47<1:17:54,  3.38it/s]
 96%|█████████▌| 355692/371472 [6:08:47<1:18:17,  3.36it/s]
 96%|█████████▌| 355693/371472 [6:08:48<1:22:29,  3.19it/s]
 96%|█████████▌| 355694/371472 [6:08:48<1:17:59,  3.37it/s]
 96%|█████████▌| 355695/371472 [6:08:48<1:17:40,  3.38it/s]
 96%|█████████▌| 355696/371472 [6:08:49<1:18:08,  3.36it/s]
 96%|█████████▌| 355697/371472 [6:08:49<1:22:55,  3.17it/s]
 96%|█████████▌| 355698/371472 [6:08:49<1:19:45,  3.30it/s]
 96%|█████████▌| 355699/371472 [6:08:49<1:15:46,  3.47it/s]
 96%|█████████▌| 355700/371472 [6:08:50<1:15:12,  3.50it/s]
                                                           
{'loss': 2.5529, 'learning_rate': 1.38232885862656e-07, 'epoch': 15.32}
+
 96%|█████████▌| 355700/371472 [6:08:50<1:15:12,  3.50it/s]
 96%|█████████▌| 355701/371472 [6:08:50<1:13:45,  3.56it/s]
 96%|█████████▌| 355702/371472 [6:08:50<1:13:08,  3.59it/s]
 96%|█████████▌| 355703/371472 [6:08:51<1:12:09,  3.64it/s]
 96%|██��██████▌| 355704/371472 [6:08:51<1:10:18,  3.74it/s]
 96%|█████████▌| 355705/371472 [6:08:51<1:11:29,  3.68it/s]
 96%|█████████▌| 355706/371472 [6:08:51<1:14:41,  3.52it/s]
 96%|█████████▌| 355707/371472 [6:08:52<1:13:29,  3.58it/s]
 96%|█████████▌| 355708/371472 [6:08:52<1:12:38,  3.62it/s]
 96%|█████████▌| 355709/371472 [6:08:52<1:12:46,  3.61it/s]
 96%|█████████▌| 355710/371472 [6:08:52<1:09:45,  3.77it/s]
 96%|█████████▌| 355711/371472 [6:08:53<1:14:02,  3.55it/s]
 96%|█████████▌| 355712/371472 [6:08:53<1:13:40,  3.57it/s]
 96%|█████████▌| 355713/371472 [6:08:53<1:13:46,  3.56it/s]
 96%|█████████▌| 355714/371472 [6:08:54<1:11:40,  3.66it/s]
 96%|█████████▌| 355715/371472 [6:08:54<1:09:23,  3.78it/s]
 96%|█████████▌| 355716/371472 [6:08:54<1:13:27,  3.58it/s]
 96%|█████████▌| 355717/371472 [6:08:54<1:14:38,  3.52it/s]
 96%|█████████▌| 355718/371472 [6:08:55<1:17:28,  3.39it/s]
 96%|█████████▌| 355719/371472 [6:08:55<1:19:39,  3.30it/s]
 96%|█████████▌| 355720/371472 [6:08:55<1:22:13,  3.19it/s]
                                                           
{'loss': 2.444, 'learning_rate': 1.3818440388717704e-07, 'epoch': 15.32}
+
 96%|█████████▌| 355720/371472 [6:08:55<1:22:13,  3.19it/s]
 96%|█████████▌| 355721/371472 [6:08:56<1:25:57,  3.05it/s]
 96%|█████████▌| 355722/371472 [6:08:56<1:20:53,  3.25it/s]
 96%|█████████▌| 355723/371472 [6:08:56<1:17:38,  3.38it/s]
 96%|█████████▌| 355724/371472 [6:08:57<1:13:46,  3.56it/s]
 96%|█████████▌| 355725/371472 [6:08:57<1:13:33,  3.57it/s]
 96%|█████████▌| 355726/371472 [6:08:57<1:12:37,  3.61it/s]
 96%|█████████▌| 355727/371472 [6:08:57<1:13:10,  3.59it/s]
 96%|█████████▌| 355728/371472 [6:08:58<1:11:22,  3.68it/s]
 96%|█████████▌| 355729/371472 [6:08:58<1:11:23,  3.68it/s]
 96%|█████████▌| 355730/371472 [6:08:58<1:11:40,  3.66it/s]
 96%|█████████▌| 355731/371472 [6:08:58<1:12:19,  3.63it/s]
 96%|█████████▌| 355732/371472 [6:08:59<1:10:39,  3.71it/s]
 96%|█████████▌| 355733/371472 [6:08:59<1:10:52,  3.70it/s]
 96%|█████████▌| 355734/371472 [6:08:59<1:12:26,  3.62it/s]
 96%|█████████▌| 355735/371472 [6:09:00<1:11:16,  3.68it/s]
 96%|█████████▌| 355736/371472 [6:09:00<1:10:06,  3.74it/s]
 96%|█████████▌| 355737/371472 [6:09:00<1:11:15,  3.68it/s]
 96%|█████████▌| 355738/371472 [6:09:00<1:17:24,  3.39it/s]
 96%|█████████▌| 355739/371472 [6:09:01<1:20:39,  3.25it/s]
 96%|█████████▌| 355740/371472 [6:09:01<1:16:09,  3.44it/s]
                                                           
{'loss': 2.5641, 'learning_rate': 1.3813592191169821e-07, 'epoch': 15.32}
+
 96%|█████████▌| 355740/371472 [6:09:01<1:16:09,  3.44it/s]
 96%|█████████▌| 355741/371472 [6:09:01<1:15:31,  3.47it/s]
 96%|█████████▌| 355742/371472 [6:09:02<1:13:29,  3.57it/s]
 96%|█████████▌| 355743/371472 [6:09:02<1:27:13,  3.01it/s]
 96%|█████████▌| 355744/371472 [6:09:02<1:22:25,  3.18it/s]
 96%|█████████▌| 355745/371472 [6:09:03<1:22:47,  3.17it/s]
 96%|█████████▌| 355746/371472 [6:09:03<1:20:37,  3.25it/s]
 96%|█████████▌| 355747/371472 [6:09:03<1:18:25,  3.34it/s]
 96%|█████████▌| 355748/371472 [6:09:03<1:16:21,  3.43it/s]
 96%|█████████▌| 355749/371472 [6:09:04<1:23:41,  3.13it/s]
 96%|█████████▌| 355750/371472 [6:09:04<1:18:18,  3.35it/s]
 96%|█████████▌| 355751/371472 [6:09:04<1:16:27,  3.43it/s]
 96%|█████████▌| 355752/371472 [6:09:05<1:15:41,  3.46it/s]
 96%|█████████▌| 355753/371472 [6:09:05<1:18:47,  3.33it/s]
 96%|█████████▌| 355754/371472 [6:09:05<1:15:53,  3.45it/s]
 96%|█████████▌| 355755/371472 [6:09:06<1:13:36,  3.56it/s]
 96%|█████████▌| 355756/371472 [6:09:06<1:11:20,  3.67it/s]
 96%|█████████▌| 355757/371472 [6:09:06<1:12:37,  3.61it/s]
 96%|█████████▌| 355758/371472 [6:09:06<1:12:20,  3.62it/s]
 96%|█████████▌| 355759/371472 [6:09:07<1:10:18,  3.72it/s]
 96%|█████████▌| 355760/371472 [6:09:07<1:13:31,  3.56it/s]
                                                           
{'loss': 2.7158, 'learning_rate': 1.3808743993621926e-07, 'epoch': 15.32}
+
 96%|���████████▌| 355760/371472 [6:09:07<1:13:31,  3.56it/s]
 96%|█████████▌| 355761/371472 [6:09:07<1:12:30,  3.61it/s]
 96%|█████████▌| 355762/371472 [6:09:07<1:16:22,  3.43it/s]
 96%|█████████▌| 355763/371472 [6:09:08<1:13:19,  3.57it/s]
 96%|█████████▌| 355764/371472 [6:09:08<1:12:32,  3.61it/s]
 96%|█████████▌| 355765/371472 [6:09:08<1:13:22,  3.57it/s]
 96%|█████████▌| 355766/371472 [6:09:09<1:11:43,  3.65it/s]
 96%|█████████▌| 355767/371472 [6:09:09<1:12:44,  3.60it/s]
 96%|█████████▌| 355768/371472 [6:09:09<1:13:26,  3.56it/s]
 96%|█████████▌| 355769/371472 [6:09:09<1:11:38,  3.65it/s]
 96%|█████████▌| 355770/371472 [6:09:10<1:09:36,  3.76it/s]
 96%|█████████▌| 355771/371472 [6:09:10<1:09:13,  3.78it/s]
 96%|█████████▌| 355772/371472 [6:09:10<1:08:21,  3.83it/s]
 96%|█████████▌| 355773/371472 [6:09:10<1:11:20,  3.67it/s]
 96%|█████████▌| 355774/371472 [6:09:11<1:11:15,  3.67it/s]
 96%|█████████▌| 355775/371472 [6:09:11<1:12:35,  3.60it/s]
 96%|█████████▌| 355776/371472 [6:09:11<1:11:12,  3.67it/s]
 96%|█████████▌| 355777/371472 [6:09:12<1:13:30,  3.56it/s]
 96%|█████████▌| 355778/371472 [6:09:12<1:18:54,  3.31it/s]
 96%|█████████▌| 355779/371472 [6:09:12<1:25:02,  3.08it/s]
 96%|█████████▌| 355780/371472 [6:09:13<1:24:58,  3.08it/s]
                                                           
{'loss': 2.612, 'learning_rate': 1.3803895796074033e-07, 'epoch': 15.32}
+
 96%|█████████▌| 355780/371472 [6:09:13<1:24:58,  3.08it/s]
 96%|█████████▌| 355781/371472 [6:09:13<1:19:32,  3.29it/s]
 96%|█████████▌| 355782/371472 [6:09:13<1:17:32,  3.37it/s]
 96%|█████████▌| 355783/371472 [6:09:13<1:15:12,  3.48it/s]
 96%|█████████▌| 355784/371472 [6:09:14<1:13:25,  3.56it/s]
 96%|█████████▌| 355785/371472 [6:09:14<1:11:29,  3.66it/s]
 96%|█████████▌| 355786/371472 [6:09:14<1:10:00,  3.73it/s]
 96%|█████████▌| 355787/371472 [6:09:14<1:11:18,  3.67it/s]
 96%|█████████▌| 355788/371472 [6:09:15<1:11:42,  3.65it/s]
 96%|█████████▌| 355789/371472 [6:09:15<1:11:44,  3.64it/s]
 96%|█████████▌| 355790/371472 [6:09:15<1:13:13,  3.57it/s]
 96%|█████████▌| 355791/371472 [6:09:16<1:11:47,  3.64it/s]
 96%|█████████▌| 355792/371472 [6:09:16<1:10:58,  3.68it/s]
 96%|█████████▌| 355793/371472 [6:09:16<1:09:40,  3.75it/s]
 96%|█████████▌| 355794/371472 [6:09:16<1:09:41,  3.75it/s]
 96%|█████████▌| 355795/371472 [6:09:17<1:08:46,  3.80it/s]
 96%|█████████▌| 355796/371472 [6:09:17<1:11:08,  3.67it/s]
 96%|█████████▌| 355797/371472 [6:09:17<1:12:13,  3.62it/s]
 96%|█████████▌| 355798/371472 [6:09:17<1:10:50,  3.69it/s]
 96%|█████████▌| 355799/371472 [6:09:18<1:09:26,  3.76it/s]
 96%|█████████▌| 355800/371472 [6:09:18<1:09:23,  3.76it/s]
                                                           
{'loss': 2.4557, 'learning_rate': 1.3799047598526148e-07, 'epoch': 15.32}
+
 96%|█████████▌| 355800/371472 [6:09:18<1:09:23,  3.76it/s]
 96%|█████████▌| 355801/371472 [6:09:18<1:08:44,  3.80it/s]
 96%|█████████▌| 355802/371472 [6:09:19<1:10:21,  3.71it/s]
 96%|█████████▌| 355803/371472 [6:09:19<1:14:36,  3.50it/s]
 96%|█████████▌| 355804/371472 [6:09:19<1:13:20,  3.56it/s]
 96%|█████████▌| 355805/371472 [6:09:19<1:11:36,  3.65it/s]
 96%|█████████▌| 355806/371472 [6:09:20<1:11:10,  3.67it/s]
 96%|█████████▌| 355807/371472 [6:09:20<1:11:16,  3.66it/s]
 96%|█████████▌| 355808/371472 [6:09:20<1:09:42,  3.74it/s]
 96%|█████████▌| 355809/371472 [6:09:20<1:10:33,  3.70it/s]
 96%|█████████▌| 355810/371472 [6:09:21<1:10:18,  3.71it/s]
 96%|█████████▌| 355811/371472 [6:09:21<1:10:39,  3.69it/s]
 96%|█████████▌| 355812/371472 [6:09:21<1:10:02,  3.73it/s]
 96%|█████████▌| 355813/371472 [6:09:22<1:08:17,  3.82it/s]
 96%|█████████▌| 355814/371472 [6:09:22<1:09:33,  3.75it/s]
 96%|█████████▌| 355815/371472 [6:09:22<1:11:13,  3.66it/s]
 96%|█████████▌| 355816/371472 [6:09:22<1:11:59,  3.62it/s]
 96%|█████████▌| 355817/371472 [6:09:23<1:10:01,  3.73it/s]
 96%|█████████▌| 355818/371472 [6:09:23<1:12:13,  3.61it/s]
 96%|█████████▌| 355819/371472 [6:09:23<1:15:53,  3.44it/s]
 96%|█████████▌| 355820/371472 [6:09:23<1:14:31,  3.50it/s]
                                                           
{'loss': 2.43, 'learning_rate': 1.3794199400978255e-07, 'epoch': 15.33}
+
 96%|█████████▌| 355820/371472 [6:09:23<1:14:31,  3.50it/s]
 96%|█████████▌| 355821/371472 [6:09:24<1:14:57,  3.48it/s]
 96%|█████████▌| 355822/371472 [6:09:24<1:11:37,  3.64it/s]
 96%|█████████▌| 355823/371472 [6:09:24<1:15:27,  3.46it/s]
 96%|█████████▌| 355824/371472 [6:09:25<1:16:16,  3.42it/s]
 96%|█████████▌| 355825/371472 [6:09:25<1:17:50,  3.35it/s]
 96%|█████████▌| 355826/371472 [6:09:25<1:13:57,  3.53it/s]
 96%|█████████▌| 355827/371472 [6:09:25<1:13:58,  3.53it/s]
 96%|█████████▌| 355828/371472 [6:09:26<1:13:26,  3.55it/s]
 96%|█████████▌| 355829/371472 [6:09:26<1:12:35,  3.59it/s]
 96%|█████████▌| 355830/371472 [6:09:26<1:14:18,  3.51it/s]
 96%|█████████▌| 355831/371472 [6:09:27<1:14:55,  3.48it/s]
 96%|█████████▌| 355832/371472 [6:09:27<1:14:47,  3.49it/s]
 96%|█████████▌| 355833/371472 [6:09:27<1:14:17,  3.51it/s]
 96%|█████████▌| 355834/371472 [6:09:27<1:12:39,  3.59it/s]
 96%|█████████▌| 355835/371472 [6:09:28<1:16:57,  3.39it/s]
 96%|█████████▌| 355836/371472 [6:09:28<1:17:07,  3.38it/s]
 96%|█████████▌| 355837/371472 [6:09:28<1:15:39,  3.44it/s]
 96%|█████████▌| 355838/371472 [6:09:29<1:15:06,  3.47it/s]
 96%|█████████▌| 355839/371472 [6:09:29<1:13:42,  3.53it/s]
 96%|█████████▌| 355840/371472 [6:09:29<1:11:10,  3.66it/s]
                                                           
{'loss': 2.7721, 'learning_rate': 1.378935120343037e-07, 'epoch': 15.33}
+
 96%|█████████▌| 355840/371472 [6:09:29<1:11:10,  3.66it/s]
 96%|█████████▌| 355841/371472 [6:09:29<1:10:31,  3.69it/s]
 96%|█████████▌| 355842/371472 [6:09:30<1:13:03,  3.57it/s]
 96%|█████████▌| 355843/371472 [6:09:30<1:11:15,  3.66it/s]
 96%|█████████▌| 355844/371472 [6:09:30<1:09:34,  3.74it/s]
 96%|█████████▌| 355845/371472 [6:09:31<1:14:00,  3.52it/s]
 96%|█████████▌| 355846/371472 [6:09:31<1:12:00,  3.62it/s]
 96%|█████████▌| 355847/371472 [6:09:31<1:10:55,  3.67it/s]
 96%|█████████▌| 355848/371472 [6:09:31<1:12:28,  3.59it/s]
 96%|█████████▌| 355849/371472 [6:09:32<1:11:00,  3.67it/s]
 96%|█████████▌| 355850/371472 [6:09:32<1:12:24,  3.60it/s]
 96%|█████████▌| 355851/371472 [6:09:32<1:10:49,  3.68it/s]
 96%|█████████▌| 355852/371472 [6:09:32<1:10:20,  3.70it/s]
 96%|█████████▌| 355853/371472 [6:09:33<1:14:14,  3.51it/s]
 96%|█████████▌| 355854/371472 [6:09:33<1:10:30,  3.69it/s]
 96%|█████████▌| 355855/371472 [6:09:33<1:10:53,  3.67it/s]
 96%|█████████▌| 355856/371472 [6:09:34<1:10:51,  3.67it/s]
 96%|█████████▌| 355857/371472 [6:09:34<1:13:49,  3.53it/s]
 96%|█████████▌| 355858/371472 [6:09:34<1:13:01,  3.56it/s]
 96%|█████████▌| 355859/371472 [6:09:34<1:14:24,  3.50it/s]
 96%|█████████▌| 355860/371472 [6:09:35<1:12:19,  3.60it/s]
                                                           
{'loss': 2.6042, 'learning_rate': 1.3784503005882475e-07, 'epoch': 15.33}
+
 96%|█████████▌| 355860/371472 [6:09:35<1:12:19,  3.60it/s]
 96%|█████████▌| 355861/371472 [6:09:35<1:11:13,  3.65it/s]
 96%|█████████▌| 355862/371472 [6:09:35<1:17:15,  3.37it/s]
 96%|█████████▌| 355863/371472 [6:09:36<1:16:43,  3.39it/s]
 96%|█████████▌| 355864/371472 [6:09:36<1:15:01,  3.47it/s]
 96%|█████████▌| 355865/371472 [6:09:36<1:15:55,  3.43it/s]
 96%|█████████▌| 355866/371472 [6:09:36<1:13:28,  3.54it/s]
 96%|█████████▌| 355867/371472 [6:09:37<1:11:27,  3.64it/s]
 96%|█████████▌| 355868/371472 [6:09:37<1:12:30,  3.59it/s]
 96%|█████████▌| 355869/371472 [6:09:37<1:13:05,  3.56it/s]
 96%|█████████▌| 355870/371472 [6:09:38<1:13:30,  3.54it/s]
 96%|█████████▌| 355871/371472 [6:09:38<1:12:50,  3.57it/s]
 96%|█████████▌| 355872/371472 [6:09:38<1:10:02,  3.71it/s]
 96%|█████████▌| 355873/371472 [6:09:38<1:07:20,  3.86it/s]
 96%|█████████▌| 355874/371472 [6:09:39<1:11:42,  3.63it/s]
 96%|█████████▌| 355875/371472 [6:09:39<1:16:38,  3.39it/s]
 96%|█████████▌| 355876/371472 [6:09:39<1:15:07,  3.46it/s]
 96%|█████████▌| 355877/371472 [6:09:40<1:24:14,  3.09it/s]
 96%|█████████▌| 355878/371472 [6:09:40<1:22:11,  3.16it/s]
 96%|█████████▌| 355879/371472 [6:09:40<1:24:19,  3.08it/s]
 96%|█████████▌| 355880/371472 [6:09:41<1:20:01,  3.25it/s]
                                                           
{'loss': 2.6828, 'learning_rate': 1.3779654808334593e-07, 'epoch': 15.33}
+
 96%|█████████▌| 355880/371472 [6:09:41<1:20:01,  3.25it/s]
 96%|█████████▌| 355881/371472 [6:09:41<1:18:22,  3.32it/s]
 96%|█████████▌| 355882/371472 [6:09:41<1:22:39,  3.14it/s]
 96%|█████████▌| 355883/371472 [6:09:42<1:21:11,  3.20it/s]
 96%|█████████▌| 355884/371472 [6:09:42<1:15:42,  3.43it/s]
 96%|█████████▌| 355885/371472 [6:09:42<1:16:28,  3.40it/s]
 96%|█████████▌| 355886/371472 [6:09:42<1:14:55,  3.47it/s]
 96%|█████████▌| 355887/371472 [6:09:43<1:20:35,  3.22it/s]
 96%|█████████▌| 355888/371472 [6:09:43<1:23:13,  3.12it/s]
 96%|█████████▌| 355889/371472 [6:09:43<1:19:16,  3.28it/s]
 96%|█████████▌| 355890/371472 [6:09:44<1:17:15,  3.36it/s]
 96%|█████████▌| 355891/371472 [6:09:44<1:17:14,  3.36it/s]
 96%|█████████▌| 355892/371472 [6:09:44<1:13:46,  3.52it/s]
 96%|█████████▌| 355893/371472 [6:09:44<1:12:17,  3.59it/s]
 96%|█████████▌| 355894/371472 [6:09:45<1:12:41,  3.57it/s]
 96%|█████████▌| 355895/371472 [6:09:45<1:14:32,  3.48it/s]
 96%|█████████▌| 355896/371472 [6:09:45<1:13:43,  3.52it/s]
 96%|█████████▌| 355897/371472 [6:09:46<1:13:32,  3.53it/s]
 96%|█████████▌| 355898/371472 [6:09:46<1:10:58,  3.66it/s]
 96%|█████████▌| 355899/371472 [6:09:46<1:10:49,  3.66it/s]
 96%|█████████▌| 355900/371472 [6:09:46<1:12:56,  3.56it/s]
                                                           
{'loss': 2.6199, 'learning_rate': 1.3774806610786697e-07, 'epoch': 15.33}
+
 96%|█████████▌| 355900/371472 [6:09:46<1:12:56,  3.56it/s]
 96%|█████████▌| 355901/371472 [6:09:47<1:23:39,  3.10it/s]
 96%|█████████▌| 355902/371472 [6:09:47<1:18:44,  3.30it/s]
 96%|█████████▌| 355903/371472 [6:09:47<1:17:00,  3.37it/s]
 96%|█████████▌| 355904/371472 [6:09:48<1:16:36,  3.39it/s]
 96%|█████████▌| 355905/371472 [6:09:48<1:16:47,  3.38it/s]
 96%|█████████▌| 355906/371472 [6:09:48<1:17:45,  3.34it/s]
 96%|█████████▌| 355907/371472 [6:09:49<1:16:49,  3.38it/s]
 96%|█████████▌| 355908/371472 [6:09:49<1:14:28,  3.48it/s]
 96%|█████████▌| 355909/371472 [6:09:49<1:15:44,  3.42it/s]
 96%|█████████▌| 355910/371472 [6:09:49<1:12:15,  3.59it/s]
 96%|█████████▌| 355911/371472 [6:09:50<1:17:31,  3.35it/s]
 96%|█████████▌| 355912/371472 [6:09:50<1:13:23,  3.53it/s]
 96%|█████████▌| 355913/371472 [6:09:50<1:10:45,  3.66it/s]
 96%|█████████▌| 355914/371472 [6:09:50<1:09:39,  3.72it/s]
 96%|█████████▌| 355915/371472 [6:09:51<1:16:04,  3.41it/s]
 96%|█████████▌| 355916/371472 [6:09:51<1:16:01,  3.41it/s]
 96%|█████████▌| 355917/371472 [6:09:51<1:22:22,  3.15it/s]
 96%|█████████▌| 355918/371472 [6:09:52<1:21:42,  3.17it/s]
 96%|█████████▌| 355919/371472 [6:09:52<1:20:11,  3.23it/s]
 96%|█████████▌| 355920/371472 [6:09:52<1:19:06,  3.28it/s]
                                                           
{'loss': 2.6048, 'learning_rate': 1.376995841323881e-07, 'epoch': 15.33}
+
 96%|█████████▌| 355920/371472 [6:09:52<1:19:06,  3.28it/s]
 96%|█████████▌| 355921/371472 [6:09:53<1:16:02,  3.41it/s]
 96%|█████████▌| 355922/371472 [6:09:53<1:20:06,  3.24it/s]
 96%|█████████▌| 355923/371472 [6:09:53<1:16:03,  3.41it/s]
 96%|█████████▌| 355924/371472 [6:09:53<1:13:45,  3.51it/s]
 96%|█████████▌| 355925/371472 [6:09:54<1:16:36,  3.38it/s]
 96%|█████████▌| 355926/371472 [6:09:54<1:16:54,  3.37it/s]
 96%|█████████▌| 355927/371472 [6:09:54<1:14:54,  3.46it/s]
 96%|█████████▌| 355928/371472 [6:09:55<1:11:40,  3.61it/s]
 96%|█████████▌| 355929/371472 [6:09:55<1:19:13,  3.27it/s]
 96%|█████████▌| 355930/371472 [6:09:55<1:20:05,  3.23it/s]
 96%|█████████▌| 355931/371472 [6:09:56<1:15:49,  3.42it/s]
 96%|█████████▌| 355932/371472 [6:09:56<1:13:42,  3.51it/s]
 96%|█████████▌| 355933/371472 [6:09:56<1:12:03,  3.59it/s]
 96%|█████████▌| 355934/371472 [6:09:56<1:12:05,  3.59it/s]
 96%|█████████▌| 355935/371472 [6:09:57<1:13:12,  3.54it/s]
 96%|█████████▌| 355936/371472 [6:09:57<1:14:43,  3.46it/s]
 96%|█████████▌| 355937/371472 [6:09:57<1:12:12,  3.59it/s]
 96%|█████████▌| 355938/371472 [6:09:57<1:11:08,  3.64it/s]
 96%|█████████▌| 355939/371472 [6:09:58<1:13:57,  3.50it/s]
 96%|█████████▌| 355940/371472 [6:09:58<1:13:33,  3.52it/s]
                                                           
{'loss': 2.5047, 'learning_rate': 1.376511021569092e-07, 'epoch': 15.33}
+
 96%|█████████▌| 355940/371472 [6:09:58<1:13:33,  3.52it/s]
 96%|█████████▌| 355941/371472 [6:09:58<1:12:16,  3.58it/s]
 96%|█████████▌| 355942/371472 [6:09:59<1:12:35,  3.57it/s]
 96%|█████████▌| 355943/371472 [6:09:59<1:11:34,  3.62it/s]
 96%|█████████▌| 355944/371472 [6:09:59<1:12:04,  3.59it/s]
 96%|█████████▌| 355945/371472 [6:09:59<1:10:29,  3.67it/s]
 96%|█████████▌| 355946/371472 [6:10:00<1:11:49,  3.60it/s]
 96%|█████████▌| 355947/371472 [6:10:00<1:09:54,  3.70it/s]
 96%|█████████▌| 355948/371472 [6:10:00<1:08:42,  3.77it/s]
 96%|█████████▌| 355949/371472 [6:10:01<1:24:18,  3.07it/s]
 96%|█████████▌| 355950/371472 [6:10:01<1:18:36,  3.29it/s]
 96%|█████████▌| 355951/371472 [6:10:01<1:20:53,  3.20it/s]
 96%|█████████▌| 355952/371472 [6:10:02<1:16:37,  3.38it/s]
 96%|█████████▌| 355953/371472 [6:10:02<1:14:29,  3.47it/s]
 96%|█████████▌| 355954/371472 [6:10:02<1:13:13,  3.53it/s]
 96%|█████████▌| 355955/371472 [6:10:02<1:11:03,  3.64it/s]
 96%|█████████▌| 355956/371472 [6:10:03<1:25:25,  3.03it/s]
 96%|█████████▌| 355957/371472 [6:10:03<1:20:55,  3.20it/s]
 96%|█████████▌| 355958/371472 [6:10:03<1:17:10,  3.35it/s]
 96%|█████████▌| 355959/371472 [6:10:04<1:15:32,  3.42it/s]
 96%|█████████▌| 355960/371472 [6:10:04<1:15:55,  3.41it/s]
                                                           
{'loss': 2.6965, 'learning_rate': 1.3760262018143034e-07, 'epoch': 15.33}
+
 96%|█████████▌| 355960/371472 [6:10:04<1:15:55,  3.41it/s]
 96%|█████████▌| 355961/371472 [6:10:04<1:14:44,  3.46it/s]
 96%|█████████▌| 355962/371472 [6:10:04<1:13:47,  3.50it/s]
 96%|█████████▌| 355963/371472 [6:10:05<1:11:39,  3.61it/s]
 96%|█████████▌| 355964/371472 [6:10:05<1:11:46,  3.60it/s]
 96%|█████████▌| 355965/371472 [6:10:05<1:11:36,  3.61it/s]
 96%|█████████▌| 355966/371472 [6:10:06<1:18:43,  3.28it/s]
 96%|█████████▌| 355967/371472 [6:10:06<1:15:04,  3.44it/s]
 96%|█████████▌| 355968/371472 [6:10:06<1:14:23,  3.47it/s]
 96%|█████████▌| 355969/371472 [6:10:07<1:19:30,  3.25it/s]
 96%|█████████▌| 355970/371472 [6:10:07<1:17:42,  3.32it/s]
 96%|█████████▌| 355971/371472 [6:10:07<1:13:47,  3.50it/s]
 96%|█████████▌| 355972/371472 [6:10:07<1:17:46,  3.32it/s]
 96%|█████████▌| 355973/371472 [6:10:08<1:19:18,  3.26it/s]
 96%|█████████▌| 355974/371472 [6:10:08<1:17:14,  3.34it/s]
 96%|█████████▌| 355975/371472 [6:10:08<1:12:13,  3.58it/s]
 96%|█████████▌| 355976/371472 [6:10:09<1:15:35,  3.42it/s]
 96%|█████████▌| 355977/371472 [6:10:09<1:12:09,  3.58it/s]
 96%|█████████▌| 355978/371472 [6:10:09<1:16:37,  3.37it/s]
 96%|█████████▌| 355979/371472 [6:10:09<1:13:14,  3.53it/s]
 96%|█████████▌| 355980/371472 [6:10:10<1:12:57,  3.54it/s]
                                                           
{'loss': 2.5248, 'learning_rate': 1.375541382059514e-07, 'epoch': 15.33}
+
 96%|█████████▌| 355980/371472 [6:10:10<1:12:57,  3.54it/s]
 96%|█████████▌| 355981/371472 [6:10:10<1:11:46,  3.60it/s]
 96%|█████████▌| 355982/371472 [6:10:10<1:11:32,  3.61it/s]
 96%|█████████▌| 355983/371472 [6:10:11<1:10:06,  3.68it/s]
 96%|█████████▌| 355984/371472 [6:10:11<1:14:36,  3.46it/s]
 96%|█████████▌| 355985/371472 [6:10:11<1:12:37,  3.55it/s]
 96%|█████████▌| 355986/371472 [6:10:11<1:10:45,  3.65it/s]
 96%|█████████▌| 355987/371472 [6:10:12<1:11:46,  3.60it/s]
 96%|█████████▌| 355988/371472 [6:10:12<1:11:04,  3.63it/s]
 96%|█████████▌| 355989/371472 [6:10:12<1:16:49,  3.36it/s]
 96%|█████████▌| 355990/371472 [6:10:13<1:24:00,  3.07it/s]
 96%|█████████▌| 355991/371472 [6:10:13<1:23:46,  3.08it/s]
 96%|█████████▌| 355992/371472 [6:10:13<1:20:04,  3.22it/s]
 96%|█████████▌| 355993/371472 [6:10:14<1:23:29,  3.09it/s]
 96%|█████████▌| 355994/371472 [6:10:14<1:21:55,  3.15it/s]
 96%|█████████▌| 355995/371472 [6:10:14<1:20:59,  3.18it/s]
 96%|█████████▌| 355996/371472 [6:10:14<1:15:43,  3.41it/s]
 96%|█████████▌| 355997/371472 [6:10:15<1:16:36,  3.37it/s]
 96%|█████████▌| 355998/371472 [6:10:15<1:15:17,  3.43it/s]
 96%|█████████▌| 355999/371472 [6:10:15<1:15:18,  3.42it/s]
 96%|█████████▌| 356000/371472 [6:10:16<1:12:16,  3.57it/s]
                                                           
{'loss': 2.5559, 'learning_rate': 1.3750565623047257e-07, 'epoch': 15.33}
+
 96%|█████████▌| 356000/371472 [6:10:16<1:12:16,  3.57it/s]
 96%|█████████▌| 356001/371472 [6:10:16<1:13:09,  3.52it/s]
 96%|█████████▌| 356002/371472 [6:10:16<1:11:45,  3.59it/s]
 96%|█████████▌| 356003/371472 [6:10:16<1:10:26,  3.66it/s]
 96%|█████████▌| 356004/371472 [6:10:17<1:09:57,  3.69it/s]
 96%|█████████▌| 356005/371472 [6:10:17<1:10:47,  3.64it/s]
 96%|█████████▌| 356006/371472 [6:10:17<1:13:12,  3.52it/s]
 96%|█████████▌| 356007/371472 [6:10:18<1:10:09,  3.67it/s]
 96%|█████████▌| 356008/371472 [6:10:18<1:13:08,  3.52it/s]
 96%|█████████▌| 356009/371472 [6:10:18<1:13:57,  3.48it/s]
 96%|█████████▌| 356010/371472 [6:10:18<1:10:21,  3.66it/s]
 96%|█████████▌| 356011/371472 [6:10:19<1:11:58,  3.58it/s]
 96%|█████████▌| 356012/371472 [6:10:19<1:17:17,  3.33it/s]
 96%|█████████▌| 356013/371472 [6:10:19<1:13:50,  3.49it/s]
 96%|█████████▌| 356014/371472 [6:10:20<1:11:29,  3.60it/s]
 96%|█████████▌| 356015/371472 [6:10:20<1:12:49,  3.54it/s]
 96%|█████████▌| 356016/371472 [6:10:20<1:12:12,  3.57it/s]
 96%|█████████▌| 356017/371472 [6:10:20<1:12:03,  3.58it/s]
 96%|█████████▌| 356018/371472 [6:10:21<1:12:20,  3.56it/s]
 96%|█████████▌| 356019/371472 [6:10:21<1:16:54,  3.35it/s]
 96%|█████████▌| 356020/371472 [6:10:21<1:14:47,  3.44it/s]
                                                           
{'loss': 2.4135, 'learning_rate': 1.374571742549936e-07, 'epoch': 15.33}
+
 96%|█████████▌| 356020/371472 [6:10:21<1:14:47,  3.44it/s]
 96%|█████████▌| 356021/371472 [6:10:22<1:19:05,  3.26it/s]
 96%|█████████▌| 356022/371472 [6:10:22<1:13:43,  3.49it/s]
 96%|█████████▌| 356023/371472 [6:10:22<1:11:21,  3.61it/s]
 96%|█████████▌| 356024/371472 [6:10:22<1:13:54,  3.48it/s]
 96%|█████████▌| 356025/371472 [6:10:23<1:15:55,  3.39it/s]
 96%|█████████▌| 356026/371472 [6:10:23<1:16:26,  3.37it/s]
 96%|█████████▌| 356027/371472 [6:10:23<1:15:55,  3.39it/s]
 96%|█████████▌| 356028/371472 [6:10:24<1:14:41,  3.45it/s]
 96%|█████████▌| 356029/371472 [6:10:24<1:11:27,  3.60it/s]
 96%|█████████▌| 356030/371472 [6:10:24<1:10:05,  3.67it/s]
 96%|█████████▌| 356031/371472 [6:10:24<1:13:20,  3.51it/s]
 96%|█████████▌| 356032/371472 [6:10:25<1:15:47,  3.40it/s]
 96%|█████████▌| 356033/371472 [6:10:25<1:17:02,  3.34it/s]
 96%|█████████▌| 356034/371472 [6:10:25<1:18:08,  3.29it/s]
 96%|█████████▌| 356035/371472 [6:10:26<1:17:28,  3.32it/s]
 96%|█████████▌| 356036/371472 [6:10:26<1:17:36,  3.31it/s]
 96%|█████████▌| 356037/371472 [6:10:26<1:21:11,  3.17it/s]
 96%|█████████▌| 356038/371472 [6:10:27<1:22:45,  3.11it/s]
 96%|█████████▌| 356039/371472 [6:10:27<1:19:49,  3.22it/s]
 96%|█████████▌| 356040/371472 [6:10:27<1:17:51,  3.30it/s]
                                                           
{'loss': 2.5815, 'learning_rate': 1.3740869227951476e-07, 'epoch': 15.34}
+
 96%|█████████▌| 356040/371472 [6:10:27<1:17:51,  3.30it/s]
 96%|█████████▌| 356041/371472 [6:10:27<1:15:28,  3.41it/s]
 96%|████████���▌| 356042/371472 [6:10:28<1:18:11,  3.29it/s]
 96%|█████████▌| 356043/371472 [6:10:28<1:17:12,  3.33it/s]
 96%|█████████▌| 356044/371472 [6:10:28<1:15:55,  3.39it/s]
 96%|█████████▌| 356045/371472 [6:10:29<1:13:57,  3.48it/s]
 96%|█████████▌| 356046/371472 [6:10:29<1:20:27,  3.20it/s]
 96%|█████████▌| 356047/371472 [6:10:29<1:25:22,  3.01it/s]
 96%|█████████▌| 356048/371472 [6:10:30<1:20:25,  3.20it/s]
 96%|█████████▌| 356049/371472 [6:10:30<1:16:42,  3.35it/s]
 96%|█████████▌| 356050/371472 [6:10:30<1:14:55,  3.43it/s]
 96%|█████████▌| 356051/371472 [6:10:30<1:13:29,  3.50it/s]
 96%|█████████▌| 356052/371472 [6:10:31<1:16:37,  3.35it/s]
 96%|█████████▌| 356053/371472 [6:10:31<1:14:06,  3.47it/s]
 96%|█████████▌| 356054/371472 [6:10:31<1:15:38,  3.40it/s]
 96%|█████████▌| 356055/371472 [6:10:32<1:20:16,  3.20it/s]
 96%|█████████▌| 356056/371472 [6:10:32<1:18:27,  3.28it/s]
 96%|█████████▌| 356057/371472 [6:10:32<1:21:06,  3.17it/s]
 96%|█████████▌| 356058/371472 [6:10:33<1:19:43,  3.22it/s]
 96%|█████████▌| 356059/371472 [6:10:33<1:21:36,  3.15it/s]
 96%|█████████▌| 356060/371472 [6:10:33<1:16:36,  3.35it/s]
                                                           
{'loss': 2.5068, 'learning_rate': 1.3736021030403583e-07, 'epoch': 15.34}
+
 96%|█████████▌| 356060/371472 [6:10:33<1:16:36,  3.35it/s]
 96%|█████████▌| 356061/371472 [6:10:34<1:15:25,  3.41it/s]
 96%|█████████▌| 356062/371472 [6:10:34<1:16:04,  3.38it/s]
 96%|█████████▌| 356063/371472 [6:10:34<1:21:10,  3.16it/s]
 96%|█████████▌| 356064/371472 [6:10:34<1:17:21,  3.32it/s]
 96%|█████████▌| 356065/371472 [6:10:35<1:13:25,  3.50it/s]
 96%|█████████▌| 356066/371472 [6:10:35<1:11:12,  3.61it/s]
 96%|█████████▌| 356067/371472 [6:10:35<1:21:22,  3.16it/s]
 96%|█████████▌| 356068/371472 [6:10:36<1:22:29,  3.11it/s]
 96%|█████████▌| 356069/371472 [6:10:36<1:16:54,  3.34it/s]
 96%|█████████▌| 356070/371472 [6:10:36<1:18:42,  3.26it/s]
 96%|█████████▌| 356071/371472 [6:10:37<1:16:19,  3.36it/s]
 96%|█████████▌| 356072/371472 [6:10:37<1:12:13,  3.55it/s]
 96%|█████████▌| 356073/371472 [6:10:37<1:10:35,  3.64it/s]
 96%|█████████▌| 356074/371472 [6:10:37<1:11:02,  3.61it/s]
 96%|█████████▌| 356075/371472 [6:10:38<1:08:51,  3.73it/s]
 96%|█████████▌| 356076/371472 [6:10:38<1:11:43,  3.58it/s]
 96%|█████████▌| 356077/371472 [6:10:38<1:11:38,  3.58it/s]
 96%|█████████▌| 356078/371472 [6:10:38<1:09:37,  3.68it/s]
 96%|█████████▌| 356079/371472 [6:10:39<1:07:47,  3.78it/s]
 96%|█████████▌| 356080/371472 [6:10:39<1:10:25,  3.64it/s]
                                                           
{'loss': 2.7345, 'learning_rate': 1.3731172832855698e-07, 'epoch': 15.34}
+
 96%|█████████▌| 356080/371472 [6:10:39<1:10:25,  3.64it/s]
 96%|█████████▌| 356081/371472 [6:10:39<1:08:55,  3.72it/s]
 96%|█████████▌| 356082/371472 [6:10:39<1:07:56,  3.78it/s]
 96%|█████████▌| 356083/371472 [6:10:40<1:08:12,  3.76it/s]
 96%|█████████▌| 356084/371472 [6:10:40<1:08:43,  3.73it/s]
 96%|█████████▌| 356085/371472 [6:10:40<1:07:13,  3.81it/s]
 96%|█████████▌| 356086/371472 [6:10:41<1:06:40,  3.85it/s]
 96%|█████████▌| 356087/371472 [6:10:41<1:06:48,  3.84it/s]
 96%|█████████▌| 356088/371472 [6:10:41<1:08:30,  3.74it/s]
 96%|█████████▌| 356089/371472 [6:10:41<1:11:35,  3.58it/s]
 96%|█████████▌| 356090/371472 [6:10:42<1:16:41,  3.34it/s]
 96%|█████████▌| 356091/371472 [6:10:42<1:16:02,  3.37it/s]
 96%|█████████▌| 356092/371472 [6:10:42<1:17:21,  3.31it/s]
 96%|█████████▌| 356093/371472 [6:10:43<1:16:17,  3.36it/s]
 96%|█████████▌| 356094/371472 [6:10:43<1:15:43,  3.38it/s]
 96%|█████████▌| 356095/371472 [6:10:43<1:11:37,  3.58it/s]
 96%|█████████▌| 356096/371472 [6:10:43<1:11:12,  3.60it/s]
 96%|█████████▌| 356097/371472 [6:10:44<1:14:46,  3.43it/s]
 96%|█████████▌| 356098/371472 [6:10:44<1:11:52,  3.57it/s]
 96%|█████████▌| 356099/371472 [6:10:44<1:17:54,  3.29it/s]
 96%|█████████▌| 356100/371472 [6:10:45<1:14:06,  3.46it/s]
                                                           
{'loss': 2.604, 'learning_rate': 1.3726324635307803e-07, 'epoch': 15.34}
+
 96%|█████████▌| 356100/371472 [6:10:45<1:14:06,  3.46it/s]
 96%|█████████▌| 356101/371472 [6:10:45<1:16:29,  3.35it/s]
 96%|█████████▌| 356102/371472 [6:10:45<1:26:44,  2.95it/s]
 96%|█████████▌| 356103/371472 [6:10:46<1:20:57,  3.16it/s]
 96%|█████████▌| 356104/371472 [6:10:46<1:17:13,  3.32it/s]
 96%|█████████▌| 356105/371472 [6:10:46<1:19:00,  3.24it/s]
 96%|█████████▌| 356106/371472 [6:10:47<1:17:35,  3.30it/s]
 96%|█████████▌| 356107/371472 [6:10:47<1:14:29,  3.44it/s]
 96%|█████████▌| 356108/371472 [6:10:47<1:12:57,  3.51it/s]
 96%|█████████▌| 356109/371472 [6:10:47<1:13:29,  3.48it/s]
 96%|█████████▌| 356110/371472 [6:10:48<1:13:55,  3.46it/s]
 96%|█████████▌| 356111/371472 [6:10:48<1:14:59,  3.41it/s]
 96%|█████████▌| 356112/371472 [6:10:48<1:13:12,  3.50it/s]
 96%|█████████▌| 356113/371472 [6:10:48<1:11:28,  3.58it/s]
 96%|█████████▌| 356114/371472 [6:10:49<1:11:48,  3.56it/s]
 96%|█████████▌| 356115/371472 [6:10:49<1:10:46,  3.62it/s]
 96%|█████████▌| 356116/371472 [6:10:49<1:12:34,  3.53it/s]
 96%|█████████▌| 356117/371472 [6:10:50<1:13:03,  3.50it/s]
 96%|█████████▌| 356118/371472 [6:10:50<1:10:03,  3.65it/s]
 96%|█████████▌| 356119/371472 [6:10:50<1:10:10,  3.65it/s]
 96%|█████████▌| 356120/371472 [6:10:50<1:08:25,  3.74it/s]
                                                           
{'loss': 2.585, 'learning_rate': 1.372147643775992e-07, 'epoch': 15.34}
+
 96%|█████████▌| 356120/371472 [6:10:50<1:08:25,  3.74it/s]
 96%|█████████▌| 356121/371472 [6:10:51<1:07:25,  3.79it/s]
 96%|█████████▌| 356122/371472 [6:10:51<1:11:12,  3.59it/s]
 96%|█████████▌| 356123/371472 [6:10:51<1:09:34,  3.68it/s]
 96%|█████████▌| 356124/371472 [6:10:51<1:07:09,  3.81it/s]
 96%|█████████▌| 356125/371472 [6:10:52<1:09:17,  3.69it/s]
 96%|█████████▌| 356126/371472 [6:10:52<1:15:12,  3.40it/s]
 96%|█████████▌| 356127/371472 [6:10:52<1:14:17,  3.44it/s]
 96%|█████████▌| 356128/371472 [6:10:53<1:13:07,  3.50it/s]
 96%|█████████▌| 356129/371472 [6:10:53<1:12:55,  3.51it/s]
 96%|█████████▌| 356130/371472 [6:10:53<1:13:58,  3.46it/s]
 96%|█████████▌| 356131/371472 [6:10:54<1:14:45,  3.42it/s]
 96%|█████████▌| 356132/371472 [6:10:54<1:13:10,  3.49it/s]
 96%|█████████▌| 356133/371472 [6:10:54<1:10:16,  3.64it/s]
 96%|█████████▌| 356134/371472 [6:10:54<1:15:43,  3.38it/s]
 96%|█████████▌| 356135/371472 [6:10:55<1:15:29,  3.39it/s]
 96%|█████████▌| 356136/371472 [6:10:55<1:11:28,  3.58it/s]
 96%|█████████▌| 356137/371472 [6:10:55<1:09:48,  3.66it/s]
 96%|█████████▌| 356138/371472 [6:10:55<1:07:47,  3.77it/s]
 96%|█████████▌| 356139/371472 [6:10:56<1:09:25,  3.68it/s]
 96%|█████████▌| 356140/371472 [6:10:56<1:09:44,  3.66it/s]
                                                           
{'loss': 2.6818, 'learning_rate': 1.3716628240212025e-07, 'epoch': 15.34}
+
 96%|█████████▌| 356140/371472 [6:10:56<1:09:44,  3.66it/s]
 96%|█████████▌| 356141/371472 [6:10:56<1:12:38,  3.52it/s]
 96%|█████████▌| 356142/371472 [6:10:57<1:11:03,  3.60it/s]
 96%|█████████▌| 356143/371472 [6:10:57<1:15:10,  3.40it/s]
 96%|█████████▌| 356144/371472 [6:10:57<1:12:49,  3.51it/s]
 96%|█████████▌| 356145/371472 [6:10:57<1:11:18,  3.58it/s]
 96%|█████████▌| 356146/371472 [6:10:58<1:11:03,  3.59it/s]
 96%|█████████▌| 356147/371472 [6:10:58<1:11:27,  3.57it/s]
 96%|█████████▌| 356148/371472 [6:10:58<1:13:20,  3.48it/s]
 96%|█████████▌| 356149/371472 [6:10:59<1:13:37,  3.47it/s]
 96%|█████████▌| 356150/371472 [6:10:59<1:15:29,  3.38it/s]
 96%|█████████▌| 356151/371472 [6:10:59<1:15:05,  3.40it/s]
 96%|█████████▌| 356152/371472 [6:10:59<1:11:03,  3.59it/s]
 96%|█████████▌| 356153/371472 [6:11:00<1:09:01,  3.70it/s]
 96%|█████████▌| 356154/371472 [6:11:00<1:11:06,  3.59it/s]
 96%|█████████▌| 356155/371472 [6:11:00<1:13:13,  3.49it/s]
 96%|█████████▌| 356156/371472 [6:11:01<1:19:38,  3.21it/s]
 96%|█████████▌| 356157/371472 [6:11:01<1:18:00,  3.27it/s]
 96%|█████████▌| 356158/371472 [6:11:01<1:15:41,  3.37it/s]
 96%|█████████▌| 356159/371472 [6:11:02<1:17:05,  3.31it/s]
 96%|█████████▌| 356160/371472 [6:11:02<1:13:11,  3.49it/s]
                                                           
{'loss': 2.708, 'learning_rate': 1.371178004266414e-07, 'epoch': 15.34}
+
 96%|█████████▌| 356160/371472 [6:11:02<1:13:11,  3.49it/s]
 96%|█████████▌| 356161/371472 [6:11:02<1:14:35,  3.42it/s]
 96%|█████████▌| 356162/371472 [6:11:02<1:12:15,  3.53it/s]
 96%|█████████▌| 356163/371472 [6:11:03<1:08:34,  3.72it/s]
 96%|█████████▌| 356164/371472 [6:11:03<1:08:52,  3.70it/s]
 96%|█████████▌| 356165/371472 [6:11:03<1:10:25,  3.62it/s]
 96%|█████████▌| 356166/371472 [6:11:03<1:09:31,  3.67it/s]
 96%|█████████▌| 356167/371472 [6:11:04<1:12:39,  3.51it/s]
 96%|█████████▌| 356168/371472 [6:11:04<1:11:38,  3.56it/s]
 96%|█████████▌| 356169/371472 [6:11:04<1:09:05,  3.69it/s]
 96%|█████████▌| 356170/371472 [6:11:05<1:09:05,  3.69it/s]
 96%|█████████▌| 356171/371472 [6:11:05<1:09:51,  3.65it/s]
 96%|█████████▌| 356172/371472 [6:11:05<1:08:39,  3.71it/s]
 96%|█████████▌| 356173/371472 [6:11:05<1:08:15,  3.74it/s]
 96%|█████████▌| 356174/371472 [6:11:06<1:08:02,  3.75it/s]
 96%|█████████▌| 356175/371472 [6:11:06<1:10:35,  3.61it/s]
 96%|█████████▌| 356176/371472 [6:11:06<1:12:15,  3.53it/s]
 96%|█████████▌| 356177/371472 [6:11:07<1:16:36,  3.33it/s]
 96%|█████████▌| 356178/371472 [6:11:07<1:14:40,  3.41it/s]
 96%|█████████▌| 356179/371472 [6:11:07<1:15:30,  3.38it/s]
 96%|█████████▌| 356180/371472 [6:11:07<1:21:06,  3.14it/s]
                                                           
{'loss': 2.519, 'learning_rate': 1.3706931845116247e-07, 'epoch': 15.34}
+
 96%|█████████▌| 356180/371472 [6:11:07<1:21:06,  3.14it/s]
 96%|█████████▌| 356181/371472 [6:11:08<1:18:42,  3.24it/s]
 96%|█████████▌| 356182/371472 [6:11:08<1:14:35,  3.42it/s]
 96%|█████████▌| 356183/371472 [6:11:08<1:12:28,  3.52it/s]
 96%|█████████▌| 356184/371472 [6:11:09<1:13:03,  3.49it/s]
 96%|█████████▌| 356185/371472 [6:11:09<1:13:57,  3.44it/s]
 96%|█████████▌| 356186/371472 [6:11:09<1:13:44,  3.45it/s]
 96%|█████████▌| 356187/371472 [6:11:09<1:13:41,  3.46it/s]
 96%|█████████▌| 356188/371472 [6:11:10<1:12:29,  3.51it/s]
 96%|█████████▌| 356189/371472 [6:11:10<1:11:05,  3.58it/s]
 96%|█████████▌| 356190/371472 [6:11:10<1:15:04,  3.39it/s]
 96%|█████████▌| 356191/371472 [6:11:11<1:11:58,  3.54it/s]
 96%|█████████▌| 356192/371472 [6:11:11<1:11:52,  3.54it/s]
 96%|█████████▌| 356193/371472 [6:11:11<1:17:29,  3.29it/s]
 96%|█████████▌| 356194/371472 [6:11:11<1:13:00,  3.49it/s]
 96%|█████████▌| 356195/371472 [6:11:12<1:09:50,  3.65it/s]
 96%|█████████▌| 356196/371472 [6:11:12<1:10:00,  3.64it/s]
 96%|█████████▌| 356197/371472 [6:11:12<1:08:45,  3.70it/s]
 96%|█████████▌| 356198/371472 [6:11:13<1:11:28,  3.56it/s]
 96%|█████████▌| 356199/371472 [6:11:13<1:07:52,  3.75it/s]
 96%|█████████▌| 356200/371472 [6:11:13<1:08:56,  3.69it/s]
                                                           
{'loss': 2.5596, 'learning_rate': 1.3702083647568363e-07, 'epoch': 15.34}
+
 96%|█████████▌| 356200/371472 [6:11:13<1:08:56,  3.69it/s]
 96%|█████████▌| 356201/371472 [6:11:13<1:10:36,  3.60it/s]
 96%|█████████▌| 356202/371472 [6:11:14<1:10:45,  3.60it/s]
 96%|█████████▌| 356203/371472 [6:11:14<1:15:25,  3.37it/s]
 96%|█████████▌| 356204/371472 [6:11:14<1:14:27,  3.42it/s]
 96%|█████████▌| 356205/371472 [6:11:15<1:18:01,  3.26it/s]
 96%|█████████▌| 356206/371472 [6:11:15<1:16:15,  3.34it/s]
 96%|█████████▌| 356207/371472 [6:11:15<1:14:14,  3.43it/s]
 96%|█████████▌| 356208/371472 [6:11:15<1:17:28,  3.28it/s]
 96%|█████████▌| 356209/371472 [6:11:16<1:15:30,  3.37it/s]
 96%|█████████▌| 356210/371472 [6:11:16<1:15:01,  3.39it/s]
 96%|█████████▌| 356211/371472 [6:11:16<1:12:59,  3.48it/s]
 96%|█████████▌| 356212/371472 [6:11:17<1:09:58,  3.63it/s]
 96%|█████████▌| 356213/371472 [6:11:17<1:10:05,  3.63it/s]
 96%|█████████▌| 356214/371472 [6:11:17<1:09:00,  3.68it/s]
 96%|█████████▌| 356215/371472 [6:11:17<1:09:34,  3.65it/s]
 96%|█████████▌| 356216/371472 [6:11:18<1:13:13,  3.47it/s]
 96%|█████████▌| 356217/371472 [6:11:18<1:11:44,  3.54it/s]
 96%|█████████▌| 356218/371472 [6:11:18<1:14:04,  3.43it/s]
 96%|█████████▌| 356219/371472 [6:11:19<1:10:46,  3.59it/s]
 96%|█████████▌| 356220/371472 [6:11:19<1:12:17,  3.52it/s]
                                                           
{'loss': 2.5721, 'learning_rate': 1.369723545002047e-07, 'epoch': 15.34}
+
 96%|█████████▌| 356220/371472 [6:11:19<1:12:17,  3.52it/s]
 96%|█████████▌| 356221/371472 [6:11:19<1:15:14,  3.38it/s]
 96%|█████████▌| 356222/371472 [6:11:20<1:20:23,  3.16it/s]
 96%|█████████▌| 356223/371472 [6:11:20<1:18:58,  3.22it/s]
 96%|█████████▌| 356224/371472 [6:11:20<1:21:40,  3.11it/s]
 96%|█████████▌| 356225/371472 [6:11:21<1:28:49,  2.86it/s]
 96%|█████████▌| 356226/371472 [6:11:21<1:24:24,  3.01it/s]
 96%|█████████▌| 356227/371472 [6:11:21<1:24:17,  3.01it/s]
 96%|█████████▌| 356228/371472 [6:11:21<1:19:01,  3.22it/s]
 96%|█████████▌| 356229/371472 [6:11:22<1:17:11,  3.29it/s]
 96%|█████████▌| 356230/371472 [6:11:22<1:16:46,  3.31it/s]
 96%|█████████▌| 356231/371472 [6:11:22<1:17:08,  3.29it/s]
 96%|█████████▌| 356232/371472 [6:11:23<1:16:59,  3.30it/s]
 96%|█████████▌| 356233/371472 [6:11:23<1:16:23,  3.32it/s]
 96%|█████████▌| 356234/371472 [6:11:23<1:12:59,  3.48it/s]
 96%|█████████▌| 356235/371472 [6:11:23<1:11:21,  3.56it/s]
 96%|█████████▌| 356236/371472 [6:11:24<1:08:20,  3.72it/s]
 96%|█████████▌| 356237/371472 [6:11:24<1:09:25,  3.66it/s]
 96%|█████████▌| 356238/371472 [6:11:24<1:13:41,  3.45it/s]
 96%|█████████▌| 356239/371472 [6:11:25<1:10:30,  3.60it/s]
 96%|█████████▌| 356240/371472 [6:11:25<1:09:29,  3.65it/s]
                                                           
{'loss': 2.5023, 'learning_rate': 1.3692387252472585e-07, 'epoch': 15.34}
+
 96%|█████████▌| 356240/371472 [6:11:25<1:09:29,  3.65it/s]
 96%|█████████▌| 356241/371472 [6:11:25<1:07:15,  3.77it/s]
 96%|█████████▌| 356242/371472 [6:11:25<1:06:41,  3.81it/s]
 96%|█████████▌| 356243/371472 [6:11:26<1:09:48,  3.64it/s]
 96%|█████████▌| 356244/371472 [6:11:26<1:13:47,  3.44it/s]
 96%|█████████▌| 356245/371472 [6:11:26<1:22:12,  3.09it/s]
 96%|█████████▌| 356246/371472 [6:11:27<1:21:02,  3.13it/s]
 96%|█████████▌| 356247/371472 [6:11:27<1:17:10,  3.29it/s]
 96%|█████████▌| 356248/371472 [6:11:27<1:16:10,  3.33it/s]
 96%|█████████▌| 356249/371472 [6:11:28<1:13:49,  3.44it/s]
 96%|█████████▌| 356250/371472 [6:11:28<1:12:26,  3.50it/s]
 96%|█████████▌| 356251/371472 [6:11:28<1:10:40,  3.59it/s]
 96%|█████████▌| 356252/371472 [6:11:28<1:13:12,  3.47it/s]
 96%|█████████▌| 356253/371472 [6:11:29<1:14:34,  3.40it/s]
 96%|█████████▌| 356254/371472 [6:11:29<1:12:59,  3.47it/s]
 96%|█████████▌| 356255/371472 [6:11:29<1:12:44,  3.49it/s]
 96%|█████████▌| 356256/371472 [6:11:30<1:13:48,  3.44it/s]
 96%|█████████▌| 356257/371472 [6:11:30<1:15:44,  3.35it/s]
 96%|█████████▌| 356258/371472 [6:11:30<1:13:05,  3.47it/s]
 96%|█████████▌| 356259/371472 [6:11:30<1:12:51,  3.48it/s]
 96%|█████████▌| 356260/371472 [6:11:31<1:12:29,  3.50it/s]
                                                           
{'loss': 2.6405, 'learning_rate': 1.3687539054924695e-07, 'epoch': 15.34}
+
 96%|█████████▌| 356260/371472 [6:11:31<1:12:29,  3.50it/s]
 96%|█████████▌| 356261/371472 [6:11:31<1:11:41,  3.54it/s]
 96%|█████████▌| 356262/371472 [6:11:31<1:08:12,  3.72it/s]
 96%|█████████▌| 356263/371472 [6:11:32<1:13:08,  3.47it/s]
 96%|█████████▌| 356264/371472 [6:11:32<1:14:35,  3.40it/s]
 96%|█████████▌| 356265/371472 [6:11:32<1:19:59,  3.17it/s]
 96%|█████████▌| 356266/371472 [6:11:33<1:18:42,  3.22it/s]
 96%|█████████▌| 356267/371472 [6:11:33<1:17:26,  3.27it/s]
 96%|█████████▌| 356268/371472 [6:11:33<1:17:01,  3.29it/s]
 96%|█████████▌| 356269/371472 [6:11:33<1:19:24,  3.19it/s]
 96%|█████████▌| 356270/371472 [6:11:34<1:14:37,  3.40it/s]
 96%|█████████▌| 356271/371472 [6:11:34<1:11:21,  3.55it/s]
 96%|█████████▌| 356272/371472 [6:11:34<1:11:02,  3.57it/s]
 96%|█████████▌| 356273/371472 [6:11:34<1:10:01,  3.62it/s]
 96%|█████████▌| 356274/371472 [6:11:35<1:07:10,  3.77it/s]
 96%|█████████▌| 356275/371472 [6:11:35<1:07:33,  3.75it/s]
 96%|█████████▌| 356276/371472 [6:11:35<1:14:03,  3.42it/s]
 96%|█████████▌| 356277/371472 [6:11:36<1:15:40,  3.35it/s]
 96%|█████████▌| 356278/371472 [6:11:36<1:14:11,  3.41it/s]
 96%|█████████▌| 356279/371472 [6:11:36<1:12:16,  3.50it/s]
 96%|█████████▌| 356280/371472 [6:11:36<1:10:28,  3.59it/s]
                                                           
{'loss': 2.6164, 'learning_rate': 1.3682690857376807e-07, 'epoch': 15.35}
+
 96%|█████████▌| 356280/371472 [6:11:36<1:10:28,  3.59it/s]
 96%|█████████▌| 356281/371472 [6:11:37<1:11:23,  3.55it/s]
 96%|█████████▌| 356282/371472 [6:11:37<1:10:50,  3.57it/s]
 96%|█████████▌| 356283/371472 [6:11:37<1:09:18,  3.65it/s]
 96%|█████████▌| 356284/371472 [6:11:38<1:08:07,  3.72it/s]
 96%|█████████▌| 356285/371472 [6:11:38<1:12:23,  3.50it/s]
 96%|█████████▌| 356286/371472 [6:11:38<1:10:05,  3.61it/s]
 96%|█████████▌| 356287/371472 [6:11:38<1:11:03,  3.56it/s]
 96%|█████████▌| 356288/371472 [6:11:39<1:13:22,  3.45it/s]
 96%|█████████▌| 356289/371472 [6:11:39<1:19:07,  3.20it/s]
 96%|█████████▌| 356290/371472 [6:11:40<1:27:47,  2.88it/s]
 96%|█████████▌| 356291/371472 [6:11:40<1:32:22,  2.74it/s]
 96%|█████████▌| 356292/371472 [6:11:40<1:24:24,  3.00it/s]
 96%|█████████▌| 356293/371472 [6:11:40<1:18:30,  3.22it/s]
 96%|█████████▌| 356294/371472 [6:11:41<1:16:48,  3.29it/s]
 96%|█████████▌| 356295/371472 [6:11:41<1:14:08,  3.41it/s]
 96%|█████████▌| 356296/371472 [6:11:41<1:14:25,  3.40it/s]
 96%|█████████▌| 356297/371472 [6:11:42<1:11:54,  3.52it/s]
 96%|█████████▌| 356298/371472 [6:11:42<1:13:28,  3.44it/s]
 96%|█████████▌| 356299/371472 [6:11:42<1:10:35,  3.58it/s]
 96%|█████████▌| 356300/371472 [6:11:42<1:08:30,  3.69it/s]
                                                           
{'loss': 2.49, 'learning_rate': 1.3677842659828912e-07, 'epoch': 15.35}
+
 96%|█████████▌| 356300/371472 [6:11:42<1:08:30,  3.69it/s]
 96%|█████████▌| 356301/371472 [6:11:43<1:14:08,  3.41it/s]
 96%|█████████▌| 356302/371472 [6:11:43<1:13:15,  3.45it/s]
 96%|█████████▌| 356303/371472 [6:11:43<1:12:01,  3.51it/s]
 96%|█████████▌| 356304/371472 [6:11:44<1:11:14,  3.55it/s]
 96%|█████████▌| 356305/371472 [6:11:44<1:11:44,  3.52it/s]
 96%|█████████▌| 356306/371472 [6:11:44<1:20:17,  3.15it/s]
 96%|█████████▌| 356307/371472 [6:11:44<1:16:09,  3.32it/s]
 96%|█████████▌| 356308/371472 [6:11:45<1:15:12,  3.36it/s]
 96%|█████████▌| 356309/371472 [6:11:45<1:12:45,  3.47it/s]
 96%|█████████▌| 356310/371472 [6:11:45<1:12:23,  3.49it/s]
 96%|█████████▌| 356311/371472 [6:11:46<1:09:15,  3.65it/s]
 96%|█████████▌| 356312/371472 [6:11:46<1:09:44,  3.62it/s]
 96%|█████████▌| 356313/371472 [6:11:46<1:14:25,  3.39it/s]
 96%|█████████▌| 356314/371472 [6:11:47<1:20:05,  3.15it/s]
 96%|█████████▌| 356315/371472 [6:11:47<1:16:39,  3.30it/s]
 96%|█████████▌| 356316/371472 [6:11:47<1:19:11,  3.19it/s]
 96%|█████████▌| 356317/371472 [6:11:47<1:15:26,  3.35it/s]
 96%|█████████▌| 356318/371472 [6:11:48<1:12:08,  3.50it/s]
 96%|█████████▌| 356319/371472 [6:11:48<1:11:44,  3.52it/s]
 96%|█████████▌| 356320/371472 [6:11:48<1:14:28,  3.39it/s]
                                                           
{'loss': 2.4903, 'learning_rate': 1.3672994462281019e-07, 'epoch': 15.35}
+
 96%|█████████▌| 356320/371472 [6:11:48<1:14:28,  3.39it/s]
 96%|█████████▌| 356321/371472 [6:11:49<1:14:06,  3.41it/s]
 96%|█████████▌| 356322/371472 [6:11:49<1:16:07,  3.32it/s]
 96%|█████████▌| 356323/371472 [6:11:49<1:16:30,  3.30it/s]
 96%|█████████▌| 356324/371472 [6:11:50<1:16:44,  3.29it/s]
 96%|█████████▌| 356325/371472 [6:11:50<1:14:05,  3.41it/s]
 96%|█████████▌| 356326/371472 [6:11:50<1:12:25,  3.49it/s]
 96%|█████████▌| 356327/371472 [6:11:50<1:15:15,  3.35it/s]
 96%|█████████▌| 356328/371472 [6:11:51<1:13:37,  3.43it/s]
 96%|█████████▌| 356329/371472 [6:11:51<1:13:26,  3.44it/s]
 96%|█████████▌| 356330/371472 [6:11:51<1:19:11,  3.19it/s]
 96%|█████████▌| 356331/371472 [6:11:52<1:14:43,  3.38it/s]
 96%|█████████▌| 356332/371472 [6:11:52<1:12:46,  3.47it/s]
 96%|█████████▌| 356333/371472 [6:11:52<1:17:05,  3.27it/s]
 96%|█████████▌| 356334/371472 [6:11:52<1:13:17,  3.44it/s]
 96%|█████████▌| 356335/371472 [6:11:53<1:12:32,  3.48it/s]
 96%|█████████▌| 356336/371472 [6:11:53<1:12:41,  3.47it/s]
 96%|█████████▌| 356337/371472 [6:11:53<1:14:54,  3.37it/s]
 96%|█████████▌| 356338/371472 [6:11:54<1:12:00,  3.50it/s]
 96%|█████████▌| 356339/371472 [6:11:54<1:09:13,  3.64it/s]
 96%|█████████▌| 356340/371472 [6:11:54<1:07:52,  3.72it/s]
                                                           
{'loss': 2.5136, 'learning_rate': 1.3668146264733134e-07, 'epoch': 15.35}
+
 96%|█████████▌| 356340/371472 [6:11:54<1:07:52,  3.72it/s]
 96%|█████████▌| 356341/371472 [6:11:54<1:08:24,  3.69it/s]
 96%|█████████▌| 356342/371472 [6:11:55<1:10:16,  3.59it/s]
 96%|█████████▌| 356343/371472 [6:11:55<1:13:21,  3.44it/s]
 96%|█████████▌| 356344/371472 [6:11:55<1:09:09,  3.65it/s]
 96%|█████████▌| 356345/371472 [6:11:56<1:09:58,  3.60it/s]
 96%|█████████▌| 356346/371472 [6:11:56<1:07:52,  3.71it/s]
 96%|█████████▌| 356347/371472 [6:11:56<1:09:05,  3.65it/s]
 96%|█████████▌| 356348/371472 [6:11:56<1:09:23,  3.63it/s]
 96%|█████████▌| 356349/371472 [6:11:57<1:09:33,  3.62it/s]
 96%|█████████▌| 356350/371472 [6:11:57<1:06:40,  3.78it/s]
 96%|█████████▌| 356351/371472 [6:11:57<1:04:31,  3.91it/s]
 96%|█████████▌| 356352/371472 [6:11:57<1:07:22,  3.74it/s]
 96%|█████████▌| 356353/371472 [6:11:58<1:07:30,  3.73it/s]
 96%|█████████▌| 356354/371472 [6:11:58<1:10:30,  3.57it/s]
 96%|█████████▌| 356355/371472 [6:11:58<1:09:27,  3.63it/s]
 96%|█████████▌| 356356/371472 [6:11:58<1:08:21,  3.69it/s]
 96%|█████████▌| 356357/371472 [6:11:59<1:08:14,  3.69it/s]
 96%|█████████▌| 356358/371472 [6:11:59<1:10:36,  3.57it/s]
 96%|█████████▌| 356359/371472 [6:11:59<1:08:19,  3.69it/s]
 96%|█████████▌| 356360/371472 [6:12:00<1:07:54,  3.71it/s]
                                                           
{'loss': 2.7022, 'learning_rate': 1.3663298067185238e-07, 'epoch': 15.35}
+
 96%|█████████▌| 356360/371472 [6:12:00<1:07:54,  3.71it/s]
 96%|█████████▌| 356361/371472 [6:12:00<1:07:03,  3.76it/s]
 96%|█████████▌| 356362/371472 [6:12:00<1:06:57,  3.76it/s]
 96%|█████████▌| 356363/371472 [6:12:00<1:05:05,  3.87it/s]
 96%|█████████▌| 356364/371472 [6:12:01<1:05:58,  3.82it/s]
 96%|█████████▌| 356365/371472 [6:12:01<1:04:29,  3.90it/s]
 96%|█████████▌| 356366/371472 [6:12:01<1:09:50,  3.61it/s]
 96%|█████████▌| 356367/371472 [6:12:01<1:11:41,  3.51it/s]
 96%|█████████▌| 356368/371472 [6:12:02<1:10:18,  3.58it/s]
 96%|█████████▌| 356369/371472 [6:12:02<1:11:48,  3.51it/s]
 96%|█████████▌| 356370/371472 [6:12:02<1:09:17,  3.63it/s]
 96%|█████████▌| 356371/371472 [6:12:03<1:11:00,  3.54it/s]
 96%|█████████▌| 356372/371472 [6:12:03<1:09:01,  3.65it/s]
 96%|█████████▌| 356373/371472 [6:12:03<1:13:07,  3.44it/s]
 96%|█████████▌| 356374/371472 [6:12:03<1:12:11,  3.49it/s]
 96%|█████████▌| 356375/371472 [6:12:04<1:13:30,  3.42it/s]
 96%|█████████▌| 356376/371472 [6:12:04<1:16:12,  3.30it/s]
 96%|█████████▌| 356377/371472 [6:12:04<1:19:20,  3.17it/s]
 96%|█████████▌| 356378/371472 [6:12:05<1:15:04,  3.35it/s]
 96%|█████████▌| 356379/371472 [6:12:05<1:12:08,  3.49it/s]
 96%|█████████▌| 356380/371472 [6:12:05<1:13:18,  3.43it/s]
                                                           
{'loss': 2.7203, 'learning_rate': 1.3658449869637356e-07, 'epoch': 15.35}
+
 96%|█████████▌| 356380/371472 [6:12:05<1:13:18,  3.43it/s]
 96%|█████████▌| 356381/371472 [6:12:06<1:12:08,  3.49it/s]
 96%|█████████▌| 356382/371472 [6:12:06<1:08:41,  3.66it/s]
 96%|█████████▌| 356383/371472 [6:12:06<1:09:49,  3.60it/s]
 96%|█████████▌| 356384/371472 [6:12:06<1:09:12,  3.63it/s]
 96%|█████████▌| 356385/371472 [6:12:07<1:07:04,  3.75it/s]
 96%|█████████▌| 356386/371472 [6:12:07<1:06:35,  3.78it/s]
 96%|█████████▌| 356387/371472 [6:12:07<1:06:15,  3.79it/s]
 96%|█████████▌| 356388/371472 [6:12:07<1:07:20,  3.73it/s]
 96%|█████████▌| 356389/371472 [6:12:08<1:08:56,  3.65it/s]
 96%|█████████▌| 356390/371472 [6:12:08<1:13:34,  3.42it/s]
 96%|█████████▌| 356391/371472 [6:12:08<1:10:05,  3.59it/s]
 96%|█████████▌| 356392/371472 [6:12:09<1:09:24,  3.62it/s]
 96%|█████████▌| 356393/371472 [6:12:09<1:09:49,  3.60it/s]
 96%|█████████▌| 356394/371472 [6:12:09<1:14:01,  3.39it/s]
 96%|█████████▌| 356395/371472 [6:12:09<1:13:08,  3.44it/s]
 96%|█████████▌| 356396/371472 [6:12:10<1:15:39,  3.32it/s]
 96%|█████████▌| 356397/371472 [6:12:10<1:20:59,  3.10it/s]
 96%|█████████▌| 356398/371472 [6:12:10<1:18:49,  3.19it/s]
 96%|█████████▌| 356399/371472 [6:12:11<1:15:36,  3.32it/s]
 96%|█████████▌| 356400/371472 [6:12:11<1:21:09,  3.09it/s]
                                                           
{'loss': 2.4629, 'learning_rate': 1.365360167208946e-07, 'epoch': 15.35}
+
 96%|█████████▌| 356400/371472 [6:12:11<1:21:09,  3.09it/s]
 96%|█████████▌| 356401/371472 [6:12:11<1:18:12,  3.21it/s]
 96%|█████████▌| 356402/371472 [6:12:12<1:14:01,  3.39it/s]
 96%|█████████▌| 356403/371472 [6:12:12<1:13:06,  3.44it/s]
 96%|█████████▌| 356404/371472 [6:12:12<1:10:58,  3.54it/s]
 96%|█████████▌| 356405/371472 [6:12:12<1:09:18,  3.62it/s]
 96%|█████████▌| 356406/371472 [6:12:13<1:07:07,  3.74it/s]
 96%|█████████▌| 356407/371472 [6:12:13<1:06:28,  3.78it/s]
 96%|█████████▌| 356408/371472 [6:12:13<1:09:05,  3.63it/s]
 96%|█████████▌| 356409/371472 [6:12:13<1:11:05,  3.53it/s]
 96%|█████████▌| 356410/371472 [6:12:14<1:10:51,  3.54it/s]
 96%|█████████▌| 356411/371472 [6:12:14<1:09:58,  3.59it/s]
 96%|█████████▌| 356412/371472 [6:12:14<1:09:42,  3.60it/s]
 96%|█████████▌| 356413/371472 [6:12:15<1:10:22,  3.57it/s]
 96%|█████████▌| 356414/371472 [6:12:15<1:10:22,  3.57it/s]
 96%|█████████▌| 356415/371472 [6:12:15<1:10:20,  3.57it/s]
 96%|█████████▌| 356416/371472 [6:12:15<1:09:25,  3.61it/s]
 96%|█████████▌| 356417/371472 [6:12:16<1:09:15,  3.62it/s]
 96%|█████████▌| 356418/371472 [6:12:16<1:11:13,  3.52it/s]
 96%|█████████▌| 356419/371472 [6:12:16<1:11:40,  3.50it/s]
 96%|█████████▌| 356420/371472 [6:12:17<1:11:30,  3.51it/s]
                                                           
{'loss': 2.6773, 'learning_rate': 1.3648753474541576e-07, 'epoch': 15.35}
+
 96%|█████████▌| 356420/371472 [6:12:17<1:11:30,  3.51it/s]
 96%|█████████▌| 356421/371472 [6:12:17<1:08:32,  3.66it/s]
 96%|█████████▌| 356422/371472 [6:12:17<1:07:36,  3.71it/s]
 96%|█████████▌| 356423/371472 [6:12:17<1:11:45,  3.50it/s]
 96%|█████████▌| 356424/371472 [6:12:18<1:08:22,  3.67it/s]
 96%|█████████▌| 356425/371472 [6:12:18<1:10:24,  3.56it/s]
 96%|█████████▌| 356426/371472 [6:12:18<1:16:13,  3.29it/s]
 96%|█████████▌| 356427/371472 [6:12:19<1:12:46,  3.45it/s]
 96%|█████████▌| 356428/371472 [6:12:19<1:10:54,  3.54it/s]
 96%|█████████▌| 356429/371472 [6:12:19<1:11:28,  3.51it/s]
 96%|█████████▌| 356430/371472 [6:12:19<1:10:46,  3.54it/s]
 96%|█████████▌| 356431/371472 [6:12:20<1:14:38,  3.36it/s]
 96%|█████████▌| 356432/371472 [6:12:20<1:14:18,  3.37it/s]
 96%|█████████▌| 356433/371472 [6:12:20<1:10:31,  3.55it/s]
 96%|█████████▌| 356434/371472 [6:12:21<1:09:22,  3.61it/s]
 96%|█████████▌| 356435/371472 [6:12:21<1:08:34,  3.65it/s]
 96%|█████████▌| 356436/371472 [6:12:21<1:12:53,  3.44it/s]
 96%|█████████▌| 356437/371472 [6:12:21<1:13:05,  3.43it/s]
 96%|█████████▌| 356438/371472 [6:12:22<1:12:56,  3.44it/s]
 96%|███��█████▌| 356439/371472 [6:12:22<1:13:27,  3.41it/s]
 96%|█████████▌| 356440/371472 [6:12:22<1:12:27,  3.46it/s]
                                                           
{'loss': 2.6087, 'learning_rate': 1.3643905276993683e-07, 'epoch': 15.35}
+
 96%|█████████▌| 356440/371472 [6:12:22<1:12:27,  3.46it/s]
 96%|█████████▌| 356441/371472 [6:12:23<1:10:09,  3.57it/s]
 96%|█████████▌| 356442/371472 [6:12:23<1:10:25,  3.56it/s]
 96%|█████████▌| 356443/371472 [6:12:23<1:08:55,  3.63it/s]
 96%|█████████▌| 356444/371472 [6:12:23<1:09:15,  3.62it/s]
 96%|█████████▌| 356445/371472 [6:12:24<1:13:09,  3.42it/s]
 96%|█████████▌| 356446/371472 [6:12:24<1:13:57,  3.39it/s]
 96%|█████████▌| 356447/371472 [6:12:24<1:13:12,  3.42it/s]
 96%|█████████▌| 356448/371472 [6:12:25<1:13:01,  3.43it/s]
 96%|█████████▌| 356449/371472 [6:12:25<1:12:12,  3.47it/s]
 96%|█████████▌| 356450/371472 [6:12:25<1:09:29,  3.60it/s]
 96%|█████████▌| 356451/371472 [6:12:25<1:12:36,  3.45it/s]
 96%|█████████▌| 356452/371472 [6:12:26<1:09:51,  3.58it/s]
 96%|█████████▌| 356453/371472 [6:12:26<1:09:10,  3.62it/s]
 96%|█████████▌| 356454/371472 [6:12:26<1:11:18,  3.51it/s]
 96%|█████████▌| 356455/371472 [6:12:27<1:10:35,  3.55it/s]
 96%|█████████▌| 356456/371472 [6:12:27<1:11:27,  3.50it/s]
 96%|█████████▌| 356457/371472 [6:12:27<1:11:56,  3.48it/s]
 96%|█████████▌| 356458/371472 [6:12:27<1:12:11,  3.47it/s]
 96%|█████████▌| 356459/371472 [6:12:28<1:10:54,  3.53it/s]
 96%|█████████▌| 356460/371472 [6:12:28<1:19:08,  3.16it/s]
                                                           
{'loss': 2.5368, 'learning_rate': 1.3639057079445798e-07, 'epoch': 15.35}
+
 96%|█████████▌| 356460/371472 [6:12:28<1:19:08,  3.16it/s]
 96%|█████████▌| 356461/371472 [6:12:28<1:16:29,  3.27it/s]
 96%|█████████▌| 356462/371472 [6:12:29<1:14:58,  3.34it/s]
 96%|█████████▌| 356463/371472 [6:12:29<1:14:17,  3.37it/s]
 96%|█████████▌| 356464/371472 [6:12:29<1:12:51,  3.43it/s]
 96%|█████████▌| 356465/371472 [6:12:29<1:11:41,  3.49it/s]
 96%|█████████▌| 356466/371472 [6:12:30<1:10:24,  3.55it/s]
 96%|█████████▌| 356467/371472 [6:12:30<1:08:16,  3.66it/s]
 96%|█████████▌| 356468/371472 [6:12:30<1:08:08,  3.67it/s]
 96%|█████████▌| 356469/371472 [6:12:31<1:09:56,  3.57it/s]
 96%|█████████▌| 356470/371472 [6:12:31<1:08:54,  3.63it/s]
 96%|█████████▌| 356471/371472 [6:12:31<1:09:12,  3.61it/s]
 96%|█████████▌| 356472/371472 [6:12:31<1:08:44,  3.64it/s]
 96%|█████████▌| 356473/371472 [6:12:32<1:08:42,  3.64it/s]
 96%|█████████▌| 356474/371472 [6:12:32<1:08:04,  3.67it/s]
 96%|█████████▌| 356475/371472 [6:12:32<1:07:25,  3.71it/s]
 96%|█████████▌| 356476/371472 [6:12:32<1:07:14,  3.72it/s]
 96%|█████████▌| 356477/371472 [6:12:33<1:05:56,  3.79it/s]
 96%|█████████▌| 356478/371472 [6:12:33<1:08:01,  3.67it/s]
 96%|█████████▌| 356479/371472 [6:12:33<1:05:54,  3.79it/s]
 96%|█████████▌| 356480/371472 [6:12:34<1:06:11,  3.77it/s]
                                                           
{'loss': 2.5748, 'learning_rate': 1.3634208881897902e-07, 'epoch': 15.35}
+
 96%|█████████▌| 356480/371472 [6:12:34<1:06:11,  3.77it/s]
 96%|█████████▌| 356481/371472 [6:12:34<1:07:50,  3.68it/s]
 96%|█████████▌| 356482/371472 [6:12:34<1:10:23,  3.55it/s]
 96%|█████████▌| 356483/371472 [6:12:34<1:08:06,  3.67it/s]
 96%|█████████▌| 356484/371472 [6:12:35<1:07:53,  3.68it/s]
 96%|█████████▌| 356485/371472 [6:12:35<1:19:10,  3.16it/s]
 96%|█████████▌| 356486/371472 [6:12:35<1:21:52,  3.05it/s]
 96%|█████████▌| 356487/371472 [6:12:36<1:17:36,  3.22it/s]
 96%|█████████▌| 356488/371472 [6:12:36<1:13:36,  3.39it/s]
 96%|█████████▌| 356489/371472 [6:12:36<1:13:44,  3.39it/s]
 96%|█████████▌| 356490/371472 [6:12:37<1:14:02,  3.37it/s]
 96%|█████████▌| 356491/371472 [6:12:37<1:12:36,  3.44it/s]
 96%|█████████▌| 356492/371472 [6:12:37<1:10:21,  3.55it/s]
 96%|█████████▌| 356493/371472 [6:12:37<1:11:13,  3.51it/s]
 96%|█████████▌| 356494/371472 [6:12:38<1:08:15,  3.66it/s]
 96%|█████████▌| 356495/371472 [6:12:38<1:09:10,  3.61it/s]
 96%|█████████▌| 356496/371472 [6:12:38<1:09:33,  3.59it/s]
 96%|█████████▌| 356497/371472 [6:12:38<1:07:47,  3.68it/s]
 96%|█████████▌| 356498/371472 [6:12:39<1:06:15,  3.77it/s]
 96%|█████████▌| 356499/371472 [6:12:39<1:09:00,  3.62it/s]
 96%|█████████▌| 356500/371472 [6:12:39<1:06:34,  3.75it/s]
                                                           
{'loss': 2.6974, 'learning_rate': 1.362936068435002e-07, 'epoch': 15.36}
+
 96%|█████████▌| 356500/371472 [6:12:39<1:06:34,  3.75it/s]
 96%|█████████▌| 356501/371472 [6:12:40<1:07:05,  3.72it/s]
 96%|█████████▌| 356502/371472 [6:12:40<1:10:59,  3.51it/s]
 96%|█████████▌| 356503/371472 [6:12:40<1:12:13,  3.45it/s]
 96%|█████████▌| 356504/371472 [6:12:40<1:14:06,  3.37it/s]
 96%|█████████▌| 356505/371472 [6:12:41<1:17:20,  3.23it/s]
 96%|█████████▌| 356506/371472 [6:12:41<1:13:08,  3.41it/s]
 96%|█████████▌| 356507/371472 [6:12:41<1:11:21,  3.50it/s]
 96%|█████████▌| 356508/371472 [6:12:42<1:15:38,  3.30it/s]
 96%|█████████▌| 356509/371472 [6:12:42<1:14:23,  3.35it/s]
 96%|█████████▌| 356510/371472 [6:12:42<1:18:59,  3.16it/s]
 96%|█████████▌| 356511/371472 [6:12:43<1:15:09,  3.32it/s]
 96%|█████████▌| 356512/371472 [6:12:43<1:13:07,  3.41it/s]
 96%|█████████▌| 356513/371472 [6:12:43<1:12:58,  3.42it/s]
 96%|█████████▌| 356514/371472 [6:12:43<1:11:17,  3.50it/s]
 96%|█████████▌| 356515/371472 [6:12:44<1:09:24,  3.59it/s]
 96%|█████████▌| 356516/371472 [6:12:44<1:07:25,  3.70it/s]
 96%|█████████▌| 356517/371472 [6:12:44<1:06:26,  3.75it/s]
 96%|█████████▌| 356518/371472 [6:12:44<1:06:22,  3.75it/s]
 96%|█████████▌| 356519/371472 [6:12:45<1:05:37,  3.80it/s]
 96%|█████████▌| 356520/371472 [6:12:45<1:04:36,  3.86it/s]
                                                           
{'loss': 2.4561, 'learning_rate': 1.3624512486802125e-07, 'epoch': 15.36}
+
 96%|█████████▌| 356520/371472 [6:12:45<1:04:36,  3.86it/s]
 96%|█████████▌| 356521/371472 [6:12:45<1:04:57,  3.84it/s]
 96%|█████████▌| 356522/371472 [6:12:45<1:03:32,  3.92it/s]
 96%|█████████▌| 356523/371472 [6:12:46<1:05:21,  3.81it/s]
 96%|█████████▌| 356524/371472 [6:12:46<1:05:34,  3.80it/s]
 96%|█████████▌| 356525/371472 [6:12:46<1:08:21,  3.64it/s]
 96%|█████████▌| 356526/371472 [6:12:47<1:08:18,  3.65it/s]
 96%|█████████▌| 356527/371472 [6:12:47<1:06:18,  3.76it/s]
 96%|█████████▌| 356528/371472 [6:12:47<1:05:36,  3.80it/s]
 96%|█████████▌| 356529/371472 [6:12:47<1:04:54,  3.84it/s]
 96%|█████████▌| 356530/371472 [6:12:48<1:06:23,  3.75it/s]
 96%|█████████▌| 356531/371472 [6:12:48<1:04:34,  3.86it/s]
 96%|█████████▌| 356532/371472 [6:12:48<1:04:37,  3.85it/s]
 96%|█████████▌| 356533/371472 [6:12:48<1:06:31,  3.74it/s]
 96%|█████████▌| 356534/371472 [6:12:49<1:11:50,  3.47it/s]
 96%|█████████▌| 356535/371472 [6:12:49<1:09:59,  3.56it/s]
 96%|█████████▌| 356536/371472 [6:12:49<1:08:29,  3.63it/s]
 96%|█████████▌| 356537/371472 [6:12:50<1:14:42,  3.33it/s]
 96%|█████████▌| 356538/371472 [6:12:50<1:11:08,  3.50it/s]
 96%|█████████▌| 356539/371472 [6:12:50<1:08:19,  3.64it/s]
 96%|█████████▌| 356540/371472 [6:12:50<1:06:30,  3.74it/s]
                                                           
{'loss': 2.6078, 'learning_rate': 1.361966428925424e-07, 'epoch': 15.36}
+
 96%|█████████▌| 356540/371472 [6:12:50<1:06:30,  3.74it/s]
 96%|█████████▌| 356541/371472 [6:12:51<1:07:04,  3.71it/s]
 96%|█████████▌| 356542/371472 [6:12:51<1:09:15,  3.59it/s]
 96%|█████████▌| 356543/371472 [6:12:51<1:07:30,  3.69it/s]
 96%|█████████▌| 356544/371472 [6:12:51<1:06:41,  3.73it/s]
 96%|█████████▌| 356545/371472 [6:12:52<1:06:34,  3.74it/s]
 96%|█████████▌| 356546/371472 [6:12:52<1:05:06,  3.82it/s]
 96%|█████████▌| 356547/371472 [6:12:52<1:03:50,  3.90it/s]
 96%|█████████▌| 356548/371472 [6:12:52<1:03:30,  3.92it/s]
 96%|█████████▌| 356549/371472 [6:12:53<1:09:17,  3.59it/s]
 96%|█████████▌| 356550/371472 [6:12:53<1:13:44,  3.37it/s]
 96%|█████████▌| 356551/371472 [6:12:53<1:15:11,  3.31it/s]
 96%|█████████▌| 356552/371472 [6:12:54<1:17:56,  3.19it/s]
 96%|█████████▌| 356553/371472 [6:12:54<1:14:17,  3.35it/s]
 96%|█████████▌| 356554/371472 [6:12:54<1:16:35,  3.25it/s]
 96%|█████████▌| 356555/371472 [6:12:55<1:16:40,  3.24it/s]
 96%|█████████▌| 356556/371472 [6:12:55<1:14:59,  3.32it/s]
 96%|█████████▌| 356557/371472 [6:12:55<1:19:57,  3.11it/s]
 96%|█████████▌| 356558/371472 [6:12:56<1:19:58,  3.11it/s]
 96%|█████████▌| 356559/371472 [6:12:56<1:21:13,  3.06it/s]
 96%|█████████▌| 356560/371472 [6:12:56<1:18:23,  3.17it/s]
                                                           
{'loss': 2.5544, 'learning_rate': 1.3614816091706347e-07, 'epoch': 15.36}
+
 96%|█████████▌| 356560/371472 [6:12:56<1:18:23,  3.17it/s]
 96%|█████████▌| 356561/371472 [6:12:57<1:16:23,  3.25it/s]
 96%|█████████▌| 356562/371472 [6:12:57<1:13:09,  3.40it/s]
 96%|█████████▌| 356563/371472 [6:12:57<1:11:49,  3.46it/s]
 96%|█████████▌| 356564/371472 [6:12:58<1:17:27,  3.21it/s]
 96%|█████████▌| 356565/371472 [6:12:58<1:14:55,  3.32it/s]
 96%|█████████▌| 356566/371472 [6:12:58<1:15:05,  3.31it/s]
 96%|█████████▌| 356567/371472 [6:12:58<1:15:23,  3.30it/s]
 96%|█████████▌| 356568/371472 [6:12:59<1:13:22,  3.39it/s]
 96%|█████████▌| 356569/371472 [6:12:59<1:10:46,  3.51it/s]
 96%|█████████▌| 356570/371472 [6:12:59<1:09:45,  3.56it/s]
 96%|█████████▌| 356571/371472 [6:13:00<1:15:08,  3.30it/s]
 96%|█████████▌| 356572/371472 [6:13:00<1:10:56,  3.50it/s]
 96%|█████████▌| 356573/371472 [6:13:00<1:09:28,  3.57it/s]
 96%|█████████▌| 356574/371472 [6:13:00<1:08:03,  3.65it/s]
 96%|█████████▌| 356575/371472 [6:13:01<1:06:46,  3.72it/s]
 96%|█████████▌| 356576/371472 [6:13:01<1:09:03,  3.60it/s]
 96%|█████████▌| 356577/371472 [6:13:01<1:08:52,  3.60it/s]
 96%|█████████▌| 356578/371472 [6:13:01<1:11:29,  3.47it/s]
 96%|█████████▌| 356579/371472 [6:13:02<1:09:02,  3.60it/s]
 96%|█████████▌| 356580/371472 [6:13:02<1:07:10,  3.69it/s]
                                                           
{'loss': 2.7059, 'learning_rate': 1.3609967894158462e-07, 'epoch': 15.36}
+
 96%|█████████▌| 356580/371472 [6:13:02<1:07:10,  3.69it/s]
 96%|█████████▌| 356581/371472 [6:13:02<1:11:19,  3.48it/s]
 96%|█████████▌| 356582/371472 [6:13:03<1:11:06,  3.49it/s]
 96%|█████████▌| 356583/371472 [6:13:03<1:08:23,  3.63it/s]
 96%|█████████▌| 356584/371472 [6:13:03<1:10:21,  3.53it/s]
 96%|█████████▌| 356585/371472 [6:13:04<1:17:23,  3.21it/s]
 96%|█████████▌| 356586/371472 [6:13:04<1:13:29,  3.38it/s]
 96%|█████████▌| 356587/371472 [6:13:04<1:13:08,  3.39it/s]
 96%|█████████▌| 356588/371472 [6:13:04<1:12:15,  3.43it/s]
 96%|█████████▌| 356589/371472 [6:13:05<1:08:49,  3.60it/s]
 96%|█████████▌| 356590/371472 [6:13:05<1:05:53,  3.76it/s]
 96%|█████████▌| 356591/371472 [6:13:05<1:07:39,  3.67it/s]
 96%|█████████▌| 356592/371472 [6:13:05<1:07:03,  3.70it/s]
 96%|█████████▌| 356593/371472 [6:13:06<1:04:53,  3.82it/s]
 96%|█████████▌| 356594/371472 [6:13:06<1:12:36,  3.41it/s]
 96%|█████████▌| 356595/371472 [6:13:06<1:17:21,  3.20it/s]
 96%|█████████▌| 356596/371472 [6:13:07<1:16:10,  3.25it/s]
 96%|█████████▌| 356597/371472 [6:13:07<1:20:52,  3.07it/s]
 96%|█████████▌| 356598/371472 [6:13:07<1:17:09,  3.21it/s]
 96%|█████████▌| 356599/371472 [6:13:08<1:15:55,  3.26it/s]
 96%|█████████▌| 356600/371472 [6:13:08<1:11:48,  3.45it/s]
                                                           
{'loss': 2.6965, 'learning_rate': 1.360511969661057e-07, 'epoch': 15.36}
+
 96%|█████████▌| 356600/371472 [6:13:08<1:11:48,  3.45it/s]
 96%|█████████▌| 356601/371472 [6:13:08<1:09:28,  3.57it/s]
 96%|█████████▌| 356602/371472 [6:13:08<1:15:50,  3.27it/s]
 96%|█████████▌| 356603/371472 [6:13:09<1:15:10,  3.30it/s]
 96%|█████████▌| 356604/371472 [6:13:09<1:17:40,  3.19it/s]
 96%|█████████▌| 356605/371472 [6:13:09<1:16:06,  3.26it/s]
 96%|█████████▌| 356606/371472 [6:13:10<1:15:42,  3.27it/s]
 96%|█████████▌| 356607/371472 [6:13:10<1:11:30,  3.46it/s]
 96%|█████████▌| 356608/371472 [6:13:10<1:11:44,  3.45it/s]
 96%|█████████▌| 356609/371472 [6:13:11<1:12:03,  3.44it/s]
 96%|█████████▌| 356610/371472 [6:13:11<1:11:55,  3.44it/s]
 96%|█████████▌| 356611/371472 [6:13:11<1:19:00,  3.14it/s]
 96%|█████████▌| 356612/371472 [6:13:11<1:16:23,  3.24it/s]
 96%|█████████▌| 356613/371472 [6:13:12<1:12:45,  3.40it/s]
 96%|█████████▌| 356614/371472 [6:13:12<1:10:57,  3.49it/s]
 96%|█████████▌| 356615/371472 [6:13:12<1:09:03,  3.59it/s]
 96%|█████████▌| 356616/371472 [6:13:13<1:12:32,  3.41it/s]
 96%|█████████▌| 356617/371472 [6:13:13<1:10:29,  3.51it/s]
 96%|█████████▌| 356618/371472 [6:13:13<1:13:33,  3.37it/s]
 96%|█████████▌| 356619/371472 [6:13:13<1:13:35,  3.36it/s]
 96%|█████████▌| 356620/371472 [6:13:14<1:11:51,  3.44it/s]
                                                           
{'loss': 2.7667, 'learning_rate': 1.3600271499062684e-07, 'epoch': 15.36}
+
 96%|█████████▌| 356620/371472 [6:13:14<1:11:51,  3.44it/s]
 96%|█████████▌| 356621/371472 [6:13:14<1:15:07,  3.30it/s]
 96%|█████████▌| 356622/371472 [6:13:14<1:13:43,  3.36it/s]
 96%|█████████▌| 356623/371472 [6:13:15<1:14:37,  3.32it/s]
 96%|█████████▌| 356624/371472 [6:13:15<1:13:13,  3.38it/s]
 96%|█████████▌| 356625/371472 [6:13:15<1:19:14,  3.12it/s]
 96%|█████████▌| 356626/371472 [6:13:16<1:19:57,  3.09it/s]
 96%|█████████▌| 356627/371472 [6:13:16<1:17:43,  3.18it/s]
 96%|█████████▌| 356628/371472 [6:13:16<1:16:23,  3.24it/s]
 96%|█████████▌| 356629/371472 [6:13:17<1:12:14,  3.42it/s]
 96%|█████████▌| 356630/371472 [6:13:17<1:09:08,  3.58it/s]
 96%|█████████▌| 356631/371472 [6:13:17<1:17:15,  3.20it/s]
 96%|█████████▌| 356632/371472 [6:13:17<1:14:59,  3.30it/s]
 96%|█████████▌| 356633/371472 [6:13:18<1:16:07,  3.25it/s]
 96%|█████████▌| 356634/371472 [6:13:18<1:11:55,  3.44it/s]
 96%|█████████▌| 356635/371472 [6:13:18<1:08:14,  3.62it/s]
 96%|█████████▌| 356636/371472 [6:13:19<1:06:38,  3.71it/s]
 96%|█████████▌| 356637/371472 [6:13:19<1:09:40,  3.55it/s]
 96%|█████████▌| 356638/371472 [6:13:19<1:09:03,  3.58it/s]
 96%|█████████▌| 356639/371472 [6:13:19<1:06:00,  3.75it/s]
 96%|█████████▌| 356640/371472 [6:13:20<1:08:21,  3.62it/s]
                                                           
{'loss': 2.613, 'learning_rate': 1.359542330151479e-07, 'epoch': 15.36}
+
 96%|█████████▌| 356640/371472 [6:13:20<1:08:21,  3.62it/s]
 96%|█████████▌| 356641/371472 [6:13:20<1:08:13,  3.62it/s]
 96%|█████████▌| 356642/371472 [6:13:20<1:14:59,  3.30it/s]
 96%|█████████▌| 356643/371472 [6:13:21<1:15:30,  3.27it/s]
 96%|█████████▌| 356644/371472 [6:13:21<1:12:40,  3.40it/s]
 96%|█████████▌| 356645/371472 [6:13:21<1:11:02,  3.48it/s]
 96%|█████████▌| 356646/371472 [6:13:21<1:12:49,  3.39it/s]
 96%|█████████▌| 356647/371472 [6:13:22<1:11:10,  3.47it/s]
 96%|█████████▌| 356648/371472 [6:13:22<1:08:55,  3.58it/s]
 96%|█████████▌| 356649/371472 [6:13:22<1:08:47,  3.59it/s]
 96%|█████████▌| 356650/371472 [6:13:23<1:06:43,  3.70it/s]
 96%|█████████▌| 356651/371472 [6:13:23<1:04:57,  3.80it/s]
 96%|█████████▌| 356652/371472 [6:13:23<1:08:46,  3.59it/s]
 96%|█████████▌| 356653/371472 [6:13:23<1:10:17,  3.51it/s]
 96%|█████████▌| 356654/371472 [6:13:24<1:08:56,  3.58it/s]
 96%|█████████▌| 356655/371472 [6:13:24<1:07:29,  3.66it/s]
 96%|█████████▌| 356656/371472 [6:13:24<1:07:05,  3.68it/s]
 96%|█████████▌| 356657/371472 [6:13:24<1:10:16,  3.51it/s]
 96%|█████████▌| 356658/371472 [6:13:25<1:13:00,  3.38it/s]
 96%|█████████▌| 356659/371472 [6:13:25<1:12:55,  3.39it/s]
 96%|█████████▌| 356660/371472 [6:13:25<1:11:34,  3.45it/s]
                                                           
{'loss': 2.4522, 'learning_rate': 1.3590575103966906e-07, 'epoch': 15.36}
+
 96%|█████████▌| 356660/371472 [6:13:25<1:11:34,  3.45it/s]
 96%|█████████▌| 356661/371472 [6:13:26<1:10:29,  3.50it/s]
 96%|█████████▌| 356662/371472 [6:13:26<1:08:00,  3.63it/s]
 96%|█████████▌| 356663/371472 [6:13:26<1:11:37,  3.45it/s]
 96%|█████████▌| 356664/371472 [6:13:26<1:09:07,  3.57it/s]
 96%|█████████▌| 356665/371472 [6:13:27<1:08:37,  3.60it/s]
 96%|█████████▌| 356666/371472 [6:13:27<1:11:23,  3.46it/s]
 96%|█████████▌| 356667/371472 [6:13:27<1:13:56,  3.34it/s]
 96%|█████████▌| 356668/371472 [6:13:28<1:11:29,  3.45it/s]
 96%|█████████▌| 356669/371472 [6:13:28<1:20:13,  3.08it/s]
 96%|█████████▌| 356670/371472 [6:13:28<1:16:53,  3.21it/s]
 96%|█████████▌| 356671/371472 [6:13:29<1:11:52,  3.43it/s]
 96%|█████████▌| 356672/371472 [6:13:29<1:10:29,  3.50it/s]
 96%|█████████▌| 356673/371472 [6:13:29<1:12:10,  3.42it/s]
 96%|█████████▌| 356674/371472 [6:13:29<1:09:39,  3.54it/s]
 96%|█████████▌| 356675/371472 [6:13:30<1:08:50,  3.58it/s]
 96%|█████████▌| 356676/371472 [6:13:30<1:07:01,  3.68it/s]
 96%|█████████▌| 356677/371472 [6:13:30<1:05:59,  3.74it/s]
 96%|█████████▌| 356678/371472 [6:13:30<1:05:54,  3.74it/s]
 96%|█████████▌| 356679/371472 [6:13:31<1:09:13,  3.56it/s]
 96%|█████████▌| 356680/371472 [6:13:31<1:08:56,  3.58it/s]
                                                           
{'loss': 2.4585, 'learning_rate': 1.358572690641901e-07, 'epoch': 15.36}
+
 96%|█████████▌| 356680/371472 [6:13:31<1:08:56,  3.58it/s]
 96%|█████████▌| 356681/371472 [6:13:31<1:07:29,  3.65it/s]
 96%|█████████▌| 356682/371472 [6:13:32<1:05:36,  3.76it/s]
 96%|█████████▌| 356683/371472 [6:13:32<1:13:46,  3.34it/s]
 96%|█████████▌| 356684/371472 [6:13:32<1:11:41,  3.44it/s]
 96%|█████████▌| 356685/371472 [6:13:33<1:11:52,  3.43it/s]
 96%|█████████▌| 356686/371472 [6:13:33<1:09:16,  3.56it/s]
 96%|█████████▌| 356687/371472 [6:13:33<1:09:26,  3.55it/s]
 96%|█████████▌| 356688/371472 [6:13:33<1:06:05,  3.73it/s]
 96%|█████████▌| 356689/371472 [6:13:34<1:09:30,  3.54it/s]
 96%|█████████▌| 356690/371472 [6:13:34<1:07:02,  3.68it/s]
 96%|█████████▌| 356691/371472 [6:13:34<1:05:44,  3.75it/s]
 96%|█████████▌| 356692/371472 [6:13:34<1:08:35,  3.59it/s]
 96%|█████████▌| 356693/371472 [6:13:35<1:07:27,  3.65it/s]
 96%|█████████▌| 356694/371472 [6:13:35<1:11:34,  3.44it/s]
 96%|█████████▌| 356695/371472 [6:13:35<1:08:45,  3.58it/s]
 96%|█████████▌| 356696/371472 [6:13:36<1:07:45,  3.63it/s]
 96%|█████████▌| 356697/371472 [6:13:36<1:07:45,  3.63it/s]
 96%|█████████▌| 356698/371472 [6:13:36<1:07:11,  3.66it/s]
 96%|█████████▌| 356699/371472 [6:13:36<1:07:42,  3.64it/s]
 96%|█████████▌| 356700/371472 [6:13:37<1:07:38,  3.64it/s]
                                                           
{'loss': 2.6564, 'learning_rate': 1.3580878708871129e-07, 'epoch': 15.36}
+
 96%|█████████▌| 356700/371472 [6:13:37<1:07:38,  3.64it/s]
 96%|█████████▌| 356701/371472 [6:13:37<1:05:54,  3.73it/s]
 96%|█████████▌| 356702/371472 [6:13:37<1:05:45,  3.74it/s]
 96%|█████████▌| 356703/371472 [6:13:37<1:12:14,  3.41it/s]
 96%|█████████▌| 356704/371472 [6:13:38<1:13:20,  3.36it/s]
 96%|█████████▌| 356705/371472 [6:13:38<1:11:42,  3.43it/s]
 96%|█████████▌| 356706/371472 [6:13:38<1:09:52,  3.52it/s]
 96%|█████████▌| 356707/371472 [6:13:39<1:08:11,  3.61it/s]
 96%|█████████▌| 356708/371472 [6:13:39<1:08:02,  3.62it/s]
 96%|█████████▌| 356709/371472 [6:13:39<1:08:35,  3.59it/s]
 96%|█████████▌| 356710/371472 [6:13:39<1:07:25,  3.65it/s]
 96%|█████████▌| 356711/371472 [6:13:40<1:07:08,  3.66it/s]
 96%|█████████▌| 356712/371472 [6:13:40<1:06:01,  3.73it/s]
 96%|█████████▌| 356713/371472 [6:13:40<1:05:25,  3.76it/s]
 96%|█████████▌| 356714/371472 [6:13:40<1:03:48,  3.85it/s]
 96%|█████████▌| 356715/371472 [6:13:41<1:04:16,  3.83it/s]
 96%|█████████▌| 356716/371472 [6:13:41<1:03:45,  3.86it/s]
 96%|█████████▌| 356717/371472 [6:13:41<1:06:40,  3.69it/s]
 96%|█████████▌| 356718/371472 [6:13:42<1:07:21,  3.65it/s]
 96%|█████████▌| 356719/371472 [6:13:42<1:07:33,  3.64it/s]
 96%|█████████▌| 356720/371472 [6:13:42<1:10:46,  3.47it/s]
                                                           
{'loss': 2.6827, 'learning_rate': 1.3576030511323233e-07, 'epoch': 15.36}
+
 96%|█████████▌| 356720/371472 [6:13:42<1:10:46,  3.47it/s]
 96%|█████████▌| 356721/371472 [6:13:42<1:09:32,  3.54it/s]
 96%|█████████▌| 356722/371472 [6:13:43<1:07:08,  3.66it/s]
 96%|█████████▌| 356723/371472 [6:13:43<1:09:54,  3.52it/s]
 96%|█████████▌| 356724/371472 [6:13:43<1:08:39,  3.58it/s]
 96%|█████████▌| 356725/371472 [6:13:44<1:13:13,  3.36it/s]
 96%|█████████▌| 356726/371472 [6:13:44<1:12:28,  3.39it/s]
 96%|█████████▌| 356727/371472 [6:13:44<1:11:44,  3.43it/s]
 96%|█████████▌| 356728/371472 [6:13:44<1:10:19,  3.49it/s]
 96%|█████████▌| 356729/371472 [6:13:45<1:11:42,  3.43it/s]
 96%|█████████▌| 356730/371472 [6:13:45<1:08:24,  3.59it/s]
 96%|█████████▌| 356731/371472 [6:13:45<1:06:44,  3.68it/s]
 96%|█████████▌| 356732/371472 [6:13:46<1:05:53,  3.73it/s]
 96%|█████████▌| 356733/371472 [6:13:46<1:10:40,  3.48it/s]
 96%|█████████▌| 356734/371472 [6:13:46<1:10:13,  3.50it/s]
 96%|█████████▌| 356735/371472 [6:13:46<1:14:42,  3.29it/s]
 96%|█████████▌| 356736/371472 [6:13:47<1:12:08,  3.40it/s]
 96%|█████████▌| 356737/371472 [6:13:47<1:18:16,  3.14it/s]
 96%|█████████▌| 356738/371472 [6:13:47<1:13:38,  3.33it/s]
 96%|█████████▌| 356739/371472 [6:13:48<1:10:13,  3.50it/s]
 96%|█████████▌| 356740/371472 [6:13:48<1:12:32,  3.38it/s]
                                                           
{'loss': 2.4151, 'learning_rate': 1.3571182313775348e-07, 'epoch': 15.37}
+
 96%|█████████▌| 356740/371472 [6:13:48<1:12:32,  3.38it/s]
 96%|█████████▌| 356741/371472 [6:13:48<1:13:26,  3.34it/s]
 96%|█████████▌| 356742/371472 [6:13:49<1:11:04,  3.45it/s]
 96%|█████████▌| 356743/371472 [6:13:49<1:13:25,  3.34it/s]
 96%|█████████▌| 356744/371472 [6:13:49<1:11:06,  3.45it/s]
 96%|█████████▌| 356745/371472 [6:13:49<1:07:57,  3.61it/s]
 96%|█████████▌| 356746/371472 [6:13:50<1:11:59,  3.41it/s]
 96%|█████████▌| 356747/371472 [6:13:50<1:10:19,  3.49it/s]
 96%|█████████▌| 356748/371472 [6:13:50<1:10:00,  3.51it/s]
 96%|█████████▌| 356749/371472 [6:13:51<1:08:11,  3.60it/s]
 96%|█████████▌| 356750/371472 [6:13:51<1:06:40,  3.68it/s]
 96%|█████████▌| 356751/371472 [6:13:51<1:04:59,  3.78it/s]
 96%|█████████▌| 356752/371472 [6:13:51<1:03:29,  3.86it/s]
 96%|█████████▌| 356753/371472 [6:13:52<1:03:31,  3.86it/s]
 96%|█████████▌| 356754/371472 [6:13:52<1:03:37,  3.86it/s]
 96%|█████████▌| 356755/371472 [6:13:52<1:04:38,  3.79it/s]
 96%|█████████▌| 356756/371472 [6:13:52<1:04:19,  3.81it/s]
 96%|█████████▌| 356757/371472 [6:13:53<1:04:39,  3.79it/s]
 96%|█████████▌| 356758/371472 [6:13:53<1:07:06,  3.65it/s]
 96%|█████████▌| 356759/371472 [6:13:53<1:05:59,  3.72it/s]
 96%|█████████▌| 356760/371472 [6:13:53<1:08:55,  3.56it/s]
                                                           
{'loss': 2.5558, 'learning_rate': 1.3566334116227455e-07, 'epoch': 15.37}
+
 96%|█████████▌| 356760/371472 [6:13:53<1:08:55,  3.56it/s]
 96%|█████████▌| 356761/371472 [6:13:54<1:06:29,  3.69it/s]
 96%|█████████▌| 356762/371472 [6:13:54<1:06:06,  3.71it/s]
 96%|█████████▌| 356763/371472 [6:13:54<1:10:28,  3.48it/s]
 96%|█████████▌| 356764/371472 [6:13:55<1:08:06,  3.60it/s]
 96%|█████████▌| 356765/371472 [6:13:55<1:10:11,  3.49it/s]
 96%|█████████▌| 356766/371472 [6:13:55<1:08:08,  3.60it/s]
 96%|█████████▌| 356767/371472 [6:13:55<1:07:46,  3.62it/s]
 96%|█████████▌| 356768/371472 [6:13:56<1:07:05,  3.65it/s]
 96%|█████████▌| 356769/371472 [6:13:56<1:05:37,  3.73it/s]
 96%|█████████▌| 356770/371472 [6:13:56<1:07:15,  3.64it/s]
 96%|█████████▌| 356771/371472 [6:13:57<1:10:49,  3.46it/s]
 96%|█████████▌| 356772/371472 [6:13:57<1:11:21,  3.43it/s]
 96%|█████████▌| 356773/371472 [6:13:57<1:13:32,  3.33it/s]
 96%|█████████▌| 356774/371472 [6:13:57<1:15:00,  3.27it/s]
 96%|█████████▌| 356775/371472 [6:13:58<1:11:15,  3.44it/s]
 96%|█████████▌| 356776/371472 [6:13:58<1:10:47,  3.46it/s]
 96%|████████���▌| 356777/371472 [6:13:58<1:09:46,  3.51it/s]
 96%|█████████▌| 356778/371472 [6:13:59<1:07:21,  3.64it/s]
 96%|█████████▌| 356779/371472 [6:13:59<1:11:24,  3.43it/s]
 96%|█████████▌| 356780/371472 [6:13:59<1:08:37,  3.57it/s]
                                                           
{'loss': 2.5457, 'learning_rate': 1.356148591867957e-07, 'epoch': 15.37}
+
 96%|█████████▌| 356780/371472 [6:13:59<1:08:37,  3.57it/s]
 96%|█████████▌| 356781/371472 [6:13:59<1:10:50,  3.46it/s]
 96%|█████████▌| 356782/371472 [6:14:00<1:11:00,  3.45it/s]
 96%|█████████▌| 356783/371472 [6:14:00<1:11:22,  3.43it/s]
 96%|█████████▌| 356784/371472 [6:14:00<1:07:16,  3.64it/s]
 96%|█████████▌| 356785/371472 [6:14:01<1:09:58,  3.50it/s]
 96%|█████████▌| 356786/371472 [6:14:01<1:07:45,  3.61it/s]
 96%|█████████▌| 356787/371472 [6:14:01<1:09:17,  3.53it/s]
 96%|█████████▌| 356788/371472 [6:14:01<1:09:37,  3.51it/s]
 96%|█████████▌| 356789/371472 [6:14:02<1:10:02,  3.49it/s]
 96%|█████████▌| 356790/371472 [6:14:02<1:12:06,  3.39it/s]
 96%|█████████▌| 356791/371472 [6:14:02<1:09:23,  3.53it/s]
 96%|█████████▌| 356792/371472 [6:14:03<1:11:07,  3.44it/s]
 96%|█████████▌| 356793/371472 [6:14:03<1:15:45,  3.23it/s]
 96%|█████████▌| 356794/371472 [6:14:03<1:11:42,  3.41it/s]
 96%|█████████▌| 356795/371472 [6:14:04<1:16:21,  3.20it/s]
 96%|█████████▌| 356796/371472 [6:14:04<1:15:13,  3.25it/s]
 96%|█████████▌| 356797/371472 [6:14:04<1:11:56,  3.40it/s]
 96%|█████████▌| 356798/371472 [6:14:04<1:14:06,  3.30it/s]
 96%|█████████▌| 356799/371472 [6:14:05<1:10:13,  3.48it/s]
 96%|█████████▌| 356800/371472 [6:14:05<1:08:10,  3.59it/s]
                                                           
{'loss': 2.6274, 'learning_rate': 1.3556637721131675e-07, 'epoch': 15.37}
+
 96%|█████████▌| 356800/371472 [6:14:05<1:08:10,  3.59it/s]
 96%|█████████▌| 356801/371472 [6:14:05<1:09:01,  3.54it/s]
 96%|█████████▌| 356802/371472 [6:14:06<1:11:04,  3.44it/s]
 96%|█████████▌| 356803/371472 [6:14:06<1:07:38,  3.61it/s]
 96%|█████████▌| 356804/371472 [6:14:06<1:07:16,  3.63it/s]
 96%|█████████▌| 356805/371472 [6:14:06<1:08:33,  3.57it/s]
 96%|█████████▌| 356806/371472 [6:14:07<1:08:29,  3.57it/s]
 96%|█████████▌| 356807/371472 [6:14:07<1:07:39,  3.61it/s]
 96%|█████████▌| 356808/371472 [6:14:07<1:05:36,  3.72it/s]
 96%|█████████▌| 356809/371472 [6:14:07<1:04:59,  3.76it/s]
 96%|█████████▌| 356810/371472 [6:14:08<1:06:01,  3.70it/s]
 96%|█████████▌| 356811/371472 [6:14:08<1:05:52,  3.71it/s]
 96%|█████████▌| 356812/371472 [6:14:08<1:06:33,  3.67it/s]
 96%|█████████▌| 356813/371472 [6:14:08<1:06:22,  3.68it/s]
 96%|█████████▌| 356814/371472 [6:14:09<1:09:56,  3.49it/s]
 96%|█████████▌| 356815/371472 [6:14:09<1:09:02,  3.54it/s]
 96%|█████████▌| 356816/371472 [6:14:09<1:07:31,  3.62it/s]
 96%|█████████▌| 356817/371472 [6:14:10<1:11:43,  3.41it/s]
 96%|█████████▌| 356818/371472 [6:14:10<1:10:02,  3.49it/s]
 96%|█████████▌| 356819/371472 [6:14:10<1:09:08,  3.53it/s]
 96%|█████████▌| 356820/371472 [6:14:10<1:07:29,  3.62it/s]
                                                           
{'loss': 2.5292, 'learning_rate': 1.3551789523583782e-07, 'epoch': 15.37}
+
 96%|█████████▌| 356820/371472 [6:14:10<1:07:29,  3.62it/s]
 96%|█████████▌| 356821/371472 [6:14:11<1:12:47,  3.35it/s]
 96%|█████████▌| 356822/371472 [6:14:11<1:12:30,  3.37it/s]
 96%|█████████▌| 356823/371472 [6:14:11<1:11:06,  3.43it/s]
 96%|█████████▌| 356824/371472 [6:14:12<1:07:55,  3.59it/s]
 96%|█████████▌| 356825/371472 [6:14:12<1:05:35,  3.72it/s]
 96%|█████████▌| 356826/371472 [6:14:12<1:11:59,  3.39it/s]
 96%|█████████▌| 356827/371472 [6:14:13<1:11:57,  3.39it/s]
 96%|█████████▌| 356828/371472 [6:14:13<1:14:24,  3.28it/s]
 96%|█████████▌| 356829/371472 [6:14:13<1:17:21,  3.16it/s]
 96%|█████████▌| 356830/371472 [6:14:13<1:14:21,  3.28it/s]
 96%|█████████▌| 356831/371472 [6:14:14<1:14:34,  3.27it/s]
 96%|█████████▌| 356832/371472 [6:14:14<1:14:31,  3.27it/s]
 96%|█████████▌| 356833/371472 [6:14:14<1:11:08,  3.43it/s]
 96%|█████████▌| 356834/371472 [6:14:15<1:09:40,  3.50it/s]
 96%|█████████▌| 356835/371472 [6:14:15<1:16:06,  3.21it/s]
 96%|█████████▌| 356836/371472 [6:14:15<1:12:56,  3.34it/s]
 96%|█████████▌| 356837/371472 [6:14:16<1:09:18,  3.52it/s]
 96%|█████████▌| 356838/371472 [6:14:16<1:07:39,  3.60it/s]
 96%|█████████▌| 356839/371472 [6:14:16<1:11:00,  3.43it/s]
 96%|█████████▌| 356840/371472 [6:14:16<1:09:23,  3.51it/s]
                                                           
{'loss': 2.4636, 'learning_rate': 1.3546941326035897e-07, 'epoch': 15.37}
+
 96%|█████████▌| 356840/371472 [6:14:16<1:09:23,  3.51it/s]
 96%|█████████▌| 356841/371472 [6:14:17<1:07:37,  3.61it/s]
 96%|█████████▌| 356842/371472 [6:14:17<1:11:52,  3.39it/s]
 96%|█████████▌| 356843/371472 [6:14:17<1:09:35,  3.50it/s]
 96%|█████████▌| 356844/371472 [6:14:18<1:09:24,  3.51it/s]
 96%|█████████▌| 356845/371472 [6:14:18<1:10:01,  3.48it/s]
 96%|█████████▌| 356846/371472 [6:14:18<1:09:35,  3.50it/s]
 96%|█████████▌| 356847/371472 [6:14:18<1:14:02,  3.29it/s]
 96%|█████████▌| 356848/371472 [6:14:19<1:11:07,  3.43it/s]
 96%|█████████▌| 356849/371472 [6:14:19<1:16:03,  3.20it/s]
 96%|█████████▌| 356850/371472 [6:14:19<1:12:48,  3.35it/s]
 96%|█████████▌| 356851/371472 [6:14:20<1:11:11,  3.42it/s]
 96%|█████████▌| 356852/371472 [6:14:20<1:09:45,  3.49it/s]
 96%|█████████▌| 356853/371472 [6:14:20<1:09:46,  3.49it/s]
 96%|█████████▌| 356854/371472 [6:14:20<1:10:33,  3.45it/s]
 96%|█████████▌| 356855/371472 [6:14:21<1:14:56,  3.25it/s]
 96%|█████████▌| 356856/371472 [6:14:21<1:16:05,  3.20it/s]
 96%|█████████▌| 356857/371472 [6:14:21<1:14:56,  3.25it/s]
 96%|█████████▌| 356858/371472 [6:14:22<1:18:26,  3.10it/s]
 96%|█████████▌| 356859/371472 [6:14:22<1:18:05,  3.12it/s]
 96%|█████████▌| 356860/371472 [6:14:22<1:15:24,  3.23it/s]
                                                           
{'loss': 2.6202, 'learning_rate': 1.3542093128488004e-07, 'epoch': 15.37}
+
 96%|█████████▌| 356860/371472 [6:14:22<1:15:24,  3.23it/s]
 96%|█████████▌| 356861/371472 [6:14:23<1:17:08,  3.16it/s]
 96%|█████████▌| 356862/371472 [6:14:23<1:14:38,  3.26it/s]
 96%|█████████▌| 356863/371472 [6:14:23<1:11:44,  3.39it/s]
 96%|█████████▌| 356864/371472 [6:14:24<1:13:04,  3.33it/s]
 96%|█████████▌| 356865/371472 [6:14:24<1:12:51,  3.34it/s]
 96%|█████████▌| 356866/371472 [6:14:24<1:09:52,  3.48it/s]
 96%|█████████▌| 356867/371472 [6:14:24<1:06:53,  3.64it/s]
 96%|█████████▌| 356868/371472 [6:14:25<1:07:32,  3.60it/s]
 96%|█████████▌| 356869/371472 [6:14:25<1:08:30,  3.55it/s]
 96%|█████████▌| 356870/371472 [6:14:25<1:08:06,  3.57it/s]
 96%|█████████▌| 356871/371472 [6:14:25<1:05:44,  3.70it/s]
 96%|█████████▌| 356872/371472 [6:14:26<1:05:24,  3.72it/s]
 96%|█████████▌| 356873/371472 [6:14:26<1:03:44,  3.82it/s]
 96%|█████████▌| 356874/371472 [6:14:26<1:05:50,  3.70it/s]
 96%|█████████▌| 356875/371472 [6:14:27<1:08:49,  3.53it/s]
 96%|█████████▌| 356876/371472 [6:14:27<1:07:58,  3.58it/s]
 96%|█████████▌| 356877/371472 [6:14:27<1:05:43,  3.70it/s]
 96%|█████████▌| 356878/371472 [6:14:27<1:04:00,  3.80it/s]
 96%|█████████▌| 356879/371472 [6:14:28<1:04:48,  3.75it/s]
 96%|█████████▌| 356880/371472 [6:14:28<1:07:20,  3.61it/s]
                                                           
{'loss': 2.6561, 'learning_rate': 1.353724493094012e-07, 'epoch': 15.37}
+
 96%|█████████▌| 356880/371472 [6:14:28<1:07:20,  3.61it/s]
 96%|█████████▌| 356881/371472 [6:14:28<1:09:39,  3.49it/s]
 96%|█████████▌| 356882/371472 [6:14:29<1:10:58,  3.43it/s]
 96%|█████████▌| 356883/371472 [6:14:29<1:08:16,  3.56it/s]
 96%|█████████▌| 356884/371472 [6:14:29<1:07:47,  3.59it/s]
 96%|█████████▌| 356885/371472 [6:14:29<1:06:57,  3.63it/s]
 96%|█████████▌| 356886/371472 [6:14:30<1:04:49,  3.75it/s]
 96%|█████████▌| 356887/371472 [6:14:30<1:09:01,  3.52it/s]
 96%|█████████▌| 356888/371472 [6:14:30<1:06:27,  3.66it/s]
 96%|███��█████▌| 356889/371472 [6:14:31<1:10:49,  3.43it/s]
 96%|█████████▌| 356890/371472 [6:14:31<1:12:27,  3.35it/s]
 96%|█████████▌| 356891/371472 [6:14:31<1:08:18,  3.56it/s]
 96%|█████████▌| 356892/371472 [6:14:31<1:05:30,  3.71it/s]
 96%|█████████▌| 356893/371472 [6:14:32<1:05:04,  3.73it/s]
 96%|█████████▌| 356894/371472 [6:14:32<1:09:10,  3.51it/s]
 96%|█████████▌| 356895/371472 [6:14:32<1:08:34,  3.54it/s]
 96%|█████████▌| 356896/371472 [6:14:32<1:12:03,  3.37it/s]
 96%|█████████▌| 356897/371472 [6:14:33<1:09:15,  3.51it/s]
 96%|█████████▌| 356898/371472 [6:14:33<1:09:11,  3.51it/s]
 96%|█████████▌| 356899/371472 [6:14:33<1:10:13,  3.46it/s]
 96%|█████████▌| 356900/371472 [6:14:34<1:10:28,  3.45it/s]
                                                           
{'loss': 2.8457, 'learning_rate': 1.3532396733392227e-07, 'epoch': 15.37}
+
 96%|█████████▌| 356900/371472 [6:14:34<1:10:28,  3.45it/s]
 96%|█████████▌| 356901/371472 [6:14:34<1:18:53,  3.08it/s]
 96%|█████████▌| 356902/371472 [6:14:34<1:16:17,  3.18it/s]
 96%|█████████▌| 356903/371472 [6:14:35<1:12:32,  3.35it/s]
 96%|█████████▌| 356904/371472 [6:14:35<1:12:32,  3.35it/s]
 96%|█████████▌| 356905/371472 [6:14:35<1:14:34,  3.26it/s]
 96%|█████████▌| 356906/371472 [6:14:36<1:14:42,  3.25it/s]
 96%|█████████▌| 356907/371472 [6:14:36<1:12:07,  3.37it/s]
 96%|█████████▌| 356908/371472 [6:14:36<1:11:09,  3.41it/s]
 96%|█████████▌| 356909/371472 [6:14:36<1:13:51,  3.29it/s]
 96%|█████████▌| 356910/371472 [6:14:37<1:16:22,  3.18it/s]
 96%|█████████▌| 356911/371472 [6:14:37<1:15:42,  3.21it/s]
 96%|█████████▌| 356912/371472 [6:14:37<1:15:35,  3.21it/s]
 96%|█████████▌| 356913/371472 [6:14:38<1:12:55,  3.33it/s]
 96%|█████████▌| 356914/371472 [6:14:38<1:08:23,  3.55it/s]
 96%|█████████▌| 356915/371472 [6:14:38<1:12:58,  3.32it/s]
 96%|█████████▌| 356916/371472 [6:14:38<1:09:32,  3.49it/s]
 96%|█████████▌| 356917/371472 [6:14:39<1:07:47,  3.58it/s]
 96%|█████████▌| 356918/371472 [6:14:39<1:07:30,  3.59it/s]
 96%|█████████▌| 356919/371472 [6:14:39<1:08:36,  3.54it/s]
 96%|█████████▌| 356920/371472 [6:14:40<1:06:19,  3.66it/s]
                                                           
{'loss': 2.436, 'learning_rate': 1.3527548535844342e-07, 'epoch': 15.37}
+
 96%|█████████▌| 356920/371472 [6:14:40<1:06:19,  3.66it/s]
 96%|█████████▌| 356921/371472 [6:14:40<1:09:50,  3.47it/s]
 96%|█████████▌| 356922/371472 [6:14:40<1:08:24,  3.55it/s]
 96%|█████████▌| 356923/371472 [6:14:40<1:08:19,  3.55it/s]
 96%|█████████▌| 356924/371472 [6:14:41<1:06:58,  3.62it/s]
 96%|█████████▌| 356925/371472 [6:14:41<1:07:37,  3.58it/s]
 96%|█████████▌| 356926/371472 [6:14:41<1:06:40,  3.64it/s]
 96%|█████████▌| 356927/371472 [6:14:42<1:12:36,  3.34it/s]
 96%|█████████▌| 356928/371472 [6:14:42<1:09:49,  3.47it/s]
 96%|█████████▌| 356929/371472 [6:14:42<1:06:23,  3.65it/s]
 96%|█████████▌| 356930/371472 [6:14:42<1:07:06,  3.61it/s]
 96%|█████████▌| 356931/371472 [6:14:43<1:13:15,  3.31it/s]
 96%|█████████▌| 356932/371472 [6:14:43<1:12:30,  3.34it/s]
 96%|█████████▌| 356933/371472 [6:14:43<1:12:39,  3.34it/s]
 96%|█████████▌| 356934/371472 [6:14:44<1:09:15,  3.50it/s]
 96%|█████████▌| 356935/371472 [6:14:44<1:06:35,  3.64it/s]
 96%|█████████▌| 356936/371472 [6:14:44<1:07:22,  3.60it/s]
 96%|█████████▌| 356937/371472 [6:14:44<1:10:17,  3.45it/s]
 96%|█████████▌| 356938/371472 [6:14:45<1:15:02,  3.23it/s]
 96%|█████████▌| 356939/371472 [6:14:45<1:12:27,  3.34it/s]
 96%|█████████▌| 356940/371472 [6:14:45<1:10:03,  3.46it/s]
                                                           
{'loss': 2.4739, 'learning_rate': 1.3522700338296446e-07, 'epoch': 15.37}
+
 96%|█████████▌| 356940/371472 [6:14:45<1:10:03,  3.46it/s]
 96%|█████████▌| 356941/371472 [6:14:46<1:08:14,  3.55it/s]
 96%|█████████▌| 356942/371472 [6:14:46<1:09:07,  3.50it/s]
 96%|█████████▌| 356943/371472 [6:14:46<1:12:26,  3.34it/s]
 96%|█████████▌| 356944/371472 [6:14:46<1:08:42,  3.52it/s]
 96%|█���███████▌| 356945/371472 [6:14:47<1:14:51,  3.23it/s]
 96%|█████████▌| 356946/371472 [6:14:47<1:14:16,  3.26it/s]
 96%|█████████▌| 356947/371472 [6:14:47<1:12:45,  3.33it/s]
 96%|█████████▌| 356948/371472 [6:14:48<1:09:39,  3.48it/s]
 96%|█████████▌| 356949/371472 [6:14:48<1:13:49,  3.28it/s]
 96%|█████████▌| 356950/371472 [6:14:48<1:10:35,  3.43it/s]
 96%|█████████▌| 356951/371472 [6:14:49<1:09:56,  3.46it/s]
 96%|█████████▌| 356952/371472 [6:14:49<1:10:39,  3.42it/s]
 96%|█████████▌| 356953/371472 [6:14:49<1:06:29,  3.64it/s]
 96%|█████████▌| 356954/371472 [6:14:49<1:07:07,  3.60it/s]
 96%|█████████▌| 356955/371472 [6:14:50<1:08:42,  3.52it/s]
 96%|█████████▌| 356956/371472 [6:14:50<1:09:20,  3.49it/s]
 96%|█████████▌| 356957/371472 [6:14:50<1:07:31,  3.58it/s]
 96%|█████████▌| 356958/371472 [6:14:51<1:06:03,  3.66it/s]
 96%|█████████▌| 356959/371472 [6:14:51<1:08:46,  3.52it/s]
 96%|█████████▌| 356960/371472 [6:14:51<1:14:13,  3.26it/s]
                                                           
{'loss': 2.5, 'learning_rate': 1.3517852140748564e-07, 'epoch': 15.37}
+
 96%|█████████▌| 356960/371472 [6:14:51<1:14:13,  3.26it/s]
 96%|█████████▌| 356961/371472 [6:14:51<1:13:59,  3.27it/s]
 96%|█████████▌| 356962/371472 [6:14:52<1:12:23,  3.34it/s]
 96%|█████████▌| 356963/371472 [6:14:52<1:11:08,  3.40it/s]
 96%|█████████▌| 356964/371472 [6:14:52<1:08:34,  3.53it/s]
 96%|█████████▌| 356965/371472 [6:14:53<1:09:43,  3.47it/s]
 96%|█████████▌| 356966/371472 [6:14:53<1:12:16,  3.34it/s]
 96%|█████████▌| 356967/371472 [6:14:53<1:11:39,  3.37it/s]
 96%|█████████▌| 356968/371472 [6:14:53<1:07:19,  3.59it/s]
 96%|█████████▌| 356969/371472 [6:14:54<1:05:30,  3.69it/s]
 96%|█████████▌| 356970/371472 [6:14:54<1:05:57,  3.66it/s]
 96%|█████████▌| 356971/371472 [6:14:54<1:05:47,  3.67it/s]
 96%|█████████▌| 356972/371472 [6:14:55<1:04:35,  3.74it/s]
 96%|█████████▌| 356973/371472 [6:14:55<1:04:12,  3.76it/s]
 96%|█████████▌| 356974/371472 [6:14:55<1:04:43,  3.73it/s]
 96%|█████████▌| 356975/371472 [6:14:55<1:03:00,  3.83it/s]
 96%|█████████▌| 356976/371472 [6:14:56<1:04:29,  3.75it/s]
 96%|█████████▌| 356977/371472 [6:14:56<1:04:01,  3.77it/s]
 96%|█████████▌| 356978/371472 [6:14:56<1:04:44,  3.73it/s]
 96%|█████████▌| 356979/371472 [6:14:56<1:07:16,  3.59it/s]
 96%|█████████▌| 356980/371472 [6:14:57<1:05:37,  3.68it/s]
                                                           
{'loss': 2.6973, 'learning_rate': 1.3513003943200668e-07, 'epoch': 15.38}
+
 96%|█████████▌| 356980/371472 [6:14:57<1:05:37,  3.68it/s]
 96%|█████████▌| 356981/371472 [6:14:57<1:06:11,  3.65it/s]
 96%|█████████▌| 356982/371472 [6:14:57<1:10:23,  3.43it/s]
 96%|█████████▌| 356983/371472 [6:14:58<1:08:07,  3.54it/s]
 96%|█████████▌| 356984/371472 [6:14:58<1:06:18,  3.64it/s]
 96%|█████████▌| 356985/371472 [6:14:58<1:08:03,  3.55it/s]
 96%|█████████▌| 356986/371472 [6:14:58<1:05:43,  3.67it/s]
 96%|█████████▌| 356987/371472 [6:14:59<1:05:34,  3.68it/s]
 96%|█████████▌| 356988/371472 [6:14:59<1:06:32,  3.63it/s]
 96%|█████████▌| 356989/371472 [6:14:59<1:06:03,  3.65it/s]
 96%|█████████▌| 356990/371472 [6:14:59<1:06:02,  3.65it/s]
 96%|█████████▌| 356991/371472 [6:15:00<1:04:24,  3.75it/s]
 96%|█████████▌| 356992/371472 [6:15:00<1:06:47,  3.61it/s]
 96%|█████████▌| 356993/371472 [6:15:00<1:12:59,  3.31it/s]
 96%|█████████▌| 356994/371472 [6:15:01<1:09:23,  3.48it/s]
 96%|█████████▌| 356995/371472 [6:15:01<1:08:12,  3.54it/s]
 96%|█████████▌| 356996/371472 [6:15:01<1:05:24,  3.69it/s]
 96%|█████████▌| 356997/371472 [6:15:01<1:06:02,  3.65it/s]
 96%|█████████▌| 356998/371472 [6:15:02<1:06:04,  3.65it/s]
 96%|█████████▌| 356999/371472 [6:15:02<1:09:30,  3.47it/s]
 96%|█████████▌| 357000/371472 [6:15:02<1:08:20,  3.53it/s]
                                                           
{'loss': 2.7393, 'learning_rate': 1.3508155745652783e-07, 'epoch': 15.38}
+
 96%|█████████▌| 357000/371472 [6:15:02<1:08:20,  3.53it/s]
 96%|█████████▌| 357001/371472 [6:15:03<1:08:44,  3.51it/s]
 96%|█████████▌| 357002/371472 [6:15:03<1:07:37,  3.57it/s]
 96%|█████████▌| 357003/371472 [6:15:03<1:09:21,  3.48it/s]
 96%|█████████▌| 357004/371472 [6:15:03<1:08:55,  3.50it/s]
 96%|█████████▌| 357005/371472 [6:15:04<1:14:07,  3.25it/s]
 96%|█████████▌| 357006/371472 [6:15:04<1:10:51,  3.40it/s]
 96%|█████████▌| 357007/371472 [6:15:04<1:07:27,  3.57it/s]
 96%|█████████▌| 357008/371472 [6:15:05<1:09:14,  3.48it/s]
 96%|█████████▌| 357009/371472 [6:15:05<1:06:12,  3.64it/s]
 96%|█████████▌| 357010/371472 [6:15:05<1:05:54,  3.66it/s]
 96%|█████████▌| 357011/371472 [6:15:05<1:07:57,  3.55it/s]
 96%|█████████▌| 357012/371472 [6:15:06<1:07:04,  3.59it/s]
 96%|█████████▌| 357013/371472 [6:15:06<1:07:55,  3.55it/s]
 96%|█████████▌| 357014/371472 [6:15:06<1:12:01,  3.35it/s]
 96%|█████████▌| 357015/371472 [6:15:07<1:10:21,  3.42it/s]
 96%|█████████▌| 357016/371472 [6:15:07<1:10:19,  3.43it/s]
 96%|█████████▌| 357017/371472 [6:15:07<1:09:02,  3.49it/s]
 96%|█████████▌| 357018/371472 [6:15:07<1:07:31,  3.57it/s]
 96%|█████████▌| 357019/371472 [6:15:08<1:08:24,  3.52it/s]
 96%|█████████▌| 357020/371472 [6:15:08<1:08:05,  3.54it/s]
                                                           
{'loss': 2.5465, 'learning_rate': 1.350330754810489e-07, 'epoch': 15.38}
+
 96%|█████████▌| 357020/371472 [6:15:08<1:08:05,  3.54it/s]
 96%|█████████▌| 357021/371472 [6:15:08<1:07:24,  3.57it/s]
 96%|█████████▌| 357022/371472 [6:15:09<1:11:57,  3.35it/s]
 96%|█████████▌| 357023/371472 [6:15:09<1:08:59,  3.49it/s]
 96%|█████████▌| 357024/371472 [6:15:09<1:07:00,  3.59it/s]
 96%|█████████▌| 357025/371472 [6:15:09<1:03:50,  3.77it/s]
 96%|█████████▌| 357026/371472 [6:15:10<1:08:55,  3.49it/s]
 96%|█████████▌| 357027/371472 [6:15:10<1:07:37,  3.56it/s]
 96%|█████████▌| 357028/371472 [6:15:10<1:09:26,  3.47it/s]
 96%|█████████▌| 357029/371472 [6:15:11<1:07:38,  3.56it/s]
 96%|█████████▌| 357030/371472 [6:15:11<1:06:41,  3.61it/s]
 96%|█████████▌| 357031/371472 [6:15:11<1:04:50,  3.71it/s]
 96%|█████████▌| 357032/371472 [6:15:11<1:05:34,  3.67it/s]
 96%|█████████▌| 357033/371472 [6:15:12<1:37:27,  2.47it/s]
 96%|█████████▌| 357034/371472 [6:15:12<1:30:59,  2.64it/s]
 96%|█████████▌| 357035/371472 [6:15:13<1:23:57,  2.87it/s]
 96%|█████████▌| 357036/371472 [6:15:13<1:17:36,  3.10it/s]
 96%|█████████▌| 357037/371472 [6:15:13<1:13:43,  3.26it/s]
 96%|█████████▌| 357038/371472 [6:15:14<1:18:18,  3.07it/s]
 96%|█████████▌| 357039/371472 [6:15:14<1:14:55,  3.21it/s]
 96%|█████████▌| 357040/371472 [6:15:14<1:12:39,  3.31it/s]
                                                           
{'loss': 2.5614, 'learning_rate': 1.3498459350557006e-07, 'epoch': 15.38}
+
 96%|█████████▌| 357040/371472 [6:15:14<1:12:39,  3.31it/s]
 96%|█████████▌| 357041/371472 [6:15:14<1:19:08,  3.04it/s]
 96%|█████████▌| 357042/371472 [6:15:15<1:27:16,  2.76it/s]
 96%|█████████▌| 357043/371472 [6:15:15<1:20:13,  3.00it/s]
 96%|█████████▌| 357044/371472 [6:15:15<1:13:23,  3.28it/s]
 96%|█████████▌| 357045/371472 [6:15:16<1:09:53,  3.44it/s]
 96%|█████████▌| 357046/371472 [6:15:16<1:12:25,  3.32it/s]
 96%|█████████▌| 357047/371472 [6:15:16<1:11:52,  3.34it/s]
 96%|█████████▌| 357048/371472 [6:15:17<1:10:22,  3.42it/s]
 96%|█████████▌| 357049/371472 [6:15:17<1:08:24,  3.51it/s]
 96%|█████████▌| 357050/371472 [6:15:17<1:06:51,  3.59it/s]
 96%|█████████▌| 357051/371472 [6:15:17<1:05:17,  3.68it/s]
 96%|█████████▌| 357052/371472 [6:15:18<1:04:13,  3.74it/s]
 96%|█████████▌| 357053/371472 [6:15:18<1:08:47,  3.49it/s]
 96%|█████████▌| 357054/371472 [6:15:18<1:07:45,  3.55it/s]
 96%|█████████▌| 357055/371472 [6:15:19<1:06:25,  3.62it/s]
 96%|█████████▌| 357056/371472 [6:15:19<1:08:54,  3.49it/s]
 96%|█████████▌| 357057/371472 [6:15:19<1:11:17,  3.37it/s]
 96%|█████████▌| 357058/371472 [6:15:19<1:07:35,  3.55it/s]
 96%|█████████▌| 357059/371472 [6:15:20<1:05:11,  3.69it/s]
 96%|█████████▌| 357060/371472 [6:15:20<1:08:36,  3.50it/s]
                                                           
{'loss': 2.6187, 'learning_rate': 1.349361115300911e-07, 'epoch': 15.38}
+
 96%|█████████▌| 357060/371472 [6:15:20<1:08:36,  3.50it/s]
 96%|█████████▌| 357061/371472 [6:15:20<1:08:47,  3.49it/s]
 96%|█████████▌| 357062/371472 [6:15:21<1:07:34,  3.55it/s]
 96%|█████████▌| 357063/371472 [6:15:21<1:06:30,  3.61it/s]
 96%|█████████▌| 357064/371472 [6:15:21<1:08:16,  3.52it/s]
 96%|█████████▌| 357065/371472 [6:15:21<1:08:40,  3.50it/s]
 96%|█████████▌| 357066/371472 [6:15:22<1:08:07,  3.52it/s]
 96%|█████████▌| 357067/371472 [6:15:22<1:11:11,  3.37it/s]
 96%|█████████▌| 357068/371472 [6:15:22<1:09:47,  3.44it/s]
 96%|█████████▌| 357069/371472 [6:15:23<1:08:15,  3.52it/s]
 96%|█████████▌| 357070/371472 [6:15:23<1:07:45,  3.54it/s]
 96%|█████████▌| 357071/371472 [6:15:23<1:08:34,  3.50it/s]
 96%|█████████▌| 357072/371472 [6:15:23<1:08:52,  3.48it/s]
 96%|█████████▌| 357073/371472 [6:15:24<1:10:25,  3.41it/s]
 96%|█████████▌| 357074/371472 [6:15:24<1:08:21,  3.51it/s]
 96%|█████████▌| 357075/371472 [6:15:24<1:07:57,  3.53it/s]
 96%|█████████▌| 357076/371472 [6:15:25<1:07:15,  3.57it/s]
 96%|█████████▌| 357077/371472 [6:15:25<1:05:29,  3.66it/s]
 96%|█████████▌| 357078/371472 [6:15:25<1:09:37,  3.45it/s]
 96%|█████████▌| 357079/371472 [6:15:25<1:09:36,  3.45it/s]
 96%|█████████▌| 357080/371472 [6:15:26<1:07:53,  3.53it/s]
                                                           
{'loss': 2.5014, 'learning_rate': 1.3488762955461228e-07, 'epoch': 15.38}
+
 96%|█████████▌| 357080/371472 [6:15:26<1:07:53,  3.53it/s]
 96%|█████████▌| 357081/371472 [6:15:26<1:07:40,  3.54it/s]
 96%|█████████▌| 357082/371472 [6:15:26<1:10:13,  3.41it/s]
 96%|█████████▌| 357083/371472 [6:15:27<1:18:17,  3.06it/s]
 96%|█████████▌| 357084/371472 [6:15:27<1:21:14,  2.95it/s]
 96%|█████████▌| 357085/371472 [6:15:27<1:17:00,  3.11it/s]
 96%|█████████▌| 357086/371472 [6:15:28<1:13:35,  3.26it/s]
 96%|█████████▌| 357087/371472 [6:15:28<1:11:28,  3.35it/s]
 96%|█████████▌| 357088/371472 [6:15:28<1:10:00,  3.42it/s]
 96%|█████████▌| 357089/371472 [6:15:28<1:07:06,  3.57it/s]
 96%|█████████▌| 357090/371472 [6:15:29<1:06:21,  3.61it/s]
 96%|█████████▌| 357091/371472 [6:15:29<1:04:32,  3.71it/s]
 96%|█████████▌| 357092/371472 [6:15:29<1:05:15,  3.67it/s]
 96%|█████████▌| 357093/371472 [6:15:29<1:05:08,  3.68it/s]
 96%|█████████▌| 357094/371472 [6:15:30<1:07:58,  3.53it/s]
 96%|█████████▌| 357095/371472 [6:15:30<1:06:07,  3.62it/s]
 96%|█████████▌| 357096/371472 [6:15:30<1:07:48,  3.53it/s]
 96%|█████████▌| 357097/371472 [6:15:31<1:10:49,  3.38it/s]
 96%|█████████▌| 357098/371472 [6:15:31<1:07:19,  3.56it/s]
 96%|█████████▌| 357099/371472 [6:15:31<1:08:52,  3.48it/s]
 96%|█████████▌| 357100/371472 [6:15:31<1:08:10,  3.51it/s]
                                                           
{'loss': 2.613, 'learning_rate': 1.348391475791333e-07, 'epoch': 15.38}
+
 96%|█████████▌| 357100/371472 [6:15:31<1:08:10,  3.51it/s]
 96%|█████████▌| 357101/371472 [6:15:32<1:06:15,  3.61it/s]
 96%|█████████▌| 357102/371472 [6:15:32<1:08:43,  3.48it/s]
 96%|█████████▌| 357103/371472 [6:15:32<1:07:05,  3.57it/s]
 96%|█████████▌| 357104/371472 [6:15:33<1:05:12,  3.67it/s]
 96%|█████████▌| 357105/371472 [6:15:33<1:05:12,  3.67it/s]
 96%|█████████▌| 357106/371472 [6:15:33<1:06:56,  3.58it/s]
 96%|█████████▌| 357107/371472 [6:15:33<1:07:42,  3.54it/s]
 96%|█████████▌| 357108/371472 [6:15:34<1:05:30,  3.65it/s]
 96%|█████████▌| 357109/371472 [6:15:34<1:06:01,  3.63it/s]
 96%|█████████▌| 357110/371472 [6:15:34<1:04:19,  3.72it/s]
 96%|█████████▌| 357111/371472 [6:15:34<1:02:39,  3.82it/s]
 96%|█████████▌| 357112/371472 [6:15:35<1:02:58,  3.80it/s]
 96%|█████████▌| 357113/371472 [6:15:35<1:06:04,  3.62it/s]
 96%|█████████▌| 357114/371472 [6:15:35<1:04:40,  3.70it/s]
 96%|█████████▌| 357115/371472 [6:15:36<1:09:13,  3.46it/s]
 96%|█████████▌| 357116/371472 [6:15:36<1:08:51,  3.47it/s]
 96%|█████████▌| 357117/371472 [6:15:36<1:09:22,  3.45it/s]
 96%|█████████▌| 357118/371472 [6:15:37<1:18:42,  3.04it/s]
 96%|█████████▌| 357119/371472 [6:15:37<1:13:41,  3.25it/s]
 96%|█████████▌| 357120/371472 [6:15:37<1:12:23,  3.30it/s]
                                                           
{'loss': 2.6171, 'learning_rate': 1.3479066560365448e-07, 'epoch': 15.38}
+
 96%|█████████▌| 357120/371472 [6:15:37<1:12:23,  3.30it/s]
 96%|█████████▌| 357121/371472 [6:15:37<1:09:41,  3.43it/s]
 96%|█████████▌| 357122/371472 [6:15:38<1:09:35,  3.44it/s]
 96%|█████████▌| 357123/371472 [6:15:38<1:07:09,  3.56it/s]
 96%|█████████▌| 357124/371472 [6:15:38<1:07:46,  3.53it/s]
 96%|█████████▌| 357125/371472 [6:15:39<1:07:48,  3.53it/s]
 96%|█████████▌| 357126/371472 [6:15:39<1:06:59,  3.57it/s]
 96%|█████████▌| 357127/371472 [6:15:39<1:08:01,  3.51it/s]
 96%|█████████▌| 357128/371472 [6:15:39<1:12:03,  3.32it/s]
 96%|█████████▌| 357129/371472 [6:15:40<1:08:41,  3.48it/s]
 96%|█████████▌| 357130/371472 [6:15:40<1:05:32,  3.65it/s]
 96%|█████████▌| 357131/371472 [6:15:40<1:03:53,  3.74it/s]
 96%|█████████▌| 357132/371472 [6:15:40<1:05:01,  3.68it/s]
 96%|█████████▌| 357133/371472 [6:15:41<1:07:05,  3.56it/s]
 96%|█████████▌| 357134/371472 [6:15:41<1:06:57,  3.57it/s]
 96%|█████████▌| 357135/371472 [6:15:41<1:08:25,  3.49it/s]
 96%|█████████▌| 357136/371472 [6:15:42<1:07:52,  3.52it/s]
 96%|█████████▌| 357137/371472 [6:15:42<1:05:52,  3.63it/s]
 96%|█████████▌| 357138/371472 [6:15:42<1:13:23,  3.26it/s]
 96%|█████████▌| 357139/371472 [6:15:43<1:09:19,  3.45it/s]
 96%|█████████▌| 357140/371472 [6:15:43<1:12:38,  3.29it/s]
                                                           
{'loss': 2.5206, 'learning_rate': 1.3474218362817555e-07, 'epoch': 15.38}
+
 96%|█████████▌| 357140/371472 [6:15:43<1:12:38,  3.29it/s]
 96%|█████████▌| 357141/371472 [6:15:43<1:13:45,  3.24it/s]
 96%|█████████▌| 357142/371472 [6:15:43<1:10:30,  3.39it/s]
 96%|█████████▌| 357143/371472 [6:15:44<1:16:45,  3.11it/s]
 96%|█████████▌| 357144/371472 [6:15:44<1:14:21,  3.21it/s]
 96%|█████████▌| 357145/371472 [6:15:44<1:16:12,  3.13it/s]
 96%|█████████▌| 357146/371472 [6:15:45<1:15:06,  3.18it/s]
 96%|█████████▌| 357147/371472 [6:15:45<1:12:45,  3.28it/s]
 96%|█████████▌| 357148/371472 [6:15:45<1:13:03,  3.27it/s]
 96%|█████████▌| 357149/371472 [6:15:46<1:14:32,  3.20it/s]
 96%|█████████▌| 357150/371472 [6:15:46<1:12:22,  3.30it/s]
 96%|█████████▌| 357151/371472 [6:15:46<1:18:35,  3.04it/s]
 96%|█████████▌| 357152/371472 [6:15:47<1:12:16,  3.30it/s]
 96%|█████████▌| 357153/371472 [6:15:47<1:08:40,  3.47it/s]
 96%|█████████▌| 357154/371472 [6:15:47<1:07:22,  3.54it/s]
 96%|█████████▌| 357155/371472 [6:15:47<1:06:39,  3.58it/s]
 96%|█████████▌| 357156/371472 [6:15:48<1:05:57,  3.62it/s]
 96%|█████████▌| 357157/371472 [6:15:48<1:04:03,  3.72it/s]
 96%|█████████▌| 357158/371472 [6:15:48<1:04:37,  3.69it/s]
 96%|█████████▌| 357159/371472 [6:15:48<1:03:07,  3.78it/s]
 96%|█████████▌| 357160/371472 [6:15:49<1:05:14,  3.66it/s]
                                                           
{'loss': 2.4861, 'learning_rate': 1.346937016526967e-07, 'epoch': 15.38}
+
 96%|█████████▌| 357160/371472 [6:15:49<1:05:14,  3.66it/s]
 96%|█████████▌| 357161/371472 [6:15:49<1:05:36,  3.64it/s]
 96%|█████████▌| 357162/371472 [6:15:49<1:04:57,  3.67it/s]
 96%|█████████▌| 357163/371472 [6:15:50<1:03:17,  3.77it/s]
 96%|█████████▌| 357164/371472 [6:15:50<1:03:23,  3.76it/s]
 96%|█████████▌| 357165/371472 [6:15:50<1:05:44,  3.63it/s]
 96%|█████████▌| 357166/371472 [6:15:50<1:11:35,  3.33it/s]
 96%|█████████▌| 357167/371472 [6:15:51<1:09:22,  3.44it/s]
 96%|█████████▌| 357168/371472 [6:15:51<1:07:29,  3.53it/s]
 96%|█████████▌| 357169/371472 [6:15:51<1:08:03,  3.50it/s]
 96%|█████████▌| 357170/371472 [6:15:52<1:05:36,  3.63it/s]
 96%|█████████▌| 357171/371472 [6:15:52<1:09:18,  3.44it/s]
 96%|█████████▌| 357172/371472 [6:15:52<1:09:32,  3.43it/s]
 96%|█████████▌| 357173/371472 [6:15:52<1:09:51,  3.41it/s]
 96%|█████████▌| 357174/371472 [6:15:53<1:13:46,  3.23it/s]
 96%|█████████▌| 357175/371472 [6:15:53<1:16:19,  3.12it/s]
 96%|█████████▌| 357176/371472 [6:15:53<1:11:15,  3.34it/s]
 96%|█████████▌| 357177/371472 [6:15:54<1:12:26,  3.29it/s]
 96%|█████████▌| 357178/371472 [6:15:54<1:10:04,  3.40it/s]
 96%|█████████▌| 357179/371472 [6:15:54<1:08:02,  3.50it/s]
 96%|█████████▌| 357180/371472 [6:15:55<1:09:09,  3.44it/s]
                                                           
{'loss': 2.4912, 'learning_rate': 1.3464521967721774e-07, 'epoch': 15.38}
+
 96%|█████████▌| 357180/371472 [6:15:55<1:09:09,  3.44it/s]
 96%|█████████▌| 357181/371472 [6:15:55<1:08:33,  3.47it/s]
 96%|█████████▌| 357182/371472 [6:15:55<1:10:29,  3.38it/s]
 96%|█████████▌| 357183/371472 [6:15:55<1:08:21,  3.48it/s]
 96%|█████████▌| 357184/371472 [6:15:56<1:05:54,  3.61it/s]
 96%|█████████▌| 357185/371472 [6:15:56<1:04:34,  3.69it/s]
 96%|█████████▌| 357186/371472 [6:15:56<1:05:37,  3.63it/s]
 96%|█████████▌| 357187/371472 [6:15:56<1:05:20,  3.64it/s]
 96%|█████████▌| 357188/371472 [6:15:57<1:05:52,  3.61it/s]
 96%|█████████▌| 357189/371472 [6:15:57<1:06:18,  3.59it/s]
 96%|█████████▌| 357190/371472 [6:15:57<1:10:11,  3.39it/s]
 96%|█████████▌| 357191/371472 [6:15:58<1:06:41,  3.57it/s]
 96%|█████████▌| 357192/371472 [6:15:58<1:04:41,  3.68it/s]
 96%|█████████▌| 357193/371472 [6:15:58<1:08:36,  3.47it/s]
 96%|█████████▌| 357194/371472 [6:15:58<1:05:11,  3.65it/s]
 96%|█████████▌| 357195/371472 [6:15:59<1:06:34,  3.57it/s]
 96%|█████████▌| 357196/371472 [6:15:59<1:05:31,  3.63it/s]
 96%|█████████▌| 357197/371472 [6:15:59<1:05:09,  3.65it/s]
 96%|█████████▌| 357198/371472 [6:16:00<1:04:06,  3.71it/s]
 96%|█████████▌| 357199/371472 [6:16:00<1:03:02,  3.77it/s]
 96%|█████████▌| 357200/371472 [6:16:00<1:03:03,  3.77it/s]
                                                           
{'loss': 2.4914, 'learning_rate': 1.3459673770173892e-07, 'epoch': 15.39}
+
 96%|█████████▌| 357200/371472 [6:16:00<1:03:03,  3.77it/s]
 96%|█████████▌| 357201/371472 [6:16:00<1:02:15,  3.82it/s]
 96%|█████████▌| 357202/371472 [6:16:01<1:09:45,  3.41it/s]
 96%|█████████▌| 357203/371472 [6:16:01<1:08:08,  3.49it/s]
 96%|█████████▌| 357204/371472 [6:16:01<1:12:24,  3.28it/s]
 96%|█████████▌| 357205/371472 [6:16:02<1:11:05,  3.34it/s]
 96%|█████████▌| 357206/371472 [6:16:02<1:09:35,  3.42it/s]
 96%|█████████▌| 357207/371472 [6:16:02<1:07:53,  3.50it/s]
 96%|█████████▌| 357208/371472 [6:16:02<1:07:28,  3.52it/s]
 96%|█████████▌| 357209/371472 [6:16:03<1:11:06,  3.34it/s]
 96%|█████████▌| 357210/371472 [6:16:03<1:09:59,  3.40it/s]
 96%|█████████▌| 357211/371472 [6:16:03<1:09:48,  3.40it/s]
 96%|█████████▌| 357212/371472 [6:16:04<1:10:39,  3.36it/s]
 96%|█████████▌| 357213/371472 [6:16:04<1:06:55,  3.55it/s]
 96%|█████████▌| 357214/371472 [6:16:04<1:05:08,  3.65it/s]
 96%|█████████▌| 357215/371472 [6:16:04<1:04:45,  3.67it/s]
 96%|█████████▌| 357216/371472 [6:16:05<1:03:15,  3.76it/s]
 96%|█████████▌| 357217/371472 [6:16:05<1:05:03,  3.65it/s]
 96%|█████████▌| 357218/371472 [6:16:05<1:04:34,  3.68it/s]
 96%|█████████▌| 357219/371472 [6:16:05<1:07:15,  3.53it/s]
 96%|█████████▌| 357220/371472 [6:16:06<1:05:43,  3.61it/s]
                                                           
{'loss': 2.4217, 'learning_rate': 1.3454825572625997e-07, 'epoch': 15.39}
+
 96%|█████████▌| 357220/371472 [6:16:06<1:05:43,  3.61it/s]
 96%|█████████▌| 357221/371472 [6:16:06<1:06:28,  3.57it/s]
 96%|█████████▌| 357222/371472 [6:16:06<1:12:18,  3.28it/s]
 96%|█████████▌| 357223/371472 [6:16:07<1:09:32,  3.42it/s]
 96%|█████████▌| 357224/371472 [6:16:07<1:09:38,  3.41it/s]
 96%|█████████▌| 357225/371472 [6:16:07<1:08:55,  3.45it/s]
 96%|█████████▌| 357226/371472 [6:16:08<1:09:54,  3.40it/s]
 96%|█████████���| 357227/371472 [6:16:08<1:07:18,  3.53it/s]
 96%|█████████▌| 357228/371472 [6:16:08<1:05:47,  3.61it/s]
 96%|█████████▌| 357229/371472 [6:16:08<1:05:37,  3.62it/s]
 96%|█████████▌| 357230/371472 [6:16:09<1:21:44,  2.90it/s]
 96%|█████████▌| 357231/371472 [6:16:09<1:16:04,  3.12it/s]
 96%|█████████▌| 357232/371472 [6:16:09<1:13:03,  3.25it/s]
 96%|█████████▌| 357233/371472 [6:16:10<1:12:31,  3.27it/s]
 96%|█████████▌| 357234/371472 [6:16:10<1:10:50,  3.35it/s]
 96%|█████████▌| 357235/371472 [6:16:10<1:06:39,  3.56it/s]
 96%|█████████▌| 357236/371472 [6:16:10<1:03:20,  3.75it/s]
 96%|█████████▌| 357237/371472 [6:16:11<1:05:58,  3.60it/s]
 96%|█████████▌| 357238/371472 [6:16:11<1:03:01,  3.76it/s]
 96%|█████████▌| 357239/371472 [6:16:11<1:09:17,  3.42it/s]
 96%|█████████▌| 357240/371472 [6:16:12<1:06:48,  3.55it/s]
                                                           
{'loss': 2.5138, 'learning_rate': 1.3449977375078112e-07, 'epoch': 15.39}
+
 96%|█████████▌| 357240/371472 [6:16:12<1:06:48,  3.55it/s]
 96%|█████████▌| 357241/371472 [6:16:12<1:04:47,  3.66it/s]
 96%|█████████▌| 357242/371472 [6:16:12<1:04:30,  3.68it/s]
 96%|█████████▌| 357243/371472 [6:16:12<1:05:05,  3.64it/s]
 96%|█████████▌| 357244/371472 [6:16:13<1:07:02,  3.54it/s]
 96%|█████████▌| 357245/371472 [6:16:13<1:04:57,  3.65it/s]
 96%|█████████▌| 357246/371472 [6:16:13<1:06:48,  3.55it/s]
 96%|█████████▌| 357247/371472 [6:16:14<1:07:37,  3.51it/s]
 96%|█████████▌| 357248/371472 [6:16:14<1:05:38,  3.61it/s]
 96%|█████████▌| 357249/371472 [6:16:14<1:05:28,  3.62it/s]
 96%|█████████▌| 357250/371472 [6:16:14<1:04:48,  3.66it/s]
 96%|█████████▌| 357251/371472 [6:16:15<1:07:44,  3.50it/s]
 96%|█████████▌| 357252/371472 [6:16:15<1:07:51,  3.49it/s]
 96%|█████████▌| 357253/371472 [6:16:15<1:06:03,  3.59it/s]
 96%|█████████▌| 357254/371472 [6:16:16<1:09:17,  3.42it/s]
 96%|█████████▌| 357255/371472 [6:16:16<1:08:43,  3.45it/s]
 96%|█████████▌| 357256/371472 [6:16:16<1:10:48,  3.35it/s]
 96%|█████████▌| 357257/371472 [6:16:16<1:13:24,  3.23it/s]
 96%|█████████▌| 357258/371472 [6:16:17<1:11:28,  3.31it/s]
 96%|█████████▌| 357259/371472 [6:16:17<1:08:19,  3.47it/s]
 96%|█████████▌| 357260/371472 [6:16:17<1:09:00,  3.43it/s]
                                                           
{'loss': 2.6025, 'learning_rate': 1.344512917753022e-07, 'epoch': 15.39}
+
 96%|█████████▌| 357260/371472 [6:16:17<1:09:00,  3.43it/s]
 96%|█████████▌| 357261/371472 [6:16:18<1:07:05,  3.53it/s]
 96%|█████████▌| 357262/371472 [6:16:18<1:11:55,  3.29it/s]
 96%|█████████▌| 357263/371472 [6:16:18<1:09:16,  3.42it/s]
 96%|█████████▌| 357264/371472 [6:16:18<1:07:15,  3.52it/s]
 96%|█████████▌| 357265/371472 [6:16:19<1:06:38,  3.55it/s]
 96%|█████████▌| 357266/371472 [6:16:19<1:08:29,  3.46it/s]
 96%|█████████▌| 357267/371472 [6:16:19<1:08:19,  3.47it/s]
 96%|█████████▌| 357268/371472 [6:16:20<1:06:13,  3.57it/s]
 96%|█████████▌| 357269/371472 [6:16:20<1:04:42,  3.66it/s]
 96%|█████████▌| 357270/371472 [6:16:20<1:09:18,  3.41it/s]
 96%|█████████▌| 357271/371472 [6:16:20<1:08:24,  3.46it/s]
 96%|█████████▌| 357272/371472 [6:16:21<1:06:53,  3.54it/s]
 96%|█████████▌| 357273/371472 [6:16:21<1:03:59,  3.70it/s]
 96%|█████████▌| 357274/371472 [6:16:21<1:04:50,  3.65it/s]
 96%|█████████▌| 357275/371472 [6:16:22<1:08:38,  3.45it/s]
 96%|█████████▌| 357276/371472 [6:16:22<1:07:04,  3.53it/s]
 96%|█████████▌| 357277/371472 [6:16:22<1:06:38,  3.55it/s]
 96%|█████████▌| 357278/371472 [6:16:22<1:06:23,  3.56it/s]
 96%|█████████▌| 357279/371472 [6:16:23<1:07:18,  3.51it/s]
 96%|█████████▌| 357280/371472 [6:16:23<1:05:05,  3.63it/s]
                                                           
{'loss': 2.6837, 'learning_rate': 1.3440280979982337e-07, 'epoch': 15.39}
+
 96%|█████████▌| 357280/371472 [6:16:23<1:05:05,  3.63it/s]
 96%|█████████▌| 357281/371472 [6:16:23<1:05:28,  3.61it/s]
 96%|█████████▌| 357282/371472 [6:16:24<1:05:17,  3.62it/s]
 96%|█████████▌| 357283/371472 [6:16:24<1:04:35,  3.66it/s]
 96%|█████████▌| 357284/371472 [6:16:24<1:04:38,  3.66it/s]
 96%|█████████▌| 357285/371472 [6:16:24<1:03:26,  3.73it/s]
 96%|█████████▌| 357286/371472 [6:16:25<1:04:54,  3.64it/s]
 96%|█████████▌| 357287/371472 [6:16:25<1:05:13,  3.62it/s]
 96%|█████████▌| 357288/371472 [6:16:25<1:06:09,  3.57it/s]
 96%|█████████▌| 357289/371472 [6:16:25<1:05:04,  3.63it/s]
 96%|█████████▌| 357290/371472 [6:16:26<1:05:52,  3.59it/s]
 96%|█████████▌| 357291/371472 [6:16:26<1:08:39,  3.44it/s]
 96%|█████████▌| 357292/371472 [6:16:26<1:05:27,  3.61it/s]
 96%|█████████▌| 357293/371472 [6:16:27<1:11:39,  3.30it/s]
 96%|█████████▌| 357294/371472 [6:16:27<1:14:07,  3.19it/s]
 96%|█████████▌| 357295/371472 [6:16:27<1:17:02,  3.07it/s]
 96%|█████████▌| 357296/371472 [6:16:28<1:13:53,  3.20it/s]
 96%|█████████▌| 357297/371472 [6:16:28<1:10:39,  3.34it/s]
 96%|█████████▌| 357298/371472 [6:16:28<1:07:12,  3.51it/s]
 96%|█████████▌| 357299/371472 [6:16:28<1:04:38,  3.65it/s]
 96%|█████████▌| 357300/371472 [6:16:29<1:05:45,  3.59it/s]
                                                           
{'loss': 2.5791, 'learning_rate': 1.343543278243444e-07, 'epoch': 15.39}
+
 96%|█████████▌| 357300/371472 [6:16:29<1:05:45,  3.59it/s]
 96%|█████████▌| 357301/371472 [6:16:29<1:06:53,  3.53it/s]
 96%|█████████▌| 357302/371472 [6:16:29<1:04:20,  3.67it/s]
 96%|█████████▌| 357303/371472 [6:16:29<1:04:40,  3.65it/s]
 96%|█████████▌| 357304/371472 [6:16:30<1:02:59,  3.75it/s]
 96%|█████████▌| 357305/371472 [6:16:30<1:03:15,  3.73it/s]
 96%|█████████▌| 357306/371472 [6:16:30<1:06:18,  3.56it/s]
 96%|█████████▌| 357307/371472 [6:16:31<1:04:19,  3.67it/s]
 96%|█████████▌| 357308/371472 [6:16:31<1:02:10,  3.80it/s]
 96%|█████████▌| 357309/371472 [6:16:31<1:04:45,  3.65it/s]
 96%|█████████▌| 357310/371472 [6:16:31<1:07:52,  3.48it/s]
 96%|█████████▌| 357311/371472 [6:16:32<1:04:32,  3.66it/s]
 96%|█████████▌| 357312/371472 [6:16:32<1:05:25,  3.61it/s]
 96%|█████████▌| 357313/371472 [6:16:32<1:05:03,  3.63it/s]
 96%|█████████▌| 357314/371472 [6:16:32<1:03:12,  3.73it/s]
 96%|█████████▌| 357315/371472 [6:16:33<1:03:23,  3.72it/s]
 96%|█████████▌| 357316/371472 [6:16:33<1:04:10,  3.68it/s]
 96%|█████████▌| 357317/371472 [6:16:33<1:06:50,  3.53it/s]
 96%|█████████▌| 357318/371472 [6:16:34<1:06:32,  3.55it/s]
 96%|█████████▌| 357319/371472 [6:16:34<1:11:04,  3.32it/s]
 96%|█████████▌| 357320/371472 [6:16:34<1:11:10,  3.31it/s]
                                                           
{'loss': 2.6322, 'learning_rate': 1.3430584584886556e-07, 'epoch': 15.39}
+
 96%|█████████▌| 357320/371472 [6:16:34<1:11:10,  3.31it/s]
 96%|█████████▌| 357321/371472 [6:16:35<1:11:11,  3.31it/s]
 96%|█████████▌| 357322/371472 [6:16:35<1:08:36,  3.44it/s]
 96%|█████████▌| 357323/371472 [6:16:35<1:08:35,  3.44it/s]
 96%|█████████▌| 357324/371472 [6:16:36<1:19:32,  2.96it/s]
 96%|█████████▌| 357325/371472 [6:16:36<1:15:12,  3.14it/s]
 96%|█████████▌| 357326/371472 [6:16:36<1:15:12,  3.13it/s]
 96%|█████████▌| 357327/371472 [6:16:36<1:13:52,  3.19it/s]
 96%|█████████▌| 357328/371472 [6:16:37<1:16:37,  3.08it/s]
 96%|█████████▌| 357329/371472 [6:16:37<1:12:07,  3.27it/s]
 96%|█████████▌| 357330/371472 [6:16:37<1:11:46,  3.28it/s]
 96%|█████████▌| 357331/371472 [6:16:38<1:08:13,  3.45it/s]
 96%|█████████▌| 357332/371472 [6:16:38<1:05:40,  3.59it/s]
 96%|█████████▌| 357333/371472 [6:16:38<1:09:20,  3.40it/s]
 96%|█████████▌| 357334/371472 [6:16:38<1:07:00,  3.52it/s]
 96%|█████████▌| 357335/371472 [6:16:39<1:10:12,  3.36it/s]
 96%|█████████▌| 357336/371472 [6:16:39<1:09:13,  3.40it/s]
 96%|█████████▌| 357337/371472 [6:16:39<1:12:24,  3.25it/s]
 96%|█████████▌| 357338/371472 [6:16:40<1:10:33,  3.34it/s]
 96%|█████████▌| 357339/371472 [6:16:40<1:07:29,  3.49it/s]
 96%|█████████▌| 357340/371472 [6:16:40<1:04:15,  3.67it/s]
                                                           
{'loss': 2.532, 'learning_rate': 1.3425736387338663e-07, 'epoch': 15.39}
+
 96%|█████████▌| 357340/371472 [6:16:40<1:04:15,  3.67it/s]
 96%|█████████▌| 357341/371472 [6:16:40<1:03:33,  3.71it/s]
 96%|█████████▌| 357342/371472 [6:16:41<1:12:53,  3.23it/s]
 96%|█████████▌| 357343/371472 [6:16:41<1:09:31,  3.39it/s]
 96%|█████████▌| 357344/371472 [6:16:41<1:10:06,  3.36it/s]
 96%|█████████▌| 357345/371472 [6:16:42<1:07:32,  3.49it/s]
 96%|█████████▌| 357346/371472 [6:16:42<1:11:42,  3.28it/s]
 96%|█████████▌| 357347/371472 [6:16:42<1:07:14,  3.50it/s]
 96%|█████████▌| 357348/371472 [6:16:43<1:05:46,  3.58it/s]
 96%|█████████▌| 357349/371472 [6:16:43<1:06:57,  3.52it/s]
 96%|█████████▌| 357350/371472 [6:16:43<1:12:04,  3.27it/s]
 96%|█████████▌| 357351/371472 [6:16:43<1:08:41,  3.43it/s]
 96%|█████████▌| 357352/371472 [6:16:44<1:06:35,  3.53it/s]
 96%|█████████▌| 357353/371472 [6:16:44<1:05:43,  3.58it/s]
 96%|█████████▌| 357354/371472 [6:16:44<1:07:00,  3.51it/s]
 96%|█████████▌| 357355/371472 [6:16:45<1:07:12,  3.50it/s]
 96%|█████████▌| 357356/371472 [6:16:45<1:05:34,  3.59it/s]
 96%|█████████▌| 357357/371472 [6:16:45<1:09:35,  3.38it/s]
 96%|█████████▌| 357358/371472 [6:16:46<1:15:00,  3.14it/s]
 96%|█████████▌| 357359/371472 [6:16:46<1:13:48,  3.19it/s]
 96%|█████████▌| 357360/371472 [6:16:46<1:13:09,  3.21it/s]
                                                           
{'loss': 2.5935, 'learning_rate': 1.3420888189790768e-07, 'epoch': 15.39}
+
 96%|█████████▌| 357360/371472 [6:16:46<1:13:09,  3.21it/s]
 96%|█████████▌| 357361/371472 [6:16:46<1:13:16,  3.21it/s]
 96%|█████████▌| 357362/371472 [6:16:47<1:15:12,  3.13it/s]
 96%|█████████▌| 357363/371472 [6:16:47<1:15:03,  3.13it/s]
 96%|█████████▌| 357364/371472 [6:16:47<1:12:47,  3.23it/s]
 96%|█████████▌| 357365/371472 [6:16:48<1:13:31,  3.20it/s]
 96%|█████████▌| 357366/371472 [6:16:48<1:11:50,  3.27it/s]
 96%|█████████▌| 357367/371472 [6:16:48<1:12:38,  3.24it/s]
 96%|█████████▌| 357368/371472 [6:16:49<1:09:59,  3.36it/s]
 96%|█████████▌| 357369/371472 [6:16:49<1:09:17,  3.39it/s]
 96%|█████████▌| 357370/371472 [6:16:49<1:07:08,  3.50it/s]
 96%|█████████▌| 357371/371472 [6:16:49<1:08:36,  3.43it/s]
 96%|█████████▌| 357372/371472 [6:16:50<1:08:24,  3.44it/s]
 96%|█████████▌| 357373/371472 [6:16:50<1:06:29,  3.53it/s]
 96%|█████████▌| 357374/371472 [6:16:50<1:04:32,  3.64it/s]
 96%|█████████▌| 357375/371472 [6:16:51<1:06:08,  3.55it/s]
 96%|█████████▌| 357376/371472 [6:16:51<1:03:52,  3.68it/s]
 96%|█████████▌| 357377/371472 [6:16:51<1:01:45,  3.80it/s]
 96%|█████████▌| 357378/371472 [6:16:51<1:00:43,  3.87it/s]
 96%|█████████▌| 357379/371472 [6:16:52<1:01:36,  3.81it/s]
 96%|█████████▌| 357380/371472 [6:16:52<1:03:42,  3.69it/s]
                                                           
{'loss': 2.5235, 'learning_rate': 1.3416039992242883e-07, 'epoch': 15.39}
+
 96%|█████████▌| 357380/371472 [6:16:52<1:03:42,  3.69it/s]
 96%|█████████▌| 357381/371472 [6:16:52<1:07:15,  3.49it/s]
 96%|█████████▌| 357382/371472 [6:16:52<1:04:32,  3.64it/s]
 96%|█████████▌| 357383/371472 [6:16:53<1:02:09,  3.78it/s]
 96%|█████████▌| 357384/371472 [6:16:53<1:02:03,  3.78it/s]
 96%|█████████▌| 357385/371472 [6:16:53<1:05:16,  3.60it/s]
 96%|█████████▌| 357386/371472 [6:16:54<1:04:42,  3.63it/s]
 96%|█████████▌| 357387/371472 [6:16:54<1:03:08,  3.72it/s]
 96%|█████████▌| 357388/371472 [6:16:54<1:02:26,  3.76it/s]
 96%|█████████▌| 357389/371472 [6:16:54<1:01:05,  3.84it/s]
 96%|█████████▌| 357390/371472 [6:16:55<1:02:32,  3.75it/s]
 96%|█████████▌| 357391/371472 [6:16:55<1:03:36,  3.69it/s]
 96%|█████████▌| 357392/371472 [6:16:55<1:06:34,  3.52it/s]
 96%|█████████▌| 357393/371472 [6:16:55<1:04:48,  3.62it/s]
 96%|█████████▌| 357394/371472 [6:16:56<1:05:27,  3.58it/s]
 96%|█████████▌| 357395/371472 [6:16:56<1:07:25,  3.48it/s]
 96%|█████████▌| 357396/371472 [6:16:56<1:04:51,  3.62it/s]
 96%|█████████▌| 357397/371472 [6:16:57<1:05:09,  3.60it/s]
 96%|█████████▌| 357398/371472 [6:16:57<1:06:09,  3.55it/s]
 96%|█████████▌| 357399/371472 [6:16:57<1:04:20,  3.65it/s]
 96%|█████████▌| 357400/371472 [6:16:57<1:03:06,  3.72it/s]
                                                           
{'loss': 2.6444, 'learning_rate': 1.341119179469499e-07, 'epoch': 15.39}
+
 96%|█████████▌| 357400/371472 [6:16:57<1:03:06,  3.72it/s]
 96%|█████████▌| 357401/371472 [6:16:58<1:01:20,  3.82it/s]
 96%|█████████▌| 357402/371472 [6:16:58<1:03:07,  3.72it/s]
 96%|█████████▌| 357403/371472 [6:16:58<1:07:54,  3.45it/s]
 96%|█████████▌| 357404/371472 [6:16:58<1:04:57,  3.61it/s]
 96%|█████████▌| 357405/371472 [6:16:59<1:04:39,  3.63it/s]
 96%|█████████▌| 357406/371472 [6:16:59<1:06:57,  3.50it/s]
 96%|█████████▌| 357407/371472 [6:16:59<1:10:22,  3.33it/s]
 96%|█████████▌| 357408/371472 [6:17:00<1:07:41,  3.46it/s]
 96%|█████████▌| 357409/371472 [6:17:00<1:06:36,  3.52it/s]
 96%|█████████▌| 357410/371472 [6:17:00<1:05:28,  3.58it/s]
 96%|█████████▌| 357411/371472 [6:17:00<1:04:00,  3.66it/s]
 96%|█████████▌| 357412/371472 [6:17:01<1:09:06,  3.39it/s]
 96%|█████████▌| 357413/371472 [6:17:01<1:16:54,  3.05it/s]
 96%|█████████▌| 357414/371472 [6:17:02<1:15:23,  3.11it/s]
 96%|█████████▌| 357415/371472 [6:17:02<1:11:41,  3.27it/s]
 96%|█████████▌| 357416/371472 [6:17:02<1:10:46,  3.31it/s]
 96%|█████████▌| 357417/371472 [6:17:02<1:13:56,  3.17it/s]
 96%|█████████▌| 357418/371472 [6:17:03<1:11:40,  3.27it/s]
 96%|█████████▌| 357419/371472 [6:17:03<1:08:29,  3.42it/s]
 96%|█████████▌| 357420/371472 [6:17:03<1:14:38,  3.14it/s]
                                                           
{'loss': 2.3656, 'learning_rate': 1.3406343597147105e-07, 'epoch': 15.39}
+
 96%|█████████▌| 357420/371472 [6:17:03<1:14:38,  3.14it/s]
 96%|█████████▌| 357421/371472 [6:17:04<1:11:58,  3.25it/s]
 96%|█████████▌| 357422/371472 [6:17:04<1:09:13,  3.38it/s]
 96%|█████████▌| 357423/371472 [6:17:04<1:07:07,  3.49it/s]
 96%|█████████▌| 357424/371472 [6:17:04<1:04:25,  3.63it/s]
 96%|█████████▌| 357425/371472 [6:17:05<1:04:56,  3.61it/s]
 96%|█████████▌| 357426/371472 [6:17:05<1:06:11,  3.54it/s]
 96%|█████████▌| 357427/371472 [6:17:05<1:07:14,  3.48it/s]
 96%|█████████▌| 357428/371472 [6:17:06<1:06:13,  3.53it/s]
 96%|█████████▌| 357429/371472 [6:17:06<1:04:50,  3.61it/s]
 96%|█████████▌| 357430/371472 [6:17:06<1:07:48,  3.45it/s]
 96%|█████████▌| 357431/371472 [6:17:06<1:06:52,  3.50it/s]
 96%|█████████▌| 357432/371472 [6:17:07<1:06:45,  3.50it/s]
 96%|█████████▌| 357433/371472 [6:17:07<1:03:58,  3.66it/s]
 96%|█████████▌| 357434/371472 [6:17:07<1:05:49,  3.55it/s]
 96%|█████████▌| 357435/371472 [6:17:08<1:06:26,  3.52it/s]
 96%|█████████▌| 357436/371472 [6:17:08<1:06:13,  3.53it/s]
 96%|█████████▌| 357437/371472 [6:17:08<1:08:56,  3.39it/s]
 96%|█████████▌| 357438/371472 [6:17:08<1:06:41,  3.51it/s]
 96%|█████████▌| 357439/371472 [6:17:09<1:08:16,  3.43it/s]
 96%|█████████▌| 357440/371472 [6:17:09<1:06:55,  3.49it/s]
                                                           
{'loss': 2.6019, 'learning_rate': 1.340149539959921e-07, 'epoch': 15.4}
+
 96%|█████████▌| 357440/371472 [6:17:09<1:06:55,  3.49it/s]
 96%|█████████▌| 357441/371472 [6:17:09<1:07:35,  3.46it/s]
 96%|█████████▌| 357442/371472 [6:17:10<1:10:11,  3.33it/s]
 96%|█████████▌| 357443/371472 [6:17:10<1:08:45,  3.40it/s]
 96%|█████████▌| 357444/371472 [6:17:10<1:09:28,  3.37it/s]
 96%|█████████▌| 357445/371472 [6:17:10<1:06:29,  3.52it/s]
 96%|█████████▌| 357446/371472 [6:17:11<1:04:00,  3.65it/s]
 96%|█████████▌| 357447/371472 [6:17:11<1:03:14,  3.70it/s]
 96%|█████████▌| 357448/371472 [6:17:11<1:07:11,  3.48it/s]
 96%|█████████▌| 357449/371472 [6:17:12<1:05:40,  3.56it/s]
 96%|█████████▌| 357450/371472 [6:17:12<1:07:07,  3.48it/s]
 96%|█████████▌| 357451/371472 [6:17:12<1:06:22,  3.52it/s]
 96%|█████████▌| 357452/371472 [6:17:12<1:05:11,  3.58it/s]
 96%|█████████▌| 357453/371472 [6:17:13<1:03:38,  3.67it/s]
 96%|█████████▌| 357454/371472 [6:17:13<1:03:54,  3.66it/s]
 96%|█████████▌| 357455/371472 [6:17:13<1:02:49,  3.72it/s]
 96%|█████████▌| 357456/371472 [6:17:13<1:04:14,  3.64it/s]
 96%|█████████▌| 357457/371472 [6:17:14<1:03:56,  3.65it/s]
 96%|█████████▌| 357458/371472 [6:17:14<1:02:35,  3.73it/s]
 96%|█████████▌| 357459/371472 [6:17:14<1:03:53,  3.65it/s]
 96%|█████████▌| 357460/371472 [6:17:15<1:02:47,  3.72it/s]
                                                           
{'loss': 2.4433, 'learning_rate': 1.3396647202051327e-07, 'epoch': 15.4}
+
 96%|█████████▌| 357460/371472 [6:17:15<1:02:47,  3.72it/s]
 96%|█████████▌| 357461/371472 [6:17:15<1:04:53,  3.60it/s]
 96%|█████████▌| 357462/371472 [6:17:15<1:05:25,  3.57it/s]
 96%|█████████▌| 357463/371472 [6:17:15<1:05:52,  3.54it/s]
 96%|█████████▌| 357464/371472 [6:17:16<1:05:28,  3.57it/s]
 96%|█████████▌| 357465/371472 [6:17:16<1:04:54,  3.60it/s]
 96%|█████████▌| 357466/371472 [6:17:16<1:05:33,  3.56it/s]
 96%|█████████▌| 357467/371472 [6:17:17<1:05:30,  3.56it/s]
 96%|█████████▌| 357468/371472 [6:17:17<1:06:31,  3.51it/s]
 96%|█████████▌| 357469/371472 [6:17:17<1:14:13,  3.14it/s]
 96%|█████████▌| 357470/371472 [6:17:18<1:17:32,  3.01it/s]
 96%|█████████▌| 357471/371472 [6:17:18<1:11:33,  3.26it/s]
 96%|█████████▌| 357472/371472 [6:17:18<1:08:38,  3.40it/s]
 96%|█████████▌| 357473/371472 [6:17:18<1:09:31,  3.36it/s]
 96%|█████████▌| 357474/371472 [6:17:19<1:07:07,  3.48it/s]
 96%|█████████▌| 357475/371472 [6:17:19<1:06:41,  3.50it/s]
 96%|█████████▌| 357476/371472 [6:17:19<1:03:45,  3.66it/s]
 96%|█████████▌| 357477/371472 [6:17:19<1:04:12,  3.63it/s]
 96%|█████████▌| 357478/371472 [6:17:20<1:07:08,  3.47it/s]
 96%|█████████▌| 357479/371472 [6:17:20<1:03:29,  3.67it/s]
 96%|█████████▌| 357480/371472 [6:17:20<1:03:23,  3.68it/s]
                                                           
{'loss': 2.5902, 'learning_rate': 1.3391799004503432e-07, 'epoch': 15.4}
+
 96%|█████████▌| 357480/371472 [6:17:20<1:03:23,  3.68it/s]
 96%|█████████▌| 357481/371472 [6:17:21<1:03:33,  3.67it/s]
 96%|█████████▌| 357482/371472 [6:17:21<1:02:47,  3.71it/s]
 96%|█████████▌| 357483/371472 [6:17:21<1:02:50,  3.71it/s]
 96%|█████████▌| 357484/371472 [6:17:21<1:03:35,  3.67it/s]
 96%|█████████▌| 357485/371472 [6:17:22<1:12:35,  3.21it/s]
 96%|█████████▌| 357486/371472 [6:17:22<1:09:57,  3.33it/s]
 96%|█████████▌| 357487/371472 [6:17:22<1:09:53,  3.33it/s]
 96%|█████████▌| 357488/371472 [6:17:23<1:08:26,  3.41it/s]
 96%|█████████▌| 357489/371472 [6:17:23<1:10:04,  3.33it/s]
 96%|█████████▌| 357490/371472 [6:17:23<1:07:02,  3.48it/s]
 96%|█████████▌| 357491/371472 [6:17:23<1:04:51,  3.59it/s]
 96%|█████████▌| 357492/371472 [6:17:24<1:04:05,  3.64it/s]
 96%|█████████▌| 357493/371472 [6:17:24<1:03:51,  3.65it/s]
 96%|█████████▌| 357494/371472 [6:17:24<1:04:12,  3.63it/s]
 96%|█████████▌| 357495/371472 [6:17:25<1:10:49,  3.29it/s]
 96%|█████████▌| 357496/371472 [6:17:25<1:06:55,  3.48it/s]
 96%|█████████▌| 357497/371472 [6:17:25<1:09:47,  3.34it/s]
 96%|█████████▌| 357498/371472 [6:17:25<1:06:01,  3.53it/s]
 96%|█████████▌| 357499/371472 [6:17:26<1:05:32,  3.55it/s]
 96%|█████████▌| 357500/371472 [6:17:26<1:05:44,  3.54it/s]
                                                           
{'loss': 2.6976, 'learning_rate': 1.3386950806955547e-07, 'epoch': 15.4}
+
 96%|█████████▌| 357500/371472 [6:17:26<1:05:44,  3.54it/s]
 96%|█████████▌| 357501/371472 [6:17:26<1:05:57,  3.53it/s]
 96%|█████████▌| 357502/371472 [6:17:27<1:02:53,  3.70it/s]
 96%|█████████▌| 357503/371472 [6:17:27<1:03:46,  3.65it/s]
 96%|█████████▌| 357504/371472 [6:17:27<1:03:45,  3.65it/s]
 96%|█████████▌| 357505/371472 [6:17:27<1:01:55,  3.76it/s]
 96%|█████████▌| 357506/371472 [6:17:28<1:02:28,  3.73it/s]
 96%|█████████▌| 357507/371472 [6:17:28<1:07:17,  3.46it/s]
 96%|█████████▌| 357508/371472 [6:17:28<1:04:26,  3.61it/s]
 96%|█████████▌| 357509/371472 [6:17:29<1:06:29,  3.50it/s]
 96%|█████████▌| 357510/371472 [6:17:29<1:04:31,  3.61it/s]
 96%|█████████▌| 357511/371472 [6:17:29<1:04:17,  3.62it/s]
 96%|█████████▌| 357512/371472 [6:17:29<1:06:13,  3.51it/s]
 96%|█████████▌| 357513/371472 [6:17:30<1:15:01,  3.10it/s]
 96%|█████████▌| 357514/371472 [6:17:30<1:10:33,  3.30it/s]
 96%|█████████▌| 357515/371472 [6:17:30<1:12:47,  3.20it/s]
 96%|█████████▌| 357516/371472 [6:17:31<1:14:17,  3.13it/s]
 96%|█████████▌| 357517/371472 [6:17:31<1:11:25,  3.26it/s]
 96%|█████████▌| 357518/371472 [6:17:31<1:07:20,  3.45it/s]
 96%|█████████▌| 357519/371472 [6:17:32<1:06:34,  3.49it/s]
 96%|█████████▌| 357520/371472 [6:17:32<1:06:10,  3.51it/s]
                                                           
{'loss': 2.5105, 'learning_rate': 1.3382102609407654e-07, 'epoch': 15.4}
+
 96%|█████████▌| 357520/371472 [6:17:32<1:06:10,  3.51it/s]
 96%|█████████▌| 357521/371472 [6:17:32<1:03:20,  3.67it/s]
 96%|█████████▌| 357522/371472 [6:17:32<1:03:34,  3.66it/s]
 96%|█████████▌| 357523/371472 [6:17:33<1:00:46,  3.82it/s]
 96%|█████████▌| 357524/371472 [6:17:33<1:02:03,  3.75it/s]
 96%|█████████▌| 357525/371472 [6:17:33<1:03:42,  3.65it/s]
 96%|█████████▌| 357526/371472 [6:17:33<1:05:58,  3.52it/s]
 96%|█████████▌| 357527/371472 [6:17:34<1:06:34,  3.49it/s]
 96%|█████████▌| 357528/371472 [6:17:34<1:05:11,  3.56it/s]
 96%|█████████▌| 357529/371472 [6:17:34<1:03:56,  3.63it/s]
 96%|█████████▌| 357530/371472 [6:17:35<1:04:19,  3.61it/s]
 96%|█████████▌| 357531/371472 [6:17:35<1:04:49,  3.58it/s]
 96%|█████████▌| 357532/371472 [6:17:35<1:06:11,  3.51it/s]
 96%|█████████▌| 357533/371472 [6:17:35<1:07:06,  3.46it/s]
 96%|█████████▌| 357534/371472 [6:17:36<1:09:11,  3.36it/s]
 96%|█████████▌| 357535/371472 [6:17:36<1:07:37,  3.43it/s]
 96%|█████████▌| 357536/371472 [6:17:36<1:14:44,  3.11it/s]
 96%|█████████▌| 357537/371472 [6:17:37<1:10:14,  3.31it/s]
 96%|█████████▌| 357538/371472 [6:17:37<1:08:58,  3.37it/s]
 96%|█████████▌| 357539/371472 [6:17:37<1:06:49,  3.47it/s]
 96%|█████████▌| 357540/371472 [6:17:37<1:03:53,  3.63it/s]
                                                           
{'loss': 2.5551, 'learning_rate': 1.337725441185977e-07, 'epoch': 15.4}
+
 96%|█████████▌| 357540/371472 [6:17:37<1:03:53,  3.63it/s]
 96%|█████████▌| 357541/371472 [6:17:38<1:03:14,  3.67it/s]
 96%|█████████▋| 357542/371472 [6:17:38<1:02:22,  3.72it/s]
 96%|█████████▋| 357543/371472 [6:17:38<1:07:52,  3.42it/s]
 96%|█████████▋| 357544/371472 [6:17:39<1:35:39,  2.43it/s]
 96%|█████████▋| 357545/371472 [6:17:39<1:25:06,  2.73it/s]
 96%|█████████▋| 357546/371472 [6:17:40<1:17:33,  2.99it/s]
 96%|█████████▋| 357547/371472 [6:17:40<1:14:12,  3.13it/s]
 96%|█████████▋| 357548/371472 [6:17:40<1:09:24,  3.34it/s]
 96%|█████████▋| 357549/371472 [6:17:40<1:09:26,  3.34it/s]
 96%|█████████▋| 357550/371472 [6:17:41<1:06:58,  3.46it/s]
 96%|█████████▋| 357551/371472 [6:17:41<1:04:12,  3.61it/s]
 96%|█████████▋| 357552/371472 [6:17:41<1:01:57,  3.74it/s]
 96%|█████████▋| 357553/371472 [6:17:41<1:02:08,  3.73it/s]
 96%|█████████▋| 357554/371472 [6:17:42<1:02:01,  3.74it/s]
 96%|█████████▋| 357555/371472 [6:17:42<1:02:18,  3.72it/s]
 96%|█████████▋| 357556/371472 [6:17:42<1:02:06,  3.73it/s]
 96%|█████████▋| 357557/371472 [6:17:42<1:01:36,  3.76it/s]
 96%|█████████▋| 357558/371472 [6:17:43<1:01:37,  3.76it/s]
 96%|█████████▋| 357559/371472 [6:17:43<1:04:53,  3.57it/s]
 96%|█████████▋| 357560/371472 [6:17:43<1:04:55,  3.57it/s]
                                                           
{'loss': 2.6433, 'learning_rate': 1.3372406214311876e-07, 'epoch': 15.4}
+
 96%|█████████▋| 357560/371472 [6:17:43<1:04:55,  3.57it/s]
 96%|█████████▋| 357561/371472 [6:17:44<1:06:42,  3.48it/s]
 96%|█████████▋| 357562/371472 [6:17:44<1:04:46,  3.58it/s]
 96%|█████████▋| 357563/371472 [6:17:44<1:02:32,  3.71it/s]
 96%|█████████▋| 357564/371472 [6:17:44<1:00:47,  3.81it/s]
 96%|█████████▋| 357565/371472 [6:17:45<1:02:06,  3.73it/s]
 96%|█████████▋| 357566/371472 [6:17:45<59:59,  3.86it/s]  
 96%|█████████▋| 357567/371472 [6:17:45<1:02:09,  3.73it/s]
 96%|█████████▋| 357568/371472 [6:17:45<1:03:42,  3.64it/s]
 96%|█████████▋| 357569/371472 [6:17:46<1:02:00,  3.74it/s]
 96%|█████████▋| 357570/371472 [6:17:46<1:01:53,  3.74it/s]
 96%|█████████▋| 357571/371472 [6:17:46<1:04:22,  3.60it/s]
 96%|█████████▋| 357572/371472 [6:17:47<1:05:54,  3.52it/s]
 96%|█████████▋| 357573/371472 [6:17:47<1:06:25,  3.49it/s]
 96%|█████████▋| 357574/371472 [6:17:47<1:03:14,  3.66it/s]
 96%|█████████▋| 357575/371472 [6:17:47<1:04:56,  3.57it/s]
 96%|█████████▋| 357576/371472 [6:17:48<1:03:19,  3.66it/s]
 96%|█████████▋| 357577/371472 [6:17:48<1:01:54,  3.74it/s]
 96%|█████████▋| 357578/371472 [6:17:48<1:05:56,  3.51it/s]
 96%|█████████▋| 357579/371472 [6:17:49<1:08:38,  3.37it/s]
 96%|█████████▋| 357580/371472 [6:17:49<1:07:16,  3.44it/s]
                                                           
{'loss': 2.7044, 'learning_rate': 1.3367558016763991e-07, 'epoch': 15.4}
+
 96%|█████████▋| 357580/371472 [6:17:49<1:07:16,  3.44it/s]
 96%|█████████▋| 357581/371472 [6:17:49<1:09:43,  3.32it/s]
 96%|█████████▋| 357582/371472 [6:17:49<1:06:41,  3.47it/s]
 96%|█████████▋| 357583/371472 [6:17:50<1:08:00,  3.40it/s]
 96%|█████████▋| 357584/371472 [6:17:50<1:10:09,  3.30it/s]
 96%|█████████▋| 357585/371472 [6:17:50<1:08:05,  3.40it/s]
 96%|█████████▋| 357586/371472 [6:17:51<1:11:33,  3.23it/s]
 96%|█████████▋| 357587/371472 [6:17:51<1:09:07,  3.35it/s]
 96%|█████████▋| 357588/371472 [6:17:51<1:07:26,  3.43it/s]
 96%|█████████▋| 357589/371472 [6:17:52<1:05:46,  3.52it/s]
 96%|█████████▋| 357590/371472 [6:17:52<1:04:45,  3.57it/s]
 96%|█████████▋| 357591/371472 [6:17:52<1:04:00,  3.61it/s]
 96%|█████████▋| 357592/371472 [6:17:52<1:01:17,  3.77it/s]
 96%|█████████▋| 357593/371472 [6:17:53<1:01:53,  3.74it/s]
 96%|█████████▋| 357594/371472 [6:17:53<1:04:56,  3.56it/s]
 96%|█████████▋| 357595/371472 [6:17:53<1:04:41,  3.57it/s]
 96%|█████████▋| 357596/371472 [6:17:53<1:06:53,  3.46it/s]
 96%|█████████▋| 357597/371472 [6:17:54<1:14:08,  3.12it/s]
 96%|█████████▋| 357598/371472 [6:17:54<1:09:14,  3.34it/s]
 96%|█████████▋| 357599/371472 [6:17:54<1:10:43,  3.27it/s]
 96%|█████████▋| 357600/371472 [6:17:55<1:08:03,  3.40it/s]
                                                           
{'loss': 2.6796, 'learning_rate': 1.3362709819216099e-07, 'epoch': 15.4}
+
 96%|█████████▋| 357600/371472 [6:17:55<1:08:03,  3.40it/s]
 96%|█████████▋| 357601/371472 [6:17:55<1:10:44,  3.27it/s]
 96%|█████████▋| 357602/371472 [6:17:55<1:08:37,  3.37it/s]
 96%|█████████▋| 357603/371472 [6:17:56<1:06:34,  3.47it/s]
 96%|█████████▋| 357604/371472 [6:17:56<1:04:35,  3.58it/s]
 96%|█████████▋| 357605/371472 [6:17:56<1:03:43,  3.63it/s]
 96%|█████████▋| 357606/371472 [6:17:56<1:03:16,  3.65it/s]
 96%|█████████▋| 357607/371472 [6:17:57<1:07:17,  3.43it/s]
 96%|█████████▋| 357608/371472 [6:17:57<1:06:04,  3.50it/s]
 96%|█████████▋| 357609/371472 [6:17:57<1:05:31,  3.53it/s]
 96%|█████████▋| 357610/371472 [6:17:58<1:05:03,  3.55it/s]
 96%|█████████▋| 357611/371472 [6:17:58<1:04:04,  3.61it/s]
 96%|█████████▋| 357612/371472 [6:17:58<1:03:28,  3.64it/s]
 96%|█████████▋| 357613/371472 [6:17:58<1:06:15,  3.49it/s]
 96%|█████████▋| 357614/371472 [6:17:59<1:02:40,  3.69it/s]
 96%|█████████▋| 357615/371472 [6:17:59<1:02:03,  3.72it/s]
 96%|█████████▋| 357616/371472 [6:17:59<1:02:05,  3.72it/s]
 96%|█████████▋| 357617/371472 [6:17:59<1:02:35,  3.69it/s]
 96%|█████████▋| 357618/371472 [6:18:00<1:01:40,  3.74it/s]
 96%|█████████▋| 357619/371472 [6:18:00<1:04:23,  3.59it/s]
 96%|█████████▋| 357620/371472 [6:18:00<1:02:54,  3.67it/s]
                                                           
{'loss': 2.5611, 'learning_rate': 1.3357861621668214e-07, 'epoch': 15.4}
+
 96%|█████████▋| 357620/371472 [6:18:00<1:02:54,  3.67it/s]
 96%|█████████▋| 357621/371472 [6:18:01<1:02:46,  3.68it/s]
 96%|█████████▋| 357622/371472 [6:18:01<1:03:31,  3.63it/s]
 96%|█████████▋| 357623/371472 [6:18:01<1:09:15,  3.33it/s]
 96%|█████████▋| 357624/371472 [6:18:01<1:08:01,  3.39it/s]
 96%|█████████▋| 357625/371472 [6:18:02<1:15:03,  3.07it/s]
 96%|█████████▋| 357626/371472 [6:18:02<1:10:21,  3.28it/s]
 96%|█████████▋| 357627/371472 [6:18:02<1:05:44,  3.51it/s]
 96%|█████████▋| 357628/371472 [6:18:03<1:02:41,  3.68it/s]
 96%|█████████▋| 357629/371472 [6:18:03<59:59,  3.85it/s]  
 96%|█████████▋| 357630/371472 [6:18:03<59:37,  3.87it/s]
 96%|█████████▋| 357631/371472 [6:18:03<59:59,  3.84it/s]
 96%|█████████▋| 357632/371472 [6:18:04<1:02:04,  3.72it/s]
 96%|█████████▋| 357633/371472 [6:18:04<1:04:05,  3.60it/s]
 96%|█████████▋| 357634/371472 [6:18:04<1:02:37,  3.68it/s]
 96%|█████████▋| 357635/371472 [6:18:04<1:03:35,  3.63it/s]
 96%|█████████▋| 357636/371472 [6:18:05<1:04:03,  3.60it/s]
 96%|█████████▋| 357637/371472 [6:18:05<1:02:11,  3.71it/s]
 96%|█████████▋| 357638/371472 [6:18:05<1:00:06,  3.84it/s]
 96%|█████████▋| 357639/371472 [6:18:05<59:08,  3.90it/s]  
 96%|█████████▋| 357640/371472 [6:18:06<59:54,  3.85it/s]
                                                         
{'loss': 2.684, 'learning_rate': 1.3353013424120318e-07, 'epoch': 15.4}
+
 96%|█████████▋| 357640/371472 [6:18:06<59:54,  3.85it/s]
 96%|█████████▋| 357641/371472 [6:18:06<59:49,  3.85it/s]
 96%|█████████▋| 357642/371472 [6:18:06<1:02:10,  3.71it/s]
 96%|█████████▋| 357643/371472 [6:18:07<1:03:07,  3.65it/s]
 96%|█████████▋| 357644/371472 [6:18:07<1:02:56,  3.66it/s]
 96%|█████████▋| 357645/371472 [6:18:07<1:03:43,  3.62it/s]
 96%|█████████▋| 357646/371472 [6:18:07<1:04:24,  3.58it/s]
 96%|█████████▋| 357647/371472 [6:18:08<1:02:57,  3.66it/s]
 96%|█████████▋| 357648/371472 [6:18:08<1:05:17,  3.53it/s]
 96%|█████████▋| 357649/371472 [6:18:08<1:07:55,  3.39it/s]
 96%|█████████▋| 357650/371472 [6:18:09<1:08:46,  3.35it/s]
 96%|█████████▋| 357651/371472 [6:18:09<1:06:15,  3.48it/s]
 96%|█████████▋| 357652/371472 [6:18:09<1:03:12,  3.64it/s]
 96%|█████████▋| 357653/371472 [6:18:09<1:02:31,  3.68it/s]
 96%|█████████▋| 357654/371472 [6:18:10<1:08:58,  3.34it/s]
 96%|█████████▋| 357655/371472 [6:18:10<1:09:23,  3.32it/s]
 96%|█████████▋| 357656/371472 [6:18:10<1:08:07,  3.38it/s]
 96%|█████████▋| 357657/371472 [6:18:11<1:04:46,  3.55it/s]
 96%|█████████▋| 357658/371472 [6:18:11<1:02:16,  3.70it/s]
 96%|█████████▋| 357659/371472 [6:18:11<1:01:55,  3.72it/s]
 96%|█████████▋| 357660/371472 [6:18:11<1:00:16,  3.82it/s]
                                                           
{'loss': 2.5946, 'learning_rate': 1.3348165226572436e-07, 'epoch': 15.41}
+
 96%|█████████▋| 357660/371472 [6:18:11<1:00:16,  3.82it/s]
 96%|█████████▋| 357661/371472 [6:18:12<1:01:26,  3.75it/s]
 96%|█████████▋| 357662/371472 [6:18:12<59:37,  3.86it/s]  
 96%|█████████▋| 357663/371472 [6:18:12<1:04:28,  3.57it/s]
 96%|█████████▋| 357664/371472 [6:18:12<1:03:56,  3.60it/s]
 96%|█████████▋| 357665/371472 [6:18:13<1:02:30,  3.68it/s]
 96%|█████████▋| 357666/371472 [6:18:13<1:02:29,  3.68it/s]
 96%|█████████▋| 357667/371472 [6:18:13<1:02:28,  3.68it/s]
 96%|█████████▋| 357668/371472 [6:18:14<1:00:45,  3.79it/s]
 96%|█████████▋| 357669/371472 [6:18:14<59:12,  3.88it/s]  
 96%|█████████▋| 357670/371472 [6:18:14<57:29,  4.00it/s]
 96%|█████████▋| 357671/371472 [6:18:14<1:01:25,  3.74it/s]
 96%|█████████▋| 357672/371472 [6:18:15<1:03:36,  3.62it/s]
 96%|█████████▋| 357673/371472 [6:18:15<1:04:08,  3.59it/s]
 96%|█████████▋| 357674/371472 [6:18:15<1:01:09,  3.76it/s]
 96%|█████████▋| 357675/371472 [6:18:15<1:02:57,  3.65it/s]
 96%|█████████▋| 357676/371472 [6:18:16<1:06:54,  3.44it/s]
 96%|█████████▋| 357677/371472 [6:18:16<1:05:52,  3.49it/s]
 96%|█████████▋| 357678/371472 [6:18:16<1:08:54,  3.34it/s]
 96%|█████████▋| 357679/371472 [6:18:17<1:07:51,  3.39it/s]
 96%|█████████���| 357680/371472 [6:18:17<1:06:12,  3.47it/s]
                                                           
{'loss': 2.4644, 'learning_rate': 1.334331702902454e-07, 'epoch': 15.41}
+
 96%|█████████▋| 357680/371472 [6:18:17<1:06:12,  3.47it/s]
 96%|█████████▋| 357681/371472 [6:18:17<1:07:58,  3.38it/s]
 96%|█████████▋| 357682/371472 [6:18:17<1:05:23,  3.51it/s]
 96%|█████████▋| 357683/371472 [6:18:18<1:03:29,  3.62it/s]
 96%|█████████▋| 357684/371472 [6:18:18<1:07:42,  3.39it/s]
 96%|█████████▋| 357685/371472 [6:18:18<1:03:33,  3.62it/s]
 96%|█████████▋| 357686/371472 [6:18:19<1:02:33,  3.67it/s]
 96%|█████████▋| 357687/371472 [6:18:19<1:03:00,  3.65it/s]
 96%|█████████▋| 357688/371472 [6:18:19<1:02:56,  3.65it/s]
 96%|█████████▋| 357689/371472 [6:18:19<1:01:44,  3.72it/s]
 96%|█████████▋| 357690/371472 [6:18:20<1:03:07,  3.64it/s]
 96%|█████████▋| 357691/371472 [6:18:20<1:02:05,  3.70it/s]
 96%|█████████▋| 357692/371472 [6:18:20<1:00:45,  3.78it/s]
 96%|█████████▋| 357693/371472 [6:18:20<1:01:11,  3.75it/s]
 96%|█████████▋| 357694/371472 [6:18:21<1:01:42,  3.72it/s]
 96%|█████████▋| 357695/371472 [6:18:21<1:00:32,  3.79it/s]
 96%|█████████▋| 357696/371472 [6:18:21<1:02:11,  3.69it/s]
 96%|█████████▋| 357697/371472 [6:18:22<1:01:58,  3.70it/s]
 96%|█████████▋| 357698/371472 [6:18:22<1:01:16,  3.75it/s]
 96%|█████████▋| 357699/371472 [6:18:22<1:02:28,  3.67it/s]
 96%|█████████▋| 357700/371472 [6:18:22<1:02:37,  3.67it/s]
                                                           
{'loss': 2.6161, 'learning_rate': 1.3338468831476653e-07, 'epoch': 15.41}
+
 96%|█████████▋| 357700/371472 [6:18:22<1:02:37,  3.67it/s]
 96%|█████████▋| 357701/371472 [6:18:23<1:01:08,  3.75it/s]
 96%|█████████▋| 357702/371472 [6:18:23<1:01:39,  3.72it/s]
 96%|█████████▋| 357703/371472 [6:18:23<1:01:58,  3.70it/s]
 96%|█████████▋| 357704/371472 [6:18:23<1:00:37,  3.79it/s]
 96%|█████████▋| 357705/371472 [6:18:24<59:17,  3.87it/s]  
 96%|█████████▋| 357706/371472 [6:18:24<1:01:26,  3.73it/s]
 96%|█████████▋| 357707/371472 [6:18:24<1:00:11,  3.81it/s]
 96%|█████████▋| 357708/371472 [6:18:25<1:05:48,  3.49it/s]
 96%|█████████▋| 357709/371472 [6:18:25<1:04:42,  3.55it/s]
 96%|█████████▋| 357710/371472 [6:18:25<1:04:37,  3.55it/s]
 96%|█████████▋| 357711/371472 [6:18:25<1:03:16,  3.62it/s]
 96%|█████████▋| 357712/371472 [6:18:26<1:02:58,  3.64it/s]
 96%|█████████▋| 357713/371472 [6:18:26<1:00:53,  3.77it/s]
 96%|█████████▋| 357714/371472 [6:18:26<1:02:08,  3.69it/s]
 96%|█████████▋| 357715/371472 [6:18:27<1:12:06,  3.18it/s]
 96%|█████████▋| 357716/371472 [6:18:27<1:14:23,  3.08it/s]
 96%|█████████▋| 357717/371472 [6:18:27<1:09:49,  3.28it/s]
 96%|█████████▋| 357718/371472 [6:18:27<1:06:28,  3.45it/s]
 96%|█████████▋| 357719/371472 [6:18:28<1:05:58,  3.47it/s]
 96%|█████████▋| 357720/371472 [6:18:28<1:03:48,  3.59it/s]
                                                           
{'loss': 2.4774, 'learning_rate': 1.3333620633928763e-07, 'epoch': 15.41}
+
 96%|█████████▋| 357720/371472 [6:18:28<1:03:48,  3.59it/s]
 96%|█████████▋| 357721/371472 [6:18:28<1:09:25,  3.30it/s]
 96%|█████████▋| 357722/371472 [6:18:29<1:18:50,  2.91it/s]
 96%|█████████▋| 357723/371472 [6:18:29<1:11:37,  3.20it/s]
 96%|█████████▋| 357724/371472 [6:18:29<1:08:56,  3.32it/s]
 96%|█████████▋| 357725/371472 [6:18:30<1:07:22,  3.40it/s]
 96%|█████████▋| 357726/371472 [6:18:30<1:04:59,  3.53it/s]
 96%|█████████▋| 357727/371472 [6:18:30<1:02:08,  3.69it/s]
 96%|█████████▋| 357728/371472 [6:18:30<1:01:54,  3.70it/s]
 96%|█████████▋| 357729/371472 [6:18:31<1:07:47,  3.38it/s]
 96%|█████████▋| 357730/371472 [6:18:31<1:14:27,  3.08it/s]
 96%|█████████▋| 357731/371472 [6:18:31<1:15:59,  3.01it/s]
 96%|█████████▋| 357732/371472 [6:18:32<1:12:09,  3.17it/s]
 96%|█████████▋| 357733/371472 [6:18:32<1:08:41,  3.33it/s]
 96%|█████████▋| 357734/371472 [6:18:32<1:08:45,  3.33it/s]
 96%|█████████▋| 357735/371472 [6:18:33<1:07:10,  3.41it/s]
 96%|█████████▋| 357736/371472 [6:18:33<1:10:33,  3.24it/s]
 96%|█████████▋| 357737/371472 [6:18:33<1:07:38,  3.38it/s]
 96%|█████████▋| 357738/371472 [6:18:33<1:06:54,  3.42it/s]
 96%|█████████▋| 357739/371472 [6:18:34<1:11:55,  3.18it/s]
 96%|█████████▋| 357740/371472 [6:18:34<1:08:20,  3.35it/s]
                                                           
{'loss': 2.5197, 'learning_rate': 1.3328772436380878e-07, 'epoch': 15.41}
+
 96%|█████████▋| 357740/371472 [6:18:34<1:08:20,  3.35it/s]
 96%|█████████▋| 357741/371472 [6:18:34<1:04:24,  3.55it/s]
 96%|█████████▋| 357742/371472 [6:18:35<1:03:51,  3.58it/s]
 96%|█████████▋| 357743/371472 [6:18:35<1:05:12,  3.51it/s]
 96%|█████████▋| 357744/371472 [6:18:35<1:06:21,  3.45it/s]
 96%|█████████▋| 357745/371472 [6:18:35<1:07:05,  3.41it/s]
 96%|█████████▋| 357746/371472 [6:18:36<1:07:45,  3.38it/s]
 96%|█████████▋| 357747/371472 [6:18:36<1:04:37,  3.54it/s]
 96%|█████████▋| 357748/371472 [6:18:36<1:03:54,  3.58it/s]
 96%|█████████▋| 357749/371472 [6:18:37<1:02:40,  3.65it/s]
 96%|█████████▋| 357750/371472 [6:18:37<1:01:18,  3.73it/s]
 96%|█████████▋| 357751/371472 [6:18:37<1:03:12,  3.62it/s]
 96%|█████████▋| 357752/371472 [6:18:37<1:02:31,  3.66it/s]
 96%|█████████▋| 357753/371472 [6:18:38<1:05:56,  3.47it/s]
 96%|█████████▋| 357754/371472 [6:18:38<1:03:15,  3.61it/s]
 96%|█████████▋| 357755/371472 [6:18:38<1:02:48,  3.64it/s]
 96%|█████████▋| 357756/371472 [6:18:38<1:02:10,  3.68it/s]
 96%|█████████▋| 357757/371472 [6:18:39<1:02:31,  3.66it/s]
 96%|█████████▋| 357758/371472 [6:18:39<1:10:33,  3.24it/s]
 96%|█████████▋| 357759/371472 [6:18:39<1:08:59,  3.31it/s]
 96%|█████████▋| 357760/371472 [6:18:40<1:11:02,  3.22it/s]
                                                           
{'loss': 2.6126, 'learning_rate': 1.3323924238832982e-07, 'epoch': 15.41}
+
 96%|█████████▋| 357760/371472 [6:18:40<1:11:02,  3.22it/s]
 96%|█████████▋| 357761/371472 [6:18:40<1:07:24,  3.39it/s]
 96%|█████████▋| 357762/371472 [6:18:40<1:05:08,  3.51it/s]
 96%|█████████▋| 357763/371472 [6:18:41<1:06:25,  3.44it/s]
 96%|█████████▋| 357764/371472 [6:18:41<1:05:23,  3.49it/s]
 96%|█████████▋| 357765/371472 [6:18:41<1:01:49,  3.70it/s]
 96%|█████████▋| 357766/371472 [6:18:41<1:02:58,  3.63it/s]
 96%|█████████▋| 357767/371472 [6:18:42<1:02:27,  3.66it/s]
 96%|█████████▋| 357768/371472 [6:18:42<1:05:41,  3.48it/s]
 96%|█████████▋| 357769/371472 [6:18:42<1:10:50,  3.22it/s]
 96%|█████████▋| 357770/371472 [6:18:43<1:07:45,  3.37it/s]
 96%|█████████▋| 357771/371472 [6:18:43<1:06:56,  3.41it/s]
 96%|█████████▋| 357772/371472 [6:18:43<1:04:50,  3.52it/s]
 96%|█████████▋| 357773/371472 [6:18:43<1:03:14,  3.61it/s]
 96%|█████████▋| 357774/371472 [6:18:44<1:04:44,  3.53it/s]
 96%|█████████▋| 357775/371472 [6:18:44<1:05:06,  3.51it/s]
 96%|█████████▋| 357776/371472 [6:18:44<1:01:43,  3.70it/s]
 96%|█████████▋| 357777/371472 [6:18:45<1:01:21,  3.72it/s]
 96%|█████████▋| 357778/371472 [6:18:45<1:00:24,  3.78it/s]
 96%|█████████▋| 357779/371472 [6:18:45<1:00:37,  3.76it/s]
 96%|█████████▋| 357780/371472 [6:18:45<1:01:24,  3.72it/s]
                                                           
{'loss': 2.5587, 'learning_rate': 1.33190760412851e-07, 'epoch': 15.41}
+
 96%|█████████▋| 357780/371472 [6:18:45<1:01:24,  3.72it/s]
 96%|█████████▋| 357781/371472 [6:18:46<1:00:46,  3.75it/s]
 96%|█████████▋| 357782/371472 [6:18:46<59:48,  3.82it/s]  
 96%|█████████▋| 357783/371472 [6:18:46<1:05:03,  3.51it/s]
 96%|█████████▋| 357784/371472 [6:18:46<1:03:33,  3.59it/s]
 96%|█████████▋| 357785/371472 [6:18:47<1:01:38,  3.70it/s]
 96%|█████████▋| 357786/371472 [6:18:47<1:03:02,  3.62it/s]
 96%|█████████▋| 357787/371472 [6:18:47<1:08:13,  3.34it/s]
 96%|█████████▋| 357788/371472 [6:18:48<1:05:04,  3.50it/s]
 96%|█████████▋| 357789/371472 [6:18:48<1:02:37,  3.64it/s]
 96%|█████████▋| 357790/371472 [6:18:48<1:02:52,  3.63it/s]
 96%|█████████▋| 357791/371472 [6:18:48<1:03:46,  3.58it/s]
 96%|█████████▋| 357792/371472 [6:18:49<1:02:14,  3.66it/s]
 96%|█████████▋| 357793/371472 [6:18:49<1:00:25,  3.77it/s]
 96%|█████████▋| 357794/371472 [6:18:49<1:04:16,  3.55it/s]
 96%|█████████▋| 357795/371472 [6:18:49<1:04:13,  3.55it/s]
 96%|█████████▋| 357796/371472 [6:18:50<1:05:49,  3.46it/s]
 96%|█████████▋| 357797/371472 [6:18:50<1:03:41,  3.58it/s]
 96%|█████████▋| 357798/371472 [6:18:50<1:01:15,  3.72it/s]
 96%|█████████▋| 357799/371472 [6:18:51<1:00:57,  3.74it/s]
 96%|█████████▋| 357800/371472 [6:18:51<1:03:23,  3.60it/s]
                                                           
{'loss': 2.7785, 'learning_rate': 1.3314227843737204e-07, 'epoch': 15.41}
+
 96%|█████████▋| 357800/371472 [6:18:51<1:03:23,  3.60it/s]
 96%|█████████▋| 357801/371472 [6:18:51<1:09:34,  3.28it/s]
 96%|█████████▋| 357802/371472 [6:18:51<1:05:25,  3.48it/s]
 96%|█████████▋| 357803/371472 [6:18:52<1:02:10,  3.66it/s]
 96%|█████████▋| 357804/371472 [6:18:52<1:02:55,  3.62it/s]
 96%|█████████▋| 357805/371472 [6:18:52<1:02:28,  3.65it/s]
 96%|█████████▋| 357806/371472 [6:18:53<1:05:25,  3.48it/s]
 96%|█████████▋| 357807/371472 [6:18:53<1:06:08,  3.44it/s]
 96%|█████████▋| 357808/371472 [6:18:53<1:05:57,  3.45it/s]
 96%|█████████▋| 357809/371472 [6:18:53<1:04:18,  3.54it/s]
 96%|█████████▋| 357810/371472 [6:18:54<1:04:00,  3.56it/s]
 96%|█████████▋| 357811/371472 [6:18:54<1:02:31,  3.64it/s]
 96%|█████████▋| 357812/371472 [6:18:54<1:03:33,  3.58it/s]
 96%|█████████▋| 357813/371472 [6:18:55<1:01:02,  3.73it/s]
 96%|█████████▋| 357814/371472 [6:18:55<1:01:38,  3.69it/s]
 96%|█████████▋| 357815/371472 [6:18:55<1:00:43,  3.75it/s]
 96%|█████████▋| 357816/371472 [6:18:55<1:01:35,  3.69it/s]
 96%|█████████▋| 357817/371472 [6:18:56<1:00:47,  3.74it/s]
 96%|█████████▋| 357818/371472 [6:18:56<1:00:06,  3.79it/s]
 96%|█████████▋| 357819/371472 [6:18:56<1:03:17,  3.60it/s]
 96%|█████████▋| 357820/371472 [6:18:56<1:02:05,  3.66it/s]
                                                           
{'loss': 2.7721, 'learning_rate': 1.330937964618932e-07, 'epoch': 15.41}
+
 96%|█████████▋| 357820/371472 [6:18:56<1:02:05,  3.66it/s]
 96%|█████████▋| 357821/371472 [6:18:57<1:11:36,  3.18it/s]
 96%|█████████▋| 357822/371472 [6:18:57<1:09:27,  3.28it/s]
 96%|█████████▋| 357823/371472 [6:18:57<1:06:30,  3.42it/s]
 96%|█████████▋| 357824/371472 [6:18:58<1:07:38,  3.36it/s]
 96%|█████████▋| 357825/371472 [6:18:58<1:08:57,  3.30it/s]
 96%|█████████▋| 357826/371472 [6:18:58<1:05:21,  3.48it/s]
 96%|█████████▋| 357827/371472 [6:18:59<1:03:24,  3.59it/s]
 96%|█████████▋| 357828/371472 [6:18:59<1:02:46,  3.62it/s]
 96%|█████████▋| 357829/371472 [6:18:59<1:00:44,  3.74it/s]
 96%|█████████▋| 357830/371472 [6:18:59<1:05:13,  3.49it/s]
 96%|█████████▋| 357831/371472 [6:19:00<1:04:56,  3.50it/s]
 96%|█████████▋| 357832/371472 [6:19:00<1:04:36,  3.52it/s]
 96%|█████████▋| 357833/371472 [6:19:00<1:02:48,  3.62it/s]
 96%|█████████▋| 357834/371472 [6:19:00<1:05:20,  3.48it/s]
 96%|█████████▋| 357835/371472 [6:19:01<1:02:07,  3.66it/s]
 96%|█████████▋| 357836/371472 [6:19:01<59:09,  3.84it/s]  
 96%|█████████▋| 357837/371472 [6:19:01<58:41,  3.87it/s]
 96%|█████████▋| 357838/371472 [6:19:02<1:05:32,  3.47it/s]
 96%|█████████▋| 357839/371472 [6:19:02<1:03:40,  3.57it/s]
 96%|█████████▋| 357840/371472 [6:19:02<1:02:53,  3.61it/s]
                                                           
{'loss': 2.545, 'learning_rate': 1.3304531448641427e-07, 'epoch': 15.41}
+
 96%|█████████▋| 357840/371472 [6:19:02<1:02:53,  3.61it/s]
 96%|█████████▋| 357841/371472 [6:19:02<1:00:28,  3.76it/s]
 96%|█████████▋| 357842/371472 [6:19:03<1:00:53,  3.73it/s]
 96%|█████████▋| 357843/371472 [6:19:03<58:46,  3.86it/s]  
 96%|█████████▋| 357844/371472 [6:19:03<59:14,  3.83it/s]
 96%|█████████▋| 357845/371472 [6:19:03<59:04,  3.84it/s]
 96%|█████████▋| 357846/371472 [6:19:04<59:59,  3.79it/s]
 96%|█████████▋| 357847/371472 [6:19:04<1:03:21,  3.58it/s]
 96%|█████��███▋| 357848/371472 [6:19:04<1:03:51,  3.56it/s]
 96%|█████████▋| 357849/371472 [6:19:04<1:00:56,  3.73it/s]
 96%|█████████▋| 357850/371472 [6:19:05<1:08:28,  3.32it/s]
 96%|█████████▋| 357851/371472 [6:19:05<1:07:43,  3.35it/s]
 96%|█████████▋| 357852/371472 [6:19:05<1:06:14,  3.43it/s]
 96%|█████████▋| 357853/371472 [6:19:06<1:11:05,  3.19it/s]
 96%|█████████▋| 357854/371472 [6:19:06<1:09:31,  3.26it/s]
 96%|█████████▋| 357855/371472 [6:19:06<1:10:53,  3.20it/s]
 96%|█████████▋| 357856/371472 [6:19:07<1:07:16,  3.37it/s]
 96%|█████████▋| 357857/371472 [6:19:07<1:08:34,  3.31it/s]
 96%|█████████▋| 357858/371472 [6:19:07<1:05:34,  3.46it/s]
 96%|█████████▋| 357859/371472 [6:19:08<1:04:23,  3.52it/s]
 96%|█████████▋| 357860/371472 [6:19:08<1:04:48,  3.50it/s]
                                                           
{'loss': 2.7441, 'learning_rate': 1.3299683251093542e-07, 'epoch': 15.41}
+
 96%|█████████▋| 357860/371472 [6:19:08<1:04:48,  3.50it/s]
 96%|█████████▋| 357861/371472 [6:19:08<1:03:47,  3.56it/s]
 96%|█████████▋| 357862/371472 [6:19:08<1:01:31,  3.69it/s]
 96%|█████████▋| 357863/371472 [6:19:09<1:03:55,  3.55it/s]
 96%|█████████▋| 357864/371472 [6:19:09<1:04:27,  3.52it/s]
 96%|█████████▋| 357865/371472 [6:19:09<1:02:28,  3.63it/s]
 96%|█████████▋| 357866/371472 [6:19:09<1:00:27,  3.75it/s]
 96%|█████████▋| 357867/371472 [6:19:10<1:01:09,  3.71it/s]
 96%|█████████▋| 357868/371472 [6:19:10<1:01:11,  3.71it/s]
 96%|█████████▋| 357869/371472 [6:19:10<1:01:51,  3.67it/s]
 96%|█████████▋| 357870/371472 [6:19:11<1:03:47,  3.55it/s]
 96%|█████████▋| 357871/371472 [6:19:11<1:01:51,  3.66it/s]
 96%|█████████▋| 357872/371472 [6:19:11<1:01:53,  3.66it/s]
 96%|█████████▋| 357873/371472 [6:19:11<1:04:25,  3.52it/s]
 96%|█████████▋| 357874/371472 [6:19:12<1:01:39,  3.68it/s]
 96%|█████████▋| 357875/371472 [6:19:12<1:01:49,  3.67it/s]
 96%|█████████▋| 357876/371472 [6:19:12<59:08,  3.83it/s]  
 96%|█████████▋| 357877/371472 [6:19:12<57:44,  3.92it/s]
 96%|█████████▋| 357878/371472 [6:19:13<58:39,  3.86it/s]
 96%|█████████▋| 357879/371472 [6:19:13<57:49,  3.92it/s]
 96%|█████████▋| 357880/371472 [6:19:13<59:51,  3.78it/s]
                                                         
{'loss': 2.5583, 'learning_rate': 1.3294835053545646e-07, 'epoch': 15.41}
+
 96%|█████████▋| 357880/371472 [6:19:13<59:51,  3.78it/s]
 96%|█████████▋| 357881/371472 [6:19:13<1:00:46,  3.73it/s]
 96%|█████████▋| 357882/371472 [6:19:14<1:03:03,  3.59it/s]
 96%|█████████▋| 357883/371472 [6:19:14<1:01:21,  3.69it/s]
 96%|█████████▋| 357884/371472 [6:19:14<1:00:44,  3.73it/s]
 96%|█████████▋| 357885/371472 [6:19:15<1:02:12,  3.64it/s]
 96%|█████████▋| 357886/371472 [6:19:15<1:01:56,  3.66it/s]
 96%|█████████▋| 357887/371472 [6:19:15<1:01:53,  3.66it/s]
 96%|█████████▋| 357888/371472 [6:19:15<1:01:38,  3.67it/s]
 96%|█████████▋| 357889/371472 [6:19:16<1:00:48,  3.72it/s]
 96%|█████████▋| 357890/371472 [6:19:16<1:01:35,  3.67it/s]
 96%|█████████▋| 357891/371472 [6:19:16<1:07:10,  3.37it/s]
 96%|█████████▋| 357892/371472 [6:19:17<1:05:41,  3.45it/s]
 96%|█████████▋| 357893/371472 [6:19:17<1:04:47,  3.49it/s]
 96%|█████████▋| 357894/371472 [6:19:17<1:04:47,  3.49it/s]
 96%|█████████▋| 357895/371472 [6:19:17<1:09:13,  3.27it/s]
 96%|█████████▋| 357896/371472 [6:19:18<1:06:26,  3.41it/s]
 96%|█████████▋| 357897/371472 [6:19:18<1:04:50,  3.49it/s]
 96%|█████████▋| 357898/371472 [6:19:18<1:07:11,  3.37it/s]
 96%|█████████▋| 357899/371472 [6:19:19<1:09:10,  3.27it/s]
 96%|█████████▋| 357900/371472 [6:19:19<1:06:21,  3.41it/s]
                                                           
{'loss': 2.5272, 'learning_rate': 1.3289986855997753e-07, 'epoch': 15.42}
+
 96%|█████████▋| 357900/371472 [6:19:19<1:06:21,  3.41it/s]
 96%|█████████▋| 357901/371472 [6:19:19<1:05:44,  3.44it/s]
 96%|█████████▋| 357902/371472 [6:19:19<1:02:11,  3.64it/s]
 96%|█████████▋| 357903/371472 [6:19:20<1:03:44,  3.55it/s]
 96%|█████████▋| 357904/371472 [6:19:20<1:02:23,  3.62it/s]
 96%|█████████▋| 357905/371472 [6:19:20<1:03:24,  3.57it/s]
 96%|█████████▋| 357906/371472 [6:19:21<1:02:29,  3.62it/s]
 96%|█████████▋| 357907/371472 [6:19:21<1:01:46,  3.66it/s]
 96%|█████████▋| 357908/371472 [6:19:21<1:04:02,  3.53it/s]
 96%|█████████▋| 357909/371472 [6:19:21<1:01:35,  3.67it/s]
 96%|█████████▋| 357910/371472 [6:19:22<1:00:38,  3.73it/s]
 96%|█████████▋| 357911/371472 [6:19:22<1:01:06,  3.70it/s]
 96%|█████████▋| 357912/371472 [6:19:22<1:01:00,  3.70it/s]
 96%|█████████▋| 357913/371472 [6:19:22<1:01:52,  3.65it/s]
 96%|█████████▋| 357914/371472 [6:19:23<1:03:31,  3.56it/s]
 96%|█████████▋| 357915/371472 [6:19:23<1:16:18,  2.96it/s]
 96%|█████████▋| 357916/371472 [6:19:24<1:11:23,  3.16it/s]
 96%|█████████▋| 357917/371472 [6:19:24<1:07:40,  3.34it/s]
 96%|█████████▋| 357918/371472 [6:19:24<1:04:11,  3.52it/s]
 96%|█████████▋| 357919/371472 [6:19:24<1:04:40,  3.49it/s]
 96%|█████████▋| 357920/371472 [6:19:25<1:09:05,  3.27it/s]
                                                           
{'loss': 2.5949, 'learning_rate': 1.3285138658449869e-07, 'epoch': 15.42}
+
 96%|█████████▋| 357920/371472 [6:19:25<1:09:05,  3.27it/s]
 96%|█████████▋| 357921/371472 [6:19:25<1:06:25,  3.40it/s]
 96%|█████████▋| 357922/371472 [6:19:25<1:02:49,  3.59it/s]
 96%|█████████▋| 357923/371472 [6:19:25<1:03:06,  3.58it/s]
 96%|█████████▋| 357924/371472 [6:19:26<1:06:34,  3.39it/s]
 96%|█████████▋| 357925/371472 [6:19:26<1:12:22,  3.12it/s]
 96%|█████████▋| 357926/371472 [6:19:26<1:11:48,  3.14it/s]
 96%|█████████▋| 357927/371472 [6:19:27<1:18:25,  2.88it/s]
 96%|█████████▋| 357928/371472 [6:19:27<1:13:00,  3.09it/s]
 96%|█████████▋| 357929/371472 [6:19:27<1:11:04,  3.18it/s]
 96%|█████████▋| 357930/371472 [6:19:28<1:07:29,  3.34it/s]
 96%|█████████▋| 357931/371472 [6:19:28<1:07:24,  3.35it/s]
 96%|█████████▋| 357932/371472 [6:19:28<1:05:44,  3.43it/s]
 96%|█████████▋| 357933/371472 [6:19:29<1:04:47,  3.48it/s]
 96%|█████████▋| 357934/371472 [6:19:29<1:02:44,  3.60it/s]
 96%|█████████▋| 357935/371472 [6:19:29<1:04:32,  3.50it/s]
 96%|█████████▋| 357936/371472 [6:19:29<1:00:41,  3.72it/s]
 96%|█████████▋| 357937/371472 [6:19:30<1:01:29,  3.67it/s]
 96%|█████████▋| 357938/371472 [6:19:30<58:18,  3.87it/s]  
 96%|█████████▋| 357939/371472 [6:19:30<58:04,  3.88it/s]
 96%|█████████▋| 357940/371472 [6:19:30<1:04:06,  3.52it/s]
                                                           
{'loss': 2.5335, 'learning_rate': 1.3280290460901976e-07, 'epoch': 15.42}
+
 96%|█████████▋| 357940/371472 [6:19:30<1:04:06,  3.52it/s]
 96%|█████████▋| 357941/371472 [6:19:31<1:01:49,  3.65it/s]
 96%|█████████▋| 357942/371472 [6:19:31<1:00:36,  3.72it/s]
 96%|█████████▋| 357943/371472 [6:19:31<1:00:51,  3.71it/s]
 96%|█████████▋| 357944/371472 [6:19:32<1:00:41,  3.72it/s]
 96%|█████████▋| 357945/371472 [6:19:32<59:58,  3.76it/s]  
 96%|█████████▋| 357946/371472 [6:19:32<59:28,  3.79it/s]
 96%|█████████▋| 357947/371472 [6:19:32<58:53,  3.83it/s]
 96%|█████████▋| 357948/371472 [6:19:33<59:35,  3.78it/s]
 96%|█████████▋| 357949/371472 [6:19:33<1:03:58,  3.52it/s]
 96%|█████████▋| 357950/371472 [6:19:33<1:01:43,  3.65it/s]
 96%|█████████▋| 357951/371472 [6:19:33<1:01:44,  3.65it/s]
 96%|█████████▋| 357952/371472 [6:19:34<1:02:58,  3.58it/s]
 96%|█████████▋| 357953/371472 [6:19:34<1:05:58,  3.42it/s]
 96%|█████████▋| 357954/371472 [6:19:34<1:11:06,  3.17it/s]
 96%|█████████▋| 357955/371472 [6:19:35<1:06:12,  3.40it/s]
 96%|█████████▋| 357956/371472 [6:19:35<1:07:05,  3.36it/s]
 96%|█████████▋| 357957/371472 [6:19:35<1:06:25,  3.39it/s]
 96%|█████████▋| 357958/371472 [6:19:35<1:04:13,  3.51it/s]
 96%|█████████▋| 357959/371472 [6:19:36<1:05:12,  3.45it/s]
 96%|█████████▋| 357960/371472 [6:19:36<1:08:48,  3.27it/s]
                                                           
{'loss': 2.5555, 'learning_rate': 1.327544226335409e-07, 'epoch': 15.42}
+
 96%|███████���█▋| 357960/371472 [6:19:36<1:08:48,  3.27it/s]
 96%|█████████▋| 357961/371472 [6:19:36<1:05:54,  3.42it/s]
 96%|█████████▋| 357962/371472 [6:19:37<1:09:47,  3.23it/s]
 96%|█████████▋| 357963/371472 [6:19:37<1:07:02,  3.36it/s]
 96%|█████████▋| 357964/371472 [6:19:37<1:05:23,  3.44it/s]
 96%|█████████▋| 357965/371472 [6:19:38<1:03:54,  3.52it/s]
 96%|█████████▋| 357966/371472 [6:19:38<1:02:17,  3.61it/s]
 96%|█████████▋| 357967/371472 [6:19:38<1:01:59,  3.63it/s]
 96%|█████████▋| 357968/371472 [6:19:38<1:00:46,  3.70it/s]
 96%|█████████▋| 357969/371472 [6:19:39<1:04:41,  3.48it/s]
 96%|█████████▋| 357970/371472 [6:19:39<1:03:38,  3.54it/s]
 96%|█████████▋| 357971/371472 [6:19:39<1:02:57,  3.57it/s]
 96%|█████████▋| 357972/371472 [6:19:39<1:02:07,  3.62it/s]
 96%|█████████▋| 357973/371472 [6:19:40<1:02:14,  3.61it/s]
 96%|█████████▋| 357974/371472 [6:19:40<1:01:48,  3.64it/s]
 96%|█████████▋| 357975/371472 [6:19:40<1:00:58,  3.69it/s]
 96%|█████████▋| 357976/371472 [6:19:41<1:00:58,  3.69it/s]
 96%|█████████▋| 357977/371472 [6:19:41<1:00:11,  3.74it/s]
 96%|█████████▋| 357978/371472 [6:19:41<1:03:01,  3.57it/s]
 96%|█████████▋| 357979/371472 [6:19:41<1:03:26,  3.54it/s]
 96%|█████████▋| 357980/371472 [6:19:42<1:02:40,  3.59it/s]
                                                           
{'loss': 2.5573, 'learning_rate': 1.3270594065806198e-07, 'epoch': 15.42}
+
 96%|█████████▋| 357980/371472 [6:19:42<1:02:40,  3.59it/s]
 96%|█████████▋| 357981/371472 [6:19:42<1:01:22,  3.66it/s]
 96%|█████████▋| 357982/371472 [6:19:42<1:04:03,  3.51it/s]
 96%|█████████▋| 357983/371472 [6:19:43<1:08:47,  3.27it/s]
 96%|█████████▋| 357984/371472 [6:19:43<1:05:38,  3.42it/s]
 96%|█████████▋| 357985/371472 [6:19:43<1:03:41,  3.53it/s]
 96%|█████████▋| 357986/371472 [6:19:43<1:02:06,  3.62it/s]
 96%|█████████▋| 357987/371472 [6:19:44<1:06:08,  3.40it/s]
 96%|█████████▋| 357988/371472 [6:19:44<1:06:58,  3.36it/s]
 96%|█████████▋| 357989/371472 [6:19:44<1:09:47,  3.22it/s]
 96%|█████████▋| 357990/371472 [6:19:45<1:04:16,  3.50it/s]
 96%|█████████▋| 357991/371472 [6:19:45<1:03:31,  3.54it/s]
 96%|█████████▋| 357992/371472 [6:19:45<1:01:38,  3.64it/s]
 96%|█████████▋| 357993/371472 [6:19:45<59:43,  3.76it/s]  
 96%|█████████▋| 357994/371472 [6:19:46<1:01:46,  3.64it/s]
 96%|█████████▋| 357995/371472 [6:19:46<1:00:38,  3.70it/s]
 96%|█████████▋| 357996/371472 [6:19:46<1:02:56,  3.57it/s]
 96%|█████████▋| 357997/371472 [6:19:47<1:01:42,  3.64it/s]
 96%|█████████▋| 357998/371472 [6:19:47<1:04:45,  3.47it/s]
 96%|█████████▋| 357999/371472 [6:19:47<1:05:20,  3.44it/s]
 96%|█████████▋| 358000/371472 [6:19:47<1:03:28,  3.54it/s]
                                                           
{'loss': 2.3475, 'learning_rate': 1.3265745868258313e-07, 'epoch': 15.42}
+
 96%|█████████▋| 358000/371472 [6:19:47<1:03:28,  3.54it/s]
 96%|█████████▋| 358001/371472 [6:19:48<1:06:44,  3.36it/s]
 96%|█████████▋| 358002/371472 [6:19:48<1:07:19,  3.33it/s]
 96%|█████████▋| 358003/371472 [6:19:48<1:04:50,  3.46it/s]
 96%|█████████▋| 358004/371472 [6:19:49<1:03:24,  3.54it/s]
 96%|█████████▋| 358005/371472 [6:19:49<1:02:24,  3.60it/s]
 96%|█████████▋| 358006/371472 [6:19:49<1:03:51,  3.51it/s]
 96%|█████████▋| 358007/371472 [6:19:49<1:02:17,  3.60it/s]
 96%|█████████▋| 358008/371472 [6:19:50<1:01:54,  3.62it/s]
 96%|█████████▋| 358009/371472 [6:19:50<1:01:10,  3.67it/s]
 96%|█████████▋| 358010/371472 [6:19:50<1:02:42,  3.58it/s]
 96%|█████████▋| 358011/371472 [6:19:51<1:03:31,  3.53it/s]
 96%|█████████▋| 358012/371472 [6:19:51<1:05:31,  3.42it/s]
 96%|█████████▋| 358013/371472 [6:19:51<1:02:59,  3.56it/s]
 96%|█████████▋| 358014/371472 [6:19:51<1:01:17,  3.66it/s]
 96%|█████████▋| 358015/371472 [6:19:52<1:00:49,  3.69it/s]
 96%|█████████▋| 358016/371472 [6:19:52<59:44,  3.75it/s]  
 96%|█████████▋| 358017/371472 [6:19:52<1:01:10,  3.67it/s]
 96%|█████████▋| 358018/371472 [6:19:52<1:00:40,  3.70it/s]
 96%|█████████▋| 358019/371472 [6:19:53<1:00:17,  3.72it/s]
 96%|█████████▋| 358020/371472 [6:19:53<58:47,  3.81it/s]  
                                                         
{'loss': 2.6319, 'learning_rate': 1.3260897670710418e-07, 'epoch': 15.42}
+
 96%|█████████▋| 358020/371472 [6:19:53<58:47,  3.81it/s]
 96%|█████████▋| 358021/371472 [6:19:53<58:41,  3.82it/s]
 96%|█████████▋| 358022/371472 [6:19:54<1:04:03,  3.50it/s]
 96%|█████████▋| 358023/371472 [6:19:54<1:19:28,  2.82it/s]
 96%|█████████▋| 358024/371472 [6:19:54<1:13:56,  3.03it/s]
 96%|█████████▋| 358025/371472 [6:19:55<1:13:22,  3.05it/s]
 96%|█████████▋| 358026/371472 [6:19:55<1:11:53,  3.12it/s]
 96%|█████████▋| 358027/371472 [6:19:55<1:18:09,  2.87it/s]
 96%|█████████▋| 358028/371472 [6:19:56<1:13:12,  3.06it/s]
 96%|█████████▋| 358029/371472 [6:19:56<1:09:16,  3.23it/s]
 96%|█████████▋| 358030/371472 [6:19:56<1:11:52,  3.12it/s]
 96%|█████████▋| 358031/371472 [6:19:57<1:07:35,  3.31it/s]
 96%|█████████▋| 358032/371472 [6:19:57<1:05:09,  3.44it/s]
 96%|█████████▋| 358033/371472 [6:19:57<1:07:19,  3.33it/s]
 96%|█████████▋| 358034/371472 [6:19:57<1:08:16,  3.28it/s]
 96%|█████████▋| 358035/371472 [6:19:58<1:05:08,  3.44it/s]
 96%|█████████▋| 358036/371472 [6:19:58<1:07:07,  3.34it/s]
 96%|█████████▋| 358037/371472 [6:19:58<1:14:28,  3.01it/s]
 96%|█████████▋| 358038/371472 [6:19:59<1:11:11,  3.15it/s]
 96%|█████████▋| 358039/371472 [6:19:59<1:09:01,  3.24it/s]
 96%|█████████▋| 358040/371472 [6:19:59<1:06:08,  3.38it/s]
                                                           
{'loss': 2.467, 'learning_rate': 1.3256049473162538e-07, 'epoch': 15.42}
+
 96%|█████████▋| 358040/371472 [6:19:59<1:06:08,  3.38it/s]
 96%|█████████▋| 358041/371472 [6:19:59<1:03:54,  3.50it/s]
 96%|█████████▋| 358042/371472 [6:20:00<1:07:09,  3.33it/s]
 96%|█████████▋| 358043/371472 [6:20:00<1:04:28,  3.47it/s]
 96%|█████████▋| 358044/371472 [6:20:00<1:08:34,  3.26it/s]
 96%|█████████▋| 358045/371472 [6:20:01<1:03:35,  3.52it/s]
 96%|█████████▋| 358046/371472 [6:20:01<1:02:30,  3.58it/s]
 96%|█████████▋| 358047/371472 [6:20:01<1:03:17,  3.54it/s]
 96%|█████████▋| 358048/371472 [6:20:01<1:00:30,  3.70it/s]
 96%|█████████▋| 358049/371472 [6:20:02<1:00:57,  3.67it/s]
 96%|█████████▋| 358050/371472 [6:20:02<59:25,  3.76it/s]  
 96%|█████████▋| 358051/371472 [6:20:02<59:25,  3.76it/s]
 96%|█████████▋| 358052/371472 [6:20:03<58:29,  3.82it/s]
 96%|█████████▋| 358053/371472 [6:20:03<56:58,  3.92it/s]
 96%|█████████▋| 358054/371472 [6:20:03<57:49,  3.87it/s]
 96%|█████████▋| 358055/371472 [6:20:03<1:00:43,  3.68it/s]
 96%|█████████▋| 358056/371472 [6:20:04<59:23,  3.76it/s]  
 96%|█████████▋| 358057/371472 [6:20:04<59:15,  3.77it/s]
 96%|█████████▋| 358058/371472 [6:20:04<1:00:06,  3.72it/s]
 96%|█████████▋| 358059/371472 [6:20:04<1:03:40,  3.51it/s]
 96%|█████████▋| 358060/371472 [6:20:05<1:03:30,  3.52it/s]
                                                           
{'loss': 2.4972, 'learning_rate': 1.325120127561464e-07, 'epoch': 15.42}
+
 96%|█████████▋| 358060/371472 [6:20:05<1:03:30,  3.52it/s]
 96%|█████████▋| 358061/371472 [6:20:05<1:01:46,  3.62it/s]
 96%|█████████▋| 358062/371472 [6:20:05<59:35,  3.75it/s]  
 96%|█████████▋| 358063/371472 [6:20:05<59:56,  3.73it/s]
 96%|█████████▋| 358064/371472 [6:20:06<1:00:36,  3.69it/s]
 96%|█████████▋| 358065/371472 [6:20:06<1:02:35,  3.57it/s]
 96%|█████████▋| 358066/371472 [6:20:06<1:00:21,  3.70it/s]
 96%|█████████▋| 358067/371472 [6:20:07<1:00:46,  3.68it/s]
 96%|█████████▋| 358068/371472 [6:20:07<59:40,  3.74it/s]  
 96%|█████████▋| 358069/371472 [6:20:07<59:00,  3.79it/s]
 96%|█████████▋| 358070/371472 [6:20:07<58:38,  3.81it/s]
 96%|█████████▋| 358071/371472 [6:20:08<56:43,  3.94it/s]
 96%|█████████▋| 358072/371472 [6:20:08<58:42,  3.80it/s]
 96%|█████████▋| 358073/371472 [6:20:08<57:55,  3.86it/s]
 96%|█████████▋| 358074/371472 [6:20:08<58:20,  3.83it/s]
 96%|████��████▋| 358075/371472 [6:20:09<1:04:16,  3.47it/s]
 96%|█████████▋| 358076/371472 [6:20:09<1:03:48,  3.50it/s]
 96%|█████████▋| 358077/371472 [6:20:09<1:04:24,  3.47it/s]
 96%|█████████▋| 358078/371472 [6:20:10<1:01:17,  3.64it/s]
 96%|█████████▋| 358079/371472 [6:20:10<1:02:23,  3.58it/s]
 96%|█████████▋| 358080/371472 [6:20:10<1:05:12,  3.42it/s]
                                                           
{'loss': 2.5584, 'learning_rate': 1.3246353078066755e-07, 'epoch': 15.42}
+
 96%|█████████▋| 358080/371472 [6:20:10<1:05:12,  3.42it/s]
 96%|█████████▋| 358081/371472 [6:20:10<1:01:33,  3.63it/s]
 96%|█████████▋| 358082/371472 [6:20:11<1:01:53,  3.61it/s]
 96%|█████████▋| 358083/371472 [6:20:11<1:02:51,  3.55it/s]
 96%|█████████▋| 358084/371472 [6:20:11<1:02:09,  3.59it/s]
 96%|█████████▋| 358085/371472 [6:20:12<1:00:22,  3.70it/s]
 96%|█████████▋| 358086/371472 [6:20:12<59:54,  3.72it/s]  
 96%|█████████▋| 358087/371472 [6:20:12<59:38,  3.74it/s]
 96%|█████████▋| 358088/371472 [6:20:12<58:42,  3.80it/s]
 96%|█████████▋| 358089/371472 [6:20:13<58:29,  3.81it/s]
 96%|█████████▋| 358090/371472 [6:20:13<58:16,  3.83it/s]
 96%|█████████▋| 358091/371472 [6:20:13<58:26,  3.82it/s]
 96%|█████████▋| 358092/371472 [6:20:13<59:00,  3.78it/s]
 96%|█████████▋| 358093/371472 [6:20:14<1:00:18,  3.70it/s]
 96%|█████████▋| 358094/371472 [6:20:14<1:00:16,  3.70it/s]
 96%|█████████▋| 358095/371472 [6:20:14<1:00:17,  3.70it/s]
 96%|█████████▋| 358096/371472 [6:20:14<1:01:48,  3.61it/s]
 96%|█████████▋| 358097/371472 [6:20:15<1:03:39,  3.50it/s]
 96%|█████████▋| 358098/371472 [6:20:15<1:01:26,  3.63it/s]
 96%|█████████▋| 358099/371472 [6:20:15<1:01:37,  3.62it/s]
 96%|█████████▋| 358100/371472 [6:20:16<1:03:01,  3.54it/s]
                                                           
{'loss': 2.6771, 'learning_rate': 1.3241504880518862e-07, 'epoch': 15.42}
+
 96%|█████████▋| 358100/371472 [6:20:16<1:03:01,  3.54it/s]
 96%|█████████▋| 358101/371472 [6:20:16<1:03:20,  3.52it/s]
 96%|█████████▋| 358102/371472 [6:20:16<1:01:36,  3.62it/s]
 96%|█████████▋| 358103/371472 [6:20:16<59:29,  3.75it/s]  
 96%|█████████▋| 358104/371472 [6:20:17<58:08,  3.83it/s]
 96%|█████████▋| 358105/371472 [6:20:17<59:31,  3.74it/s]
 96%|█████████▋| 358106/371472 [6:20:17<1:01:40,  3.61it/s]
 96%|█████████▋| 358107/371472 [6:20:18<1:06:17,  3.36it/s]
 96%|█████████▋| 358108/371472 [6:20:18<1:06:22,  3.36it/s]
 96%|█████████▋| 358109/371472 [6:20:18<1:04:34,  3.45it/s]
 96%|█████████▋| 358110/371472 [6:20:18<1:03:50,  3.49it/s]
 96%|█████████▋| 358111/371472 [6:20:19<1:01:14,  3.64it/s]
 96%|█████████▋| 358112/371472 [6:20:19<58:43,  3.79it/s]  
 96%|█████████▋| 358113/371472 [6:20:19<1:00:29,  3.68it/s]
 96%|█████████▋| 358114/371472 [6:20:19<59:18,  3.75it/s]  
 96%|█████████▋| 358115/371472 [6:20:20<1:01:58,  3.59it/s]
 96%|█████████▋| 358116/371472 [6:20:20<59:57,  3.71it/s]  
 96%|█████████▋| 358117/371472 [6:20:20<1:02:20,  3.57it/s]
 96%|█████████▋| 358118/371472 [6:20:21<1:01:48,  3.60it/s]
 96%|█████████▋| 358119/371472 [6:20:21<1:00:41,  3.67it/s]
 96%|█████████▋| 358120/371472 [6:20:21<1:02:57,  3.53it/s]
                                                           
{'loss': 2.6383, 'learning_rate': 1.3236656682970977e-07, 'epoch': 15.42}
+
 96%|█████████▋| 358120/371472 [6:20:21<1:02:57,  3.53it/s]
 96%|█████████▋| 358121/371472 [6:20:21<1:03:33,  3.50it/s]
 96%|█████████▋| 358122/371472 [6:20:22<1:02:30,  3.56it/s]
 96%|█████████▋| 358123/371472 [6:20:22<1:08:23,  3.25it/s]
 96%|█████████▋| 358124/371472 [6:20:22<1:06:11,  3.36it/s]
 96%|█████████▋| 358125/371472 [6:20:23<1:03:24,  3.51it/s]
 96%|█████████▋| 358126/371472 [6:20:23<1:02:37,  3.55it/s]
 96%|█████████▋| 358127/371472 [6:20:23<1:05:04,  3.42it/s]
 96%|█████████▋| 358128/371472 [6:20:23<1:03:21,  3.51it/s]
 96%|█████████▋| 358129/371472 [6:20:24<1:02:39,  3.55it/s]
 96%|█████████▋| 358130/371472 [6:20:24<1:02:07,  3.58it/s]
 96%|███████���█▋| 358131/371472 [6:20:24<1:01:06,  3.64it/s]
 96%|█████████▋| 358132/371472 [6:20:25<1:01:37,  3.61it/s]
 96%|█████████▋| 358133/371472 [6:20:25<59:31,  3.74it/s]  
 96%|█████████▋| 358134/371472 [6:20:25<59:33,  3.73it/s]
 96%|█████████▋| 358135/371472 [6:20:25<1:00:48,  3.66it/s]
 96%|█████████▋| 358136/371472 [6:20:26<1:01:19,  3.62it/s]
 96%|█████████▋| 358137/371472 [6:20:26<1:00:35,  3.67it/s]
 96%|█████████▋| 358138/371472 [6:20:26<1:02:58,  3.53it/s]
 96%|█████████▋| 358139/371472 [6:20:26<1:00:41,  3.66it/s]
 96%|█████████▋| 358140/371472 [6:20:27<1:02:24,  3.56it/s]
                                                           
{'loss': 2.8025, 'learning_rate': 1.3231808485423082e-07, 'epoch': 15.43}
+
 96%|█████████▋| 358140/371472 [6:20:27<1:02:24,  3.56it/s]
 96%|█████████▋| 358141/371472 [6:20:27<1:04:12,  3.46it/s]
 96%|█████████▋| 358142/371472 [6:20:27<1:04:33,  3.44it/s]
 96%|█████████▋| 358143/371472 [6:20:28<1:04:28,  3.45it/s]
 96%|█████████▋| 358144/371472 [6:20:28<1:05:34,  3.39it/s]
 96%|█████████▋| 358145/371472 [6:20:28<1:06:22,  3.35it/s]
 96%|█████████▋| 358146/371472 [6:20:29<1:04:05,  3.47it/s]
 96%|█████████▋| 358147/371472 [6:20:29<1:04:09,  3.46it/s]
 96%|█████████▋| 358148/371472 [6:20:29<1:10:11,  3.16it/s]
 96%|█████████▋| 358149/371472 [6:20:29<1:06:46,  3.33it/s]
 96%|█████████▋| 358150/371472 [6:20:30<1:02:49,  3.53it/s]
 96%|█████████▋| 358151/371472 [6:20:30<1:03:03,  3.52it/s]
 96%|█████████▋| 358152/371472 [6:20:30<1:02:06,  3.57it/s]
 96%|█████████▋| 358153/371472 [6:20:31<1:02:34,  3.55it/s]
 96%|█████████▋| 358154/371472 [6:20:31<1:00:51,  3.65it/s]
 96%|█████████▋| 358155/371472 [6:20:31<1:04:20,  3.45it/s]
 96%|█████████▋| 358156/371472 [6:20:31<1:02:18,  3.56it/s]
 96%|█████████▋| 358157/371472 [6:20:32<1:07:46,  3.27it/s]
 96%|█████████▋| 358158/371472 [6:20:32<1:07:28,  3.29it/s]
 96%|█████████▋| 358159/371472 [6:20:32<1:06:37,  3.33it/s]
 96%|█████████▋| 358160/371472 [6:20:33<1:07:45,  3.27it/s]
                                                           
{'loss': 2.5147, 'learning_rate': 1.32269602878752e-07, 'epoch': 15.43}
+
 96%|█████████▋| 358160/371472 [6:20:33<1:07:45,  3.27it/s]
 96%|█████████▋| 358161/371472 [6:20:33<1:06:09,  3.35it/s]
 96%|█████████▋| 358162/371472 [6:20:33<1:03:15,  3.51it/s]
 96%|█████████▋| 358163/371472 [6:20:34<1:04:52,  3.42it/s]
 96%|█████████▋| 358164/371472 [6:20:34<1:09:01,  3.21it/s]
 96%|█████████▋| 358165/371472 [6:20:34<1:06:21,  3.34it/s]
 96%|█████████▋| 358166/371472 [6:20:34<1:06:48,  3.32it/s]
 96%|█████████▋| 358167/371472 [6:20:35<1:05:16,  3.40it/s]
 96%|█████████▋| 358168/371472 [6:20:35<1:04:04,  3.46it/s]
 96%|█████████▋| 358169/371472 [6:20:35<1:05:39,  3.38it/s]
 96%|█████████▋| 358170/371472 [6:20:36<1:06:06,  3.35it/s]
 96%|█████████▋| 358171/371472 [6:20:36<1:02:54,  3.52it/s]
 96%|█████████▋| 358172/371472 [6:20:36<1:06:45,  3.32it/s]
 96%|█████████▋| 358173/371472 [6:20:36<1:04:40,  3.43it/s]
 96%|█████████▋| 358174/371472 [6:20:37<1:03:59,  3.46it/s]
 96%|█████████▋| 358175/371472 [6:20:37<1:04:33,  3.43it/s]
 96%|█████████▋| 358176/371472 [6:20:37<1:08:39,  3.23it/s]
 96%|█████████▋| 358177/371472 [6:20:38<1:06:37,  3.33it/s]
 96%|█████████▋| 358178/371472 [6:20:38<1:06:17,  3.34it/s]
 96%|█████████▋| 358179/371472 [6:20:38<1:02:55,  3.52it/s]
 96%|█████████▋| 358180/371472 [6:20:38<1:00:26,  3.67it/s]
                                                           
{'loss': 2.5479, 'learning_rate': 1.3222112090327304e-07, 'epoch': 15.43}
+
 96%|█████████▋| 358180/371472 [6:20:38<1:00:26,  3.67it/s]
 96%|█████████▋| 358181/371472 [6:20:39<57:46,  3.83it/s]  
 96%|█████████▋| 358182/371472 [6:20:39<57:08,  3.88it/s]
 96%|█████████▋| 358183/371472 [6:20:39<56:24,  3.93it/s]
 96%|█████████▋| 358184/371472 [6:20:40<58:48,  3.77it/s]
 96%|█████████▋| 358185/371472 [6:20:40<59:43,  3.71it/s]
 96%|█████████▋| 358186/371472 [6:20:40<1:00:05,  3.68it/s]
 96%|████████��▋| 358187/371472 [6:20:40<1:00:42,  3.65it/s]
 96%|█████████▋| 358188/371472 [6:20:41<59:22,  3.73it/s]  
 96%|█████████▋| 358189/371472 [6:20:41<59:31,  3.72it/s]
 96%|█████████▋| 358190/371472 [6:20:41<1:02:03,  3.57it/s]
 96%|█████████▋| 358191/371472 [6:20:41<1:01:08,  3.62it/s]
 96%|█████████▋| 358192/371472 [6:20:42<1:01:43,  3.59it/s]
 96%|█████████▋| 358193/371472 [6:20:42<1:00:16,  3.67it/s]
 96%|█████████▋| 358194/371472 [6:20:42<1:02:00,  3.57it/s]
 96%|█████████▋| 358195/371472 [6:20:43<1:00:37,  3.65it/s]
 96%|█████████▋| 358196/371472 [6:20:43<57:59,  3.82it/s]  
 96%|█████████▋| 358197/371472 [6:20:43<1:00:38,  3.65it/s]
 96%|█████████▋| 358198/371472 [6:20:43<1:02:13,  3.56it/s]
 96%|█████████▋| 358199/371472 [6:20:44<1:04:03,  3.45it/s]
 96%|█████████▋| 358200/371472 [6:20:44<1:03:30,  3.48it/s]
                                                           
{'loss': 2.6118, 'learning_rate': 1.321726389277942e-07, 'epoch': 15.43}
+
 96%|█████████▋| 358200/371472 [6:20:44<1:03:30,  3.48it/s]
 96%|█████████▋| 358201/371472 [6:20:44<1:03:32,  3.48it/s]
 96%|█████████▋| 358202/371472 [6:20:45<1:03:03,  3.51it/s]
 96%|█████████▋| 358203/371472 [6:20:45<1:01:54,  3.57it/s]
 96%|█████████▋| 358204/371472 [6:20:45<1:03:11,  3.50it/s]
 96%|█████████▋| 358205/371472 [6:20:45<1:01:25,  3.60it/s]
 96%|█████████▋| 358206/371472 [6:20:46<1:06:17,  3.34it/s]
 96%|█████████▋| 358207/371472 [6:20:46<1:06:23,  3.33it/s]
 96%|█████████▋| 358208/371472 [6:20:46<1:06:00,  3.35it/s]
 96%|█████████▋| 358209/371472 [6:20:47<1:08:26,  3.23it/s]
 96%|█████████▋| 358210/371472 [6:20:47<1:05:04,  3.40it/s]
 96%|█████████▋| 358211/371472 [6:20:47<1:02:52,  3.52it/s]
 96%|█████████▋| 358212/371472 [6:20:47<1:01:58,  3.57it/s]
 96%|█████████▋| 358213/371472 [6:20:48<1:02:02,  3.56it/s]
 96%|█████████▋| 358214/371472 [6:20:48<59:17,  3.73it/s]  
 96%|█████████▋| 358215/371472 [6:20:48<1:00:49,  3.63it/s]
 96%|█████████▋| 358216/371472 [6:20:49<1:01:09,  3.61it/s]
 96%|█████████▋| 358217/371472 [6:20:49<1:02:42,  3.52it/s]
 96%|█████████▋| 358218/371472 [6:20:49<1:00:40,  3.64it/s]
 96%|█████████▋| 358219/371472 [6:20:49<1:00:05,  3.68it/s]
 96%|█████████▋| 358220/371472 [6:20:50<58:06,  3.80it/s]  
                                                         
{'loss': 2.6103, 'learning_rate': 1.3212415695231526e-07, 'epoch': 15.43}
+
 96%|█████████▋| 358220/371472 [6:20:50<58:06,  3.80it/s]
 96%|█████████▋| 358221/371472 [6:20:50<1:08:12,  3.24it/s]
 96%|█████████▋| 358222/371472 [6:20:50<1:08:16,  3.23it/s]
 96%|█████████▋| 358223/371472 [6:20:51<1:05:25,  3.37it/s]
 96%|█████████▋| 358224/371472 [6:20:51<1:02:14,  3.55it/s]
 96%|█████████▋| 358225/371472 [6:20:51<1:01:43,  3.58it/s]
 96%|█████████▋| 358226/371472 [6:20:51<1:05:14,  3.38it/s]
 96%|█████████▋| 358227/371472 [6:20:52<1:03:36,  3.47it/s]
 96%|█████████▋| 358228/371472 [6:20:52<1:07:45,  3.26it/s]
 96%|█████████▋| 358229/371472 [6:20:52<1:05:53,  3.35it/s]
 96%|█████████▋| 358230/371472 [6:20:53<1:02:08,  3.55it/s]
 96%|█████████▋| 358231/371472 [6:20:53<1:00:09,  3.67it/s]
 96%|█████████▋| 358232/371472 [6:20:53<59:49,  3.69it/s]  
 96%|█████████▋| 358233/371472 [6:20:53<59:21,  3.72it/s]
 96%|█████████▋| 358234/371472 [6:20:54<58:28,  3.77it/s]
 96%|█████████▋| 358235/371472 [6:20:54<56:12,  3.93it/s]
 96%|█████████▋| 358236/371472 [6:20:54<57:54,  3.81it/s]
 96%|█████████▋| 358237/371472 [6:20:54<57:35,  3.83it/s]
 96%|█████████▋| 358238/371472 [6:20:55<57:24,  3.84it/s]
 96%|█████████▋| 358239/371472 [6:20:55<1:01:18,  3.60it/s]
 96%|█████████▋| 358240/371472 [6:20:55<1:03:09,  3.49it/s]
                                                           
{'loss': 2.7325, 'learning_rate': 1.320756749768364e-07, 'epoch': 15.43}
+
 96%|█████████▋| 358240/371472 [6:20:55<1:03:09,  3.49it/s]
 96%|█████████▋| 358241/371472 [6:20:56<1:02:59,  3.50it/s]
 96%|█████████▋| 358242/371472 [6:20:56<1:06:29,  3.32it/s]
 96%|█████████▋| 358243/371472 [6:20:56<1:06:10,  3.33it/s]
 96%|█████████▋| 358244/371472 [6:20:56<1:07:06,  3.29it/s]
 96%|█████████▋| 358245/371472 [6:20:57<1:07:38,  3.26it/s]
 96%|█████████▋| 358246/371472 [6:20:57<1:06:03,  3.34it/s]
 96%|█████████▋| 358247/371472 [6:20:57<1:03:36,  3.47it/s]
 96%|█████████▋| 358248/371472 [6:20:58<1:01:03,  3.61it/s]
 96%|█████████▋| 358249/371472 [6:20:58<59:54,  3.68it/s]  
 96%|█████████▋| 358250/371472 [6:20:58<59:35,  3.70it/s]
 96%|█████████▋| 358251/371472 [6:20:58<58:34,  3.76it/s]
 96%|█████████▋| 358252/371472 [6:20:59<57:58,  3.80it/s]
 96%|█████████▋| 358253/371472 [6:20:59<1:00:31,  3.64it/s]
 96%|█████████▋| 358254/371472 [6:20:59<1:00:29,  3.64it/s]
 96%|█████████▋| 358255/371472 [6:21:00<1:03:04,  3.49it/s]
 96%|█████████▋| 358256/371472 [6:21:00<1:01:30,  3.58it/s]
 96%|█████████▋| 358257/371472 [6:21:00<1:04:03,  3.44it/s]
 96%|█████████▋| 358258/371472 [6:21:00<1:01:44,  3.57it/s]
 96%|█████████▋| 358259/371472 [6:21:01<1:09:06,  3.19it/s]
 96%|█████████▋| 358260/371472 [6:21:01<1:07:37,  3.26it/s]
                                                           
{'loss': 2.5245, 'learning_rate': 1.3202719300135746e-07, 'epoch': 15.43}
+
 96%|█████████▋| 358260/371472 [6:21:01<1:07:37,  3.26it/s]
 96%|█████████▋| 358261/371472 [6:21:01<1:14:50,  2.94it/s]
 96%|█████████▋| 358262/371472 [6:21:02<1:10:19,  3.13it/s]
 96%|█████████▋| 358263/371472 [6:21:02<1:07:50,  3.25it/s]
 96%|█████████▋| 358264/371472 [6:21:02<1:05:11,  3.38it/s]
 96%|█████████▋| 358265/371472 [6:21:03<1:04:19,  3.42it/s]
 96%|█████████▋| 358266/371472 [6:21:03<1:02:01,  3.55it/s]
 96%|█████████▋| 358267/371472 [6:21:03<1:05:08,  3.38it/s]
 96%|█████████▋| 358268/371472 [6:21:03<1:03:14,  3.48it/s]
 96%|█████████▋| 358269/371472 [6:21:04<1:05:28,  3.36it/s]
 96%|█████████▋| 358270/371472 [6:21:04<1:02:43,  3.51it/s]
 96%|█████████▋| 358271/371472 [6:21:04<1:02:37,  3.51it/s]
 96%|█████████▋| 358272/371472 [6:21:05<1:08:29,  3.21it/s]
 96%|█████████▋| 358273/371472 [6:21:05<1:06:35,  3.30it/s]
 96%|█████████▋| 358274/371472 [6:21:05<1:05:48,  3.34it/s]
 96%|█████████▋| 358275/371472 [6:21:06<1:04:19,  3.42it/s]
 96%|█████████▋| 358276/371472 [6:21:06<1:02:34,  3.51it/s]
 96%|█████████▋| 358277/371472 [6:21:06<1:05:03,  3.38it/s]
 96%|█████████▋| 358278/371472 [6:21:06<1:04:51,  3.39it/s]
 96%|█████████▋| 358279/371472 [6:21:07<1:03:34,  3.46it/s]
 96%|█████████▋| 358280/371472 [6:21:07<1:04:25,  3.41it/s]
                                                           
{'loss': 2.7162, 'learning_rate': 1.3197871102587863e-07, 'epoch': 15.43}
+
 96%|█████████▋| 358280/371472 [6:21:07<1:04:25,  3.41it/s]
 96%|█████████▋| 358281/371472 [6:21:07<1:06:36,  3.30it/s]
 96%|█████████▋| 358282/371472 [6:21:08<1:15:34,  2.91it/s]
 96%|█████████▋| 358283/371472 [6:21:08<1:17:00,  2.85it/s]
 96%|█████████▋| 358284/371472 [6:21:08<1:18:18,  2.81it/s]
 96%|█████████▋| 358285/371472 [6:21:09<1:12:27,  3.03it/s]
 96%|█████████▋| 358286/371472 [6:21:09<1:07:22,  3.26it/s]
 96%|█████████▋| 358287/371472 [6:21:09<1:02:36,  3.51it/s]
 96%|█████████▋| 358288/371472 [6:21:09<1:01:00,  3.60it/s]
 96%|█████████▋| 358289/371472 [6:21:10<59:41,  3.68it/s]  
 96%|█████████▋| 358290/371472 [6:21:10<59:22,  3.70it/s]
 96%|█████████▋| 358291/371472 [6:21:10<58:32,  3.75it/s]
 96%|█████████▋| 358292/371472 [6:21:11<1:04:45,  3.39it/s]
 96%|█████████▋| 358293/371472 [6:21:11<1:02:53,  3.49it/s]
 96%|█████████▋| 358294/371472 [6:21:11<1:01:40,  3.56it/s]
 96%|█████████▋| 358295/371472 [6:21:11<59:58,  3.66it/s]  
 96%|█████████▋| 358296/371472 [6:21:12<1:00:24,  3.64it/s]
 96%|█████████▋| 358297/371472 [6:21:12<1:00:10,  3.65it/s]
 96%|█████████▋| 358298/371472 [6:21:12<59:02,  3.72it/s]  
 96%|█████████▋| 358299/371472 [6:21:13<1:00:12,  3.65it/s]
 96%|█████████▋| 358300/371472 [6:21:13<1:00:05,  3.65it/s]
                                                           
{'loss': 2.6455, 'learning_rate': 1.3193022905039968e-07, 'epoch': 15.43}
+
 96%|█████████▋| 358300/371472 [6:21:13<1:00:05,  3.65it/s]
 96%|█████████▋| 358301/371472 [6:21:13<59:13,  3.71it/s]  
 96%|█████████▋| 358302/371472 [6:21:13<59:54,  3.66it/s]
 96%|█████████▋| 358303/371472 [6:21:14<58:50,  3.73it/s]
 96%|█████████▋| 358304/371472 [6:21:14<58:24,  3.76it/s]
 96%|█████████▋| 358305/371472 [6:21:14<57:10,  3.84it/s]
 96%|█████████▋| 358306/371472 [6:21:14<56:53,  3.86it/s]
 96%|█████████▋| 358307/371472 [6:21:15<56:07,  3.91it/s]
 96%|█████████▋| 358308/371472 [6:21:15<58:00,  3.78it/s]
 96%|█████████▋| 358309/371472 [6:21:15<59:20,  3.70it/s]
 96%|█████████▋| 358310/371472 [6:21:15<1:00:54,  3.60it/s]
 96%|█████████▋| 358311/371472 [6:21:16<1:00:09,  3.65it/s]
 96%|█████████▋| 358312/371472 [6:21:16<1:02:04,  3.53it/s]
 96%|█████████▋| 358313/371472 [6:21:16<1:02:19,  3.52it/s]
 96%|█████████▋| 358314/371472 [6:21:17<1:00:14,  3.64it/s]
 96%|█████████▋| 358315/371472 [6:21:17<1:00:39,  3.61it/s]
 96%|█████████▋| 358316/371472 [6:21:17<1:04:28,  3.40it/s]
 96%|█████████▋| 358317/371472 [6:21:17<1:01:52,  3.54it/s]
 96%|█████████▋| 358318/371472 [6:21:18<1:00:44,  3.61it/s]
 96%|█████████▋| 358319/371472 [6:21:18<1:03:43,  3.44it/s]
 96%|█████████▋| 358320/371472 [6:21:18<1:04:56,  3.37it/s]
                                                           
{'loss': 2.6967, 'learning_rate': 1.3188174707492083e-07, 'epoch': 15.43}
+
 96%|█████████▋| 358320/371472 [6:21:18<1:04:56,  3.37it/s]
 96%|█████████▋| 358321/371472 [6:21:19<1:02:05,  3.53it/s]
 96%|█████████▋| 358322/371472 [6:21:19<1:02:48,  3.49it/s]
 96%|█████████▋| 358323/371472 [6:21:19<1:00:39,  3.61it/s]
 96%|█████████▋| 358324/371472 [6:21:19<1:00:13,  3.64it/s]
 96%|█████████▋| 358325/371472 [6:21:20<58:41,  3.73it/s]  
 96%|█████████▋| 358326/371472 [6:21:20<59:04,  3.71it/s]
 96%|█████████▋| 358327/371472 [6:21:20<1:03:33,  3.45it/s]
 96%|█████████▋| 358328/371472 [6:21:21<1:03:23,  3.46it/s]
 96%|█████████▋| 358329/371472 [6:21:21<1:01:51,  3.54it/s]
 96%|█████████▋| 358330/371472 [6:21:21<1:00:21,  3.63it/s]
 96%|█████████▋| 358331/371472 [6:21:21<1:03:53,  3.43it/s]
 96%|█████████▋| 358332/371472 [6:21:22<1:06:24,  3.30it/s]
 96%|█████████▋| 358333/371472 [6:21:22<1:11:22,  3.07it/s]
 96%|█████████▋| 358334/371472 [6:21:22<1:05:57,  3.32it/s]
 96%|█████████▋| 358335/371472 [6:21:23<1:02:32,  3.50it/s]
 96%|█████████▋| 358336/371472 [6:21:23<1:00:49,  3.60it/s]
 96%|█████████▋| 358337/371472 [6:21:23<1:01:19,  3.57it/s]
 96%|█████████▋| 358338/371472 [6:21:23<1:00:25,  3.62it/s]
 96%|█████████▋| 358339/371472 [6:21:24<1:07:26,  3.25it/s]
 96%|█████████▋| 358340/371472 [6:21:24<1:05:34,  3.34it/s]
                                                           
{'loss': 2.4969, 'learning_rate': 1.318332650994419e-07, 'epoch': 15.43}
+
 96%|█████████▋| 358340/371472 [6:21:24<1:05:34,  3.34it/s]
 96%|█████████▋| 358341/371472 [6:21:24<1:02:32,  3.50it/s]
 96%|█████████▋| 358342/371472 [6:21:25<1:02:05,  3.52it/s]
 96%|█████████▋| 358343/371472 [6:21:25<1:01:55,  3.53it/s]
 96%|█████████▋| 358344/371472 [6:21:25<1:02:11,  3.52it/s]
 96%|█████████▋| 358345/371472 [6:21:25<1:00:32,  3.61it/s]
 96%|█████████▋| 358346/371472 [6:21:26<59:14,  3.69it/s]  
 96%|█████████▋| 358347/371472 [6:21:26<1:03:48,  3.43it/s]
 96%|█████████▋| 358348/371472 [6:21:26<1:02:07,  3.52it/s]
 96%|█████████▋| 358349/371472 [6:21:27<1:06:15,  3.30it/s]
 96%|█████████▋| 358350/371472 [6:21:27<1:07:53,  3.22it/s]
 96%|█████████▋| 358351/371472 [6:21:27<1:04:44,  3.38it/s]
 96%|█████████▋| 358352/371472 [6:21:27<1:00:51,  3.59it/s]
 96%|█████████▋| 358353/371472 [6:21:28<1:07:01,  3.26it/s]
 96%|█████████▋| 358354/371472 [6:21:28<1:04:52,  3.37it/s]
 96%|█████████▋| 358355/371472 [6:21:28<1:04:30,  3.39it/s]
 96%|█████████▋| 358356/371472 [6:21:29<1:03:56,  3.42it/s]
 96%|█████████▋| 358357/371472 [6:21:29<1:03:29,  3.44it/s]
 96%|█████████▋| 358358/371472 [6:21:29<1:02:30,  3.50it/s]
 96%|█████████▋| 358359/371472 [6:21:30<1:03:35,  3.44it/s]
 96%|█████████▋| 358360/371472 [6:21:30<1:05:01,  3.36it/s]
                                                           
{'loss': 2.5631, 'learning_rate': 1.3178478312396305e-07, 'epoch': 15.44}
+
 96%|█████████▋| 358360/371472 [6:21:30<1:05:01,  3.36it/s]
 96%|█████████▋| 358361/371472 [6:21:30<1:06:09,  3.30it/s]
 96%|█████████▋| 358362/371472 [6:21:30<1:02:59,  3.47it/s]
 96%|█████████▋| 358363/371472 [6:21:31<1:04:47,  3.37it/s]
 96%|█████████▋| 358364/371472 [6:21:31<1:03:32,  3.44it/s]
 96%|█████████▋| 358365/371472 [6:21:31<1:01:49,  3.53it/s]
 96%|█████████▋| 358366/371472 [6:21:32<1:02:08,  3.52it/s]
 96%|█████████▋| 358367/371472 [6:21:32<1:00:56,  3.58it/s]
 96%|█████████▋| 358368/371472 [6:21:32<59:26,  3.67it/s]  
 96%|█████████▋| 358369/371472 [6:21:32<1:02:29,  3.49it/s]
 96%|█████████▋| 358370/371472 [6:21:33<1:01:05,  3.57it/s]
 96%|█████████▋| 358371/371472 [6:21:33<1:02:12,  3.51it/s]
 96%|█████████▋| 358372/371472 [6:21:33<1:03:44,  3.43it/s]
 96%|█████████▋| 358373/371472 [6:21:34<1:01:31,  3.55it/s]
 96%|█████████▋| 358374/371472 [6:21:34<59:46,  3.65it/s]  
 96%|█████████▋| 358375/371472 [6:21:34<1:02:12,  3.51it/s]
 96%|█████████▋| 358376/371472 [6:21:34<1:01:31,  3.55it/s]
 96%|█████████▋| 358377/371472 [6:21:35<59:04,  3.69it/s]  
 96%|█████████▋| 358378/371472 [6:21:35<1:00:02,  3.64it/s]
 96%|█████████▋| 358379/371472 [6:21:35<59:41,  3.66it/s]  
 96%|█████████▋| 358380/371472 [6:21:35<58:37,  3.72it/s]
                                                         
{'loss': 2.3669, 'learning_rate': 1.3173630114848412e-07, 'epoch': 15.44}
+
 96%|█████████▋| 358380/371472 [6:21:35<58:37,  3.72it/s]
 96%|█████████▋| 358381/371472 [6:21:36<59:51,  3.65it/s]
 96%|█████████▋| 358382/371472 [6:21:36<1:08:57,  3.16it/s]
 96%|█████████▋| 358383/371472 [6:21:36<1:06:20,  3.29it/s]
 96%|█████████▋| 358384/371472 [6:21:37<1:03:09,  3.45it/s]
 96%|█████████▋| 358385/371472 [6:21:37<1:01:11,  3.56it/s]
 96%|█████████▋| 358386/371472 [6:21:37<59:23,  3.67it/s]  
 96%|█████████▋| 358387/371472 [6:21:37<58:28,  3.73it/s]
 96%|█████████▋| 358388/371472 [6:21:38<58:37,  3.72it/s]
 96%|█████████▋| 358389/371472 [6:21:38<1:03:00,  3.46it/s]
 96%|█████████▋| 358390/371472 [6:21:38<1:02:41,  3.48it/s]
 96%|█████████▋| 358391/371472 [6:21:39<1:02:14,  3.50it/s]
 96%|█████████▋| 358392/371472 [6:21:39<1:03:22,  3.44it/s]
 96%|█████████▋| 358393/371472 [6:21:39<1:00:35,  3.60it/s]
 96%|█████████▋| 358394/371472 [6:21:39<1:01:47,  3.53it/s]
 96%|█████████▋| 358395/371472 [6:21:40<1:09:47,  3.12it/s]
 96%|█████████▋| 358396/371472 [6:21:40<1:04:51,  3.36it/s]
 96%|█████████▋| 358397/371472 [6:21:40<1:03:48,  3.42it/s]
 96%|█████████▋| 358398/371472 [6:21:41<1:04:08,  3.40it/s]
 96%|█████████▋| 358399/371472 [6:21:41<1:07:33,  3.23it/s]
 96%|█████████▋| 358400/371472 [6:21:41<1:09:17,  3.14it/s]
                                                           
{'loss': 2.6126, 'learning_rate': 1.3168781917300527e-07, 'epoch': 15.44}
+
 96%|█████████▋| 358400/371472 [6:21:41<1:09:17,  3.14it/s]
 96%|█████████▋| 358401/371472 [6:21:42<1:05:58,  3.30it/s]
 96%|█████████▋| 358402/371472 [6:21:42<1:03:44,  3.42it/s]
 96%|█████████▋| 358403/371472 [6:21:42<1:04:02,  3.40it/s]
 96%|█████████▋| 358404/371472 [6:21:42<1:00:41,  3.59it/s]
 96%|█████████▋| 358405/371472 [6:21:43<59:59,  3.63it/s]  
 96%|█████████▋| 358406/371472 [6:21:43<1:00:23,  3.61it/s]
 96%|█████████▋| 358407/371472 [6:21:43<59:11,  3.68it/s]  
 96%|█████████▋| 358408/371472 [6:21:44<58:41,  3.71it/s]
 96%|█████████▋| 358409/371472 [6:21:44<56:15,  3.87it/s]
 96%|█████████▋| 358410/371472 [6:21:44<55:32,  3.92it/s]
 96%|█████████▋| 358411/371472 [6:21:44<1:01:02,  3.57it/s]
 96%|█████████▋| 358412/371472 [6:21:45<1:03:02,  3.45it/s]
 96%|█████████▋| 358413/371472 [6:21:45<1:04:29,  3.37it/s]
 96%|█████████▋| 358414/371472 [6:21:45<1:04:08,  3.39it/s]
 96%|█████████▋| 358415/371472 [6:21:46<1:00:14,  3.61it/s]
 96%|█████████▋| 358416/371472 [6:21:46<59:10,  3.68it/s]  
 96%|█████████▋| 358417/371472 [6:21:46<1:01:45,  3.52it/s]
 96%|█████████▋| 358418/371472 [6:21:46<1:01:29,  3.54it/s]
 96%|█████████▋| 358419/371472 [6:21:47<1:01:38,  3.53it/s]
 96%|█████████▋| 358420/371472 [6:21:47<1:00:00,  3.63it/s]
                                                           
{'loss': 2.4964, 'learning_rate': 1.3163933719752635e-07, 'epoch': 15.44}
+
 96%|█████████▋| 358420/371472 [6:21:47<1:00:00,  3.63it/s]
 96%|█████████▋| 358421/371472 [6:21:47<1:00:28,  3.60it/s]
 96%|█████████▋| 358422/371472 [6:21:48<1:01:44,  3.52it/s]
 96%|█████████▋| 358423/371472 [6:21:48<59:29,  3.66it/s]  
 96%|█████████▋| 358424/371472 [6:21:48<1:00:45,  3.58it/s]
 96%|█████████▋| 358425/371472 [6:21:48<57:43,  3.77it/s]  
 96%|█████████▋| 358426/371472 [6:21:49<1:03:30,  3.42it/s]
 96%|█████████▋| 358427/371472 [6:21:49<1:05:21,  3.33it/s]
 96%|█████████▋| 358428/371472 [6:21:49<1:03:34,  3.42it/s]
 96%|█████████▋| 358429/371472 [6:21:50<1:04:10,  3.39it/s]
 96%|█████████▋| 358430/371472 [6:21:50<1:01:57,  3.51it/s]
 96%|█████████▋| 358431/371472 [6:21:50<1:01:14,  3.55it/s]
 96%|█████████▋| 358432/371472 [6:21:50<1:00:52,  3.57it/s]
 96%|█████████▋| 358433/371472 [6:21:51<59:47,  3.63it/s]  
 96%|█████████▋| 358434/371472 [6:21:51<1:00:01,  3.62it/s]
 96%|█████████▋| 358435/371472 [6:21:51<1:05:23,  3.32it/s]
 96%|█████████▋| 358436/371472 [6:21:52<1:05:52,  3.30it/s]
 96%|█████████▋| 358437/371472 [6:21:52<1:06:00,  3.29it/s]
 96%|█████████▋| 358438/371472 [6:21:52<1:03:18,  3.43it/s]
 96%|█████████▋| 358439/371472 [6:21:52<1:07:28,  3.22it/s]
 96%|█████████▋| 358440/371472 [6:21:53<1:05:06,  3.34it/s]
                                                           
{'loss': 2.4935, 'learning_rate': 1.315908552220474e-07, 'epoch': 15.44}
+
 96%|█████████▋| 358440/371472 [6:21:53<1:05:06,  3.34it/s]
 96%|█████████▋| 358441/371472 [6:21:53<1:03:17,  3.43it/s]
 96%|█████████▋| 358442/371472 [6:21:53<1:08:39,  3.16it/s]
 96%|█████████▋| 358443/371472 [6:21:54<1:05:24,  3.32it/s]
 96%|█████████▋| 358444/371472 [6:21:54<1:16:08,  2.85it/s]
 96%|█████████▋| 358445/371472 [6:21:55<1:17:47,  2.79it/s]
 96%|█████████▋| 358446/371472 [6:21:55<1:12:35,  2.99it/s]
 96%|█████████▋| 358447/371472 [6:21:55<1:12:00,  3.01it/s]
 96%|█████████▋| 358448/371472 [6:21:55<1:06:58,  3.24it/s]
 96%|█████████▋| 358449/371472 [6:21:56<1:04:37,  3.36it/s]
 96%|█████████▋| 358450/371472 [6:21:56<1:03:12,  3.43it/s]
 96%|█████████▋| 358451/371472 [6:21:56<1:05:09,  3.33it/s]
 96%|█████████▋| 358452/371472 [6:21:57<1:07:22,  3.22it/s]
 96%|█████████▋| 358453/371472 [6:21:57<1:03:44,  3.40it/s]
 96%|█████████▋| 358454/371472 [6:21:57<1:01:27,  3.53it/s]
 96%|█████████▋| 358455/371472 [6:21:57<1:00:42,  3.57it/s]
 96%|█████████▋| 358456/371472 [6:21:58<59:22,  3.65it/s]  
 96%|█████████▋| 358457/371472 [6:21:58<58:19,  3.72it/s]
 96%|█████████▋| 358458/371472 [6:21:58<57:18,  3.78it/s]
 96%|█████████▋| 358459/371472 [6:21:58<57:23,  3.78it/s]
 96%|█████████▋| 358460/371472 [6:21:59<57:05,  3.80it/s]
                                                         
{'loss': 2.6113, 'learning_rate': 1.3154237324656857e-07, 'epoch': 15.44}
+
 96%|█████████▋| 358460/371472 [6:21:59<57:05,  3.80it/s]
 96%|█████████▋| 358461/371472 [6:21:59<59:06,  3.67it/s]
 96%|█████████▋| 358462/371472 [6:21:59<1:01:13,  3.54it/s]
 96%|█████████▋| 358463/371472 [6:22:00<1:00:14,  3.60it/s]
 96%|█████████▋| 358464/371472 [6:22:00<1:00:03,  3.61it/s]
 96%|█████████▋| 358465/371472 [6:22:00<1:00:26,  3.59it/s]
 96%|█████████▋| 358466/371472 [6:22:00<59:24,  3.65it/s]  
 96%|█████████▋| 358467/371472 [6:22:01<59:30,  3.64it/s]
 96%|█████████▋| 358468/371472 [6:22:01<59:47,  3.62it/s]
 96%|█████████▋| 358469/371472 [6:22:01<1:00:02,  3.61it/s]
 96%|█████████▋| 358470/371472 [6:22:02<1:06:32,  3.26it/s]
 97%|█████████▋| 358471/371472 [6:22:02<1:05:13,  3.32it/s]
 97%|█████████▋| 358472/371472 [6:22:02<1:03:30,  3.41it/s]
 97%|█████████▋| 358473/371472 [6:22:02<1:01:06,  3.55it/s]
 97%|█████████▋| 358474/371472 [6:22:03<59:22,  3.65it/s]  
 97%|█████████▋| 358475/371472 [6:22:03<1:01:46,  3.51it/s]
 97%|█████████▋| 358476/371472 [6:22:03<1:04:40,  3.35it/s]
 97%|█████████▋| 358477/371472 [6:22:04<1:02:13,  3.48it/s]
 97%|█████████▋| 358478/371472 [6:22:04<1:03:02,  3.44it/s]
 97%|█████████▋| 358479/371472 [6:22:04<1:01:48,  3.50it/s]
 97%|█████████▋| 358480/371472 [6:22:04<1:00:39,  3.57it/s]
                                                           
{'loss': 2.568, 'learning_rate': 1.3149389127108961e-07, 'epoch': 15.44}
+
 97%|█████████▋| 358480/371472 [6:22:04<1:00:39,  3.57it/s]
 97%|█████████▋| 358481/371472 [6:22:05<1:00:37,  3.57it/s]
 97%|█████████▋| 358482/371472 [6:22:05<59:11,  3.66it/s]  
 97%|█████████▋| 358483/371472 [6:22:05<1:00:22,  3.59it/s]
 97%|█████████▋| 358484/371472 [6:22:05<59:05,  3.66it/s]  
 97%|█████████▋| 358485/371472 [6:22:06<57:59,  3.73it/s]
 97%|█████████▋| 358486/371472 [6:22:06<59:42,  3.63it/s]
 97%|█████████▋| 358487/371472 [6:22:06<1:01:04,  3.54it/s]
 97%|█████████▋| 358488/371472 [6:22:07<1:00:40,  3.57it/s]
 97%|█████████▋| 358489/371472 [6:22:07<1:00:22,  3.58it/s]
 97%|█████████▋| 358490/371472 [6:22:07<1:00:19,  3.59it/s]
 97%|█████████▋| 358491/371472 [6:22:07<1:00:41,  3.56it/s]
 97%|█████████▋| 358492/371472 [6:22:08<1:02:05,  3.48it/s]
 97%|█████████▋| 358493/371472 [6:22:08<1:06:04,  3.27it/s]
 97%|█████████▋| 358494/371472 [6:22:08<1:01:55,  3.49it/s]
 97%|█████████▋| 358495/371472 [6:22:09<1:02:02,  3.49it/s]
 97%|█████████▋| 358496/371472 [6:22:09<1:04:28,  3.35it/s]
 97%|█████████▋| 358497/371472 [6:22:09<1:02:15,  3.47it/s]
 97%|█████████▋| 358498/371472 [6:22:09<1:00:26,  3.58it/s]
 97%|█████████▋| 358499/371472 [6:22:10<1:00:42,  3.56it/s]
 97%|█████████▋| 358500/371472 [6:22:10<59:53,  3.61it/s]  
                                                         
{'loss': 2.6067, 'learning_rate': 1.3144540929561076e-07, 'epoch': 15.44}
+
 97%|█████████▋| 358500/371472 [6:22:10<59:53,  3.61it/s]
 97%|█████████▋| 358501/371472 [6:22:10<1:00:53,  3.55it/s]
 97%|█████████▋| 358502/371472 [6:22:11<1:01:19,  3.53it/s]
 97%|█████████▋| 358503/371472 [6:22:11<1:05:44,  3.29it/s]
 97%|█████████▋| 358504/371472 [6:22:11<1:04:59,  3.33it/s]
 97%|█████████▋| 358505/371472 [6:22:11<1:03:06,  3.42it/s]
 97%|█████████▋| 358506/371472 [6:22:12<1:01:10,  3.53it/s]
 97%|█████████▋| 358507/371472 [6:22:12<1:02:41,  3.45it/s]
 97%|█████████▋| 358508/371472 [6:22:13<1:19:04,  2.73it/s]
 97%|█████████▋| 358509/371472 [6:22:13<1:13:58,  2.92it/s]
 97%|█████████▋| 358510/371472 [6:22:13<1:11:51,  3.01it/s]
 97%|█████████▋| 358511/371472 [6:22:13<1:06:16,  3.26it/s]
 97%|█████████▋| 358512/371472 [6:22:14<1:02:47,  3.44it/s]
 97%|█████████▋| 358513/371472 [6:22:14<1:01:44,  3.50it/s]
 97%|█████████▋| 358514/371472 [6:22:14<59:14,  3.65it/s]  
 97%|█████████▋| 358515/371472 [6:22:14<57:25,  3.76it/s]
 97%|█████████▋| 358516/371472 [6:22:15<56:46,  3.80it/s]
 97%|█████████▋| 358517/371472 [6:22:15<57:07,  3.78it/s]
 97%|█████████▋| 358518/371472 [6:22:15<1:02:28,  3.46it/s]
 97%|█████████▋| 358519/371472 [6:22:16<1:02:09,  3.47it/s]
 97%|█████████▋| 358520/371472 [6:22:16<1:00:16,  3.58it/s]
                                                           
{'loss': 2.5474, 'learning_rate': 1.313969273201318e-07, 'epoch': 15.44}
+
 97%|█████████▋| 358520/371472 [6:22:16<1:00:16,  3.58it/s]
 97%|█████████▋| 358521/371472 [6:22:16<1:02:17,  3.46it/s]
 97%|█████████▋| 358522/371472 [6:22:16<1:00:11,  3.59it/s]
 97%|█████████▋| 358523/371472 [6:22:17<58:59,  3.66it/s]  
 97%|█████████▋| 358524/371472 [6:22:17<57:23,  3.76it/s]
 97%|█████████▋| 358525/371472 [6:22:17<57:03,  3.78it/s]
 97%|█████████▋| 358526/371472 [6:22:18<1:03:42,  3.39it/s]
 97%|█████████▋| 358527/371472 [6:22:18<1:01:10,  3.53it/s]
 97%|█████████▋| 358528/371472 [6:22:18<57:47,  3.73it/s]  
 97%|█████████▋| 358529/371472 [6:22:18<58:15,  3.70it/s]
 97%|█████████▋| 358530/371472 [6:22:19<57:20,  3.76it/s]
 97%|█████████▋| 358531/371472 [6:22:19<56:23,  3.82it/s]
 97%|█████████▋| 358532/371472 [6:22:19<59:15,  3.64it/s]
 97%|█████████▋| 358533/371472 [6:22:19<56:49,  3.79it/s]
 97%|█████████▋| 358534/371472 [6:22:20<1:00:26,  3.57it/s]
 97%|█████████▋| 358535/371472 [6:22:20<1:02:43,  3.44it/s]
 97%|█████████▋| 358536/371472 [6:22:20<1:00:54,  3.54it/s]
 97%|█████████▋| 358537/371472 [6:22:21<1:00:37,  3.56it/s]
 97%|█████████▋| 358538/371472 [6:22:21<58:18,  3.70it/s]  
 97%|█████████▋| 358539/371472 [6:22:21<1:01:09,  3.52it/s]
 97%|█████████▋| 358540/371472 [6:22:21<1:04:57,  3.32it/s]
                                                           
{'loss': 2.6256, 'learning_rate': 1.31348445344653e-07, 'epoch': 15.44}
+
 97%|█████████▋| 358540/371472 [6:22:21<1:04:57,  3.32it/s]
 97%|█████████▋| 358541/371472 [6:22:22<1:06:23,  3.25it/s]
 97%|█████████▋| 358542/371472 [6:22:22<1:01:58,  3.48it/s]
 97%|█████████▋| 358543/371472 [6:22:22<1:01:51,  3.48it/s]
 97%|█████████▋| 358544/371472 [6:22:23<1:00:42,  3.55it/s]
 97%|█████████▋| 358545/371472 [6:22:23<59:06,  3.65it/s]  
 97%|█████████▋| 358546/371472 [6:22:23<58:43,  3.67it/s]
 97%|█████████▋| 358547/371472 [6:22:23<59:01,  3.65it/s]
 97%|█████████▋| 358548/371472 [6:22:24<57:56,  3.72it/s]
 97%|█████████▋| 358549/371472 [6:22:24<57:59,  3.71it/s]
 97%|█████████▋| 358550/371472 [6:22:24<58:50,  3.66it/s]
 97%|█████████▋| 358551/371472 [6:22:25<1:01:08,  3.52it/s]
 97%|█████████▋| 358552/371472 [6:22:25<1:03:07,  3.41it/s]
 97%|█████████▋| 358553/371472 [6:22:25<1:02:32,  3.44it/s]
 97%|█████████▋| 358554/371472 [6:22:25<1:02:59,  3.42it/s]
 97%|█████████▋| 358555/371472 [6:22:26<1:02:25,  3.45it/s]
 97%|█████████▋| 358556/371472 [6:22:26<1:03:31,  3.39it/s]
 97%|█████████▋| 358557/371472 [6:22:26<1:02:45,  3.43it/s]
 97%|█████████▋| 358558/371472 [6:22:27<1:02:53,  3.42it/s]
 97%|█████████▋| 358559/371472 [6:22:27<1:01:44,  3.49it/s]
 97%|█████████▋| 358560/371472 [6:22:27<1:00:53,  3.53it/s]
                                                           
{'loss': 2.4989, 'learning_rate': 1.3129996336917403e-07, 'epoch': 15.44}
+
 97%|█████████▋| 358560/371472 [6:22:27<1:00:53,  3.53it/s]
 97%|█████████▋| 358561/371472 [6:22:27<1:02:12,  3.46it/s]
 97%|█████████▋| 358562/371472 [6:22:28<1:00:51,  3.54it/s]
 97%|█████████▋| 358563/371472 [6:22:28<59:09,  3.64it/s]  
 97%|█████████▋| 358564/371472 [6:22:28<59:32,  3.61it/s]
 97%|█████████▋| 358565/371472 [6:22:29<59:38,  3.61it/s]
 97%|█████████▋| 358566/371472 [6:22:29<1:00:58,  3.53it/s]
 97%|█████████▋| 358567/371472 [6:22:29<59:02,  3.64it/s]  
 97%|█████████▋| 358568/371472 [6:22:29<57:06,  3.77it/s]
 97%|█████████▋| 358569/371472 [6:22:30<57:40,  3.73it/s]
 97%|█████████▋| 358570/371472 [6:22:30<58:08,  3.70it/s]
 97%|█████████▋| 358571/371472 [6:22:30<58:20,  3.69it/s]
 97%|█████████▋| 358572/371472 [6:22:30<58:30,  3.68it/s]
 97%|█████████▋| 358573/371472 [6:22:31<58:59,  3.64it/s]
 97%|█████████▋| 358574/371472 [6:22:31<1:00:05,  3.58it/s]
 97%|█████████▋| 358575/371472 [6:22:31<1:01:10,  3.51it/s]
 97%|█████████▋| 358576/371472 [6:22:32<1:00:34,  3.55it/s]
 97%|█████████▋| 358577/371472 [6:22:32<1:04:50,  3.31it/s]
 97%|█████████▋| 358578/371472 [6:22:32<1:05:25,  3.28it/s]
 97%|█████████▋| 358579/371472 [6:22:33<1:08:11,  3.15it/s]
 97%|█████████▋| 358580/371472 [6:22:33<1:06:47,  3.22it/s]
                                                           
{'loss': 2.5126, 'learning_rate': 1.3125148139369518e-07, 'epoch': 15.44}
+
 97%|█████████▋| 358580/371472 [6:22:33<1:06:47,  3.22it/s]
 97%|█████████▋| 358581/371472 [6:22:33<1:04:49,  3.31it/s]
 97%|█████████▋| 358582/371472 [6:22:33<1:03:38,  3.38it/s]
 97%|███���█████▋| 358583/371472 [6:22:34<1:02:31,  3.44it/s]
 97%|█████████▋| 358584/371472 [6:22:34<1:00:10,  3.57it/s]
 97%|█████████▋| 358585/371472 [6:22:34<58:41,  3.66it/s]  
 97%|█████████▋| 358586/371472 [6:22:34<58:54,  3.65it/s]
 97%|█████████▋| 358587/371472 [6:22:35<1:00:19,  3.56it/s]
 97%|█████████▋| 358588/371472 [6:22:35<1:01:33,  3.49it/s]
 97%|█████████▋| 358589/371472 [6:22:35<59:12,  3.63it/s]  
 97%|█████████▋| 358590/371472 [6:22:36<1:05:52,  3.26it/s]
 97%|█████████▋| 358591/371472 [6:22:36<1:01:45,  3.48it/s]
 97%|█████████▋| 358592/371472 [6:22:36<1:01:47,  3.47it/s]
 97%|█████████▋| 358593/371472 [6:22:37<1:01:31,  3.49it/s]
 97%|█████████▋| 358594/371472 [6:22:37<59:16,  3.62it/s]  
 97%|█████████▋| 358595/371472 [6:22:37<1:00:07,  3.57it/s]
 97%|█████████▋| 358596/371472 [6:22:37<1:00:08,  3.57it/s]
 97%|█████████▋| 358597/371472 [6:22:38<57:32,  3.73it/s]  
 97%|█████████▋| 358598/371472 [6:22:38<1:07:13,  3.19it/s]
 97%|█████████▋| 358599/371472 [6:22:38<1:03:24,  3.38it/s]
 97%|█████████▋| 358600/371472 [6:22:39<1:00:26,  3.55it/s]
                                                           
{'loss': 2.4675, 'learning_rate': 1.3120299941821625e-07, 'epoch': 15.45}
+
 97%|█████████▋| 358600/371472 [6:22:39<1:00:26,  3.55it/s]
 97%|█████████▋| 358601/371472 [6:22:39<59:06,  3.63it/s]  
 97%|█████████▋| 358602/371472 [6:22:39<58:58,  3.64it/s]
 97%|█████████▋| 358603/371472 [6:22:39<59:31,  3.60it/s]
 97%|█████████▋| 358604/371472 [6:22:40<57:47,  3.71it/s]
 97%|█████████▋| 358605/371472 [6:22:40<59:37,  3.60it/s]
 97%|█████████▋| 358606/371472 [6:22:40<58:45,  3.65it/s]
 97%|█████████▋| 358607/371472 [6:22:40<1:02:04,  3.45it/s]
 97%|█████████▋| 358608/371472 [6:22:41<1:01:13,  3.50it/s]
 97%|█████████▋| 358609/371472 [6:22:41<1:03:10,  3.39it/s]
 97%|█████████▋| 358610/371472 [6:22:41<1:01:55,  3.46it/s]
 97%|█████████▋| 358611/371472 [6:22:42<1:04:25,  3.33it/s]
 97%|█████████▋| 358612/371472 [6:22:42<1:02:32,  3.43it/s]
 97%|█████████▋| 358613/371472 [6:22:42<1:02:33,  3.43it/s]
 97%|█████████▋| 358614/371472 [6:22:43<1:02:25,  3.43it/s]
 97%|█████████▋| 358615/371472 [6:22:43<1:02:40,  3.42it/s]
 97%|█████████▋| 358616/371472 [6:22:43<1:04:55,  3.30it/s]
 97%|█████████▋| 358617/371472 [6:22:43<1:05:57,  3.25it/s]
 97%|█████████▋| 358618/371472 [6:22:44<1:05:45,  3.26it/s]
 97%|█████████▋| 358619/371472 [6:22:44<1:07:52,  3.16it/s]
 97%|█████████▋| 358620/371472 [6:22:44<1:04:16,  3.33it/s]
                                                           
{'loss': 2.5771, 'learning_rate': 1.311545174427374e-07, 'epoch': 15.45}
+
 97%|█████████▋| 358620/371472 [6:22:44<1:04:16,  3.33it/s]
 97%|█████████▋| 358621/371472 [6:22:45<1:02:42,  3.42it/s]
 97%|█████████▋| 358622/371472 [6:22:45<1:01:04,  3.51it/s]
 97%|█████████▋| 358623/371472 [6:22:45<1:00:04,  3.56it/s]
 97%|█████████▋| 358624/371472 [6:22:45<58:42,  3.65it/s]  
 97%|█████████▋| 358625/371472 [6:22:46<57:49,  3.70it/s]
 97%|█████████▋| 358626/371472 [6:22:46<58:17,  3.67it/s]
 97%|█████████▋| 358627/371472 [6:22:46<1:00:43,  3.53it/s]
 97%|█████████▋| 358628/371472 [6:22:47<1:05:36,  3.26it/s]
 97%|█████████▋| 358629/371472 [6:22:47<1:04:09,  3.34it/s]
 97%|█████████▋| 358630/371472 [6:22:47<1:02:55,  3.40it/s]
 97%|█████████▋| 358631/371472 [6:22:47<1:02:08,  3.44it/s]
 97%|█████████▋| 358632/371472 [6:22:48<1:09:02,  3.10it/s]
 97%|█████████▋| 358633/371472 [6:22:48<1:04:34,  3.31it/s]
 97%|█████████▋| 358634/371472 [6:22:48<1:01:04,  3.50it/s]
 97%|█████████▋| 358635/371472 [6:22:49<58:42,  3.64it/s]  
 97%|█████████▋| 358636/371472 [6:22:49<59:20,  3.60it/s]
 97%|█████████▋| 358637/371472 [6:22:49<58:22,  3.66it/s]
 97%|█████████▋| 358638/371472 [6:22:49<59:28,  3.60it/s]
 97%|█████████▋| 358639/371472 [6:22:50<57:49,  3.70it/s]
 97%|█████████▋| 358640/371472 [6:22:50<56:53,  3.76it/s]
                                                         
{'loss': 2.696, 'learning_rate': 1.3110603546725848e-07, 'epoch': 15.45}
+
 97%|█████████▋| 358640/371472 [6:22:50<56:53,  3.76it/s]
 97%|█████████▋| 358641/371472 [6:22:50<1:02:01,  3.45it/s]
 97%|█████████▋| 358642/371472 [6:22:51<59:50,  3.57it/s]  
 97%|█████████▋| 358643/371472 [6:22:51<1:03:47,  3.35it/s]
 97%|█████████▋| 358644/371472 [6:22:51<1:02:08,  3.44it/s]
 97%|█████████▋| 358645/371472 [6:22:51<1:01:38,  3.47it/s]
 97%|█████████▋| 358646/371472 [6:22:52<1:02:32,  3.42it/s]
 97%|█████████▋| 358647/371472 [6:22:52<1:05:08,  3.28it/s]
 97%|█████████▋| 358648/371472 [6:22:52<1:06:50,  3.20it/s]
 97%|█████████▋| 358649/371472 [6:22:53<1:06:49,  3.20it/s]
 97%|█████████▋| 358650/371472 [6:22:53<1:10:19,  3.04it/s]
 97%|█████████▋| 358651/371472 [6:22:53<1:05:47,  3.25it/s]
 97%|█████████▋| 358652/371472 [6:22:54<1:01:45,  3.46it/s]
 97%|█████████▋| 358653/371472 [6:22:54<1:00:03,  3.56it/s]
 97%|█████████▋| 358654/371472 [6:22:54<1:01:21,  3.48it/s]
 97%|█████████▋| 358655/371472 [6:22:54<59:24,  3.60it/s]  
 97%|█████████▋| 358656/371472 [6:22:55<57:23,  3.72it/s]
 97%|█████████▋| 358657/371472 [6:22:55<56:52,  3.76it/s]
 97%|█████████▋| 358658/371472 [6:22:55<1:02:36,  3.41it/s]
 97%|█████████▋| 358659/371472 [6:22:56<1:04:44,  3.30it/s]
 97%|█████████▋| 358660/371472 [6:22:56<1:03:05,  3.38it/s]
                                                           
{'loss': 2.5282, 'learning_rate': 1.3105755349177963e-07, 'epoch': 15.45}
+
 97%|█████████▋| 358660/371472 [6:22:56<1:03:05,  3.38it/s]
 97%|█████████▋| 358661/371472 [6:22:56<1:03:58,  3.34it/s]
 97%|█████████▋| 358662/371472 [6:22:57<1:02:05,  3.44it/s]
 97%|█████████▋| 358663/371472 [6:22:57<1:01:31,  3.47it/s]
 97%|█████████▋| 358664/371472 [6:22:57<59:36,  3.58it/s]  
 97%|█████████▋| 358665/371472 [6:22:57<1:01:22,  3.48it/s]
 97%|█████████▋| 358666/371472 [6:22:58<59:56,  3.56it/s]  
 97%|█████████▋| 358667/371472 [6:22:58<58:55,  3.62it/s]
 97%|█████████▋| 358668/371472 [6:22:58<57:14,  3.73it/s]
 97%|█████████▋| 358669/371472 [6:22:58<55:32,  3.84it/s]
 97%|█████████▋| 358670/371472 [6:22:59<58:50,  3.63it/s]
 97%|█████████▋| 358671/371472 [6:22:59<1:05:43,  3.25it/s]
 97%|█████████▋| 358672/371472 [6:22:59<1:06:50,  3.19it/s]
 97%|█████████▋| 358673/371472 [6:23:00<1:06:07,  3.23it/s]
 97%|█████████▋| 358674/371472 [6:23:00<1:02:22,  3.42it/s]
 97%|█████████▋| 358675/371472 [6:23:00<1:04:14,  3.32it/s]
 97%|█████████▋| 358676/371472 [6:23:01<1:02:12,  3.43it/s]
 97%|█████████▋| 358677/371472 [6:23:01<59:49,  3.56it/s]  
 97%|█████████▋| 358678/371472 [6:23:01<58:36,  3.64it/s]
 97%|█████████▋| 358679/371472 [6:23:01<58:37,  3.64it/s]
 97%|█████████▋| 358680/371472 [6:23:02<58:28,  3.65it/s]
                                                         
{'loss': 2.4702, 'learning_rate': 1.310090715163007e-07, 'epoch': 15.45}
+
 97%|█████████▋| 358680/371472 [6:23:02<58:28,  3.65it/s]
 97%|█████████▋| 358681/371472 [6:23:02<59:21,  3.59it/s]
 97%|█████████▋| 358682/371472 [6:23:02<1:03:33,  3.35it/s]
 97%|█████████▋| 358683/371472 [6:23:03<1:02:11,  3.43it/s]
 97%|█████████▋| 358684/371472 [6:23:03<1:00:44,  3.51it/s]
 97%|█████████▋| 358685/371472 [6:23:03<58:49,  3.62it/s]  
 97%|█████████▋| 358686/371472 [6:23:03<55:59,  3.81it/s]
 97%|█████████▋| 358687/371472 [6:23:04<1:03:58,  3.33it/s]
 97%|█████████▋| 358688/371472 [6:23:04<1:01:43,  3.45it/s]
 97%|█████████▋| 358689/371472 [6:23:04<1:01:39,  3.45it/s]
 97%|█████████▋| 358690/371472 [6:23:04<1:01:04,  3.49it/s]
 97%|█████████▋| 358691/371472 [6:23:05<1:03:16,  3.37it/s]
 97%|█████████▋| 358692/371472 [6:23:05<1:02:48,  3.39it/s]
 97%|█████████▋| 358693/371472 [6:23:05<59:13,  3.60it/s]  
 97%|█████████▋| 358694/371472 [6:23:06<1:02:09,  3.43it/s]
 97%|█████████▋| 358695/371472 [6:23:06<1:02:44,  3.39it/s]
 97%|█████████▋| 358696/371472 [6:23:06<1:00:58,  3.49it/s]
 97%|█████████▋| 358697/371472 [6:23:06<59:17,  3.59it/s]  
 97%|█████████��| 358698/371472 [6:23:07<1:03:48,  3.34it/s]
 97%|█████████▋| 358699/371472 [6:23:07<1:01:01,  3.49it/s]
 97%|█████████▋| 358700/371472 [6:23:07<58:20,  3.65it/s]  
                                                         
{'loss': 2.6523, 'learning_rate': 1.3096058954082185e-07, 'epoch': 15.45}
+
 97%|█████████▋| 358700/371472 [6:23:07<58:20,  3.65it/s]
 97%|█████████▋| 358701/371472 [6:23:08<57:30,  3.70it/s]
 97%|█████████▋| 358702/371472 [6:23:08<57:45,  3.69it/s]
 97%|█████████▋| 358703/371472 [6:23:08<58:56,  3.61it/s]
 97%|█████████▋| 358704/371472 [6:23:08<56:59,  3.73it/s]
 97%|█████████▋| 358705/371472 [6:23:09<57:16,  3.72it/s]
 97%|█████████▋| 358706/371472 [6:23:09<57:13,  3.72it/s]
 97%|█████████▋| 358707/371472 [6:23:09<1:00:40,  3.51it/s]
 97%|█████████▋| 358708/371472 [6:23:10<58:06,  3.66it/s]  
 97%|█████████▋| 358709/371472 [6:23:10<57:01,  3.73it/s]
 97%|█████████▋| 358710/371472 [6:23:10<1:00:36,  3.51it/s]
 97%|█████████▋| 358711/371472 [6:23:10<1:03:39,  3.34it/s]
 97%|█████████▋| 358712/371472 [6:23:11<1:02:31,  3.40it/s]
 97%|█████████▋| 358713/371472 [6:23:11<1:00:21,  3.52it/s]
 97%|█████████▋| 358714/371472 [6:23:11<58:43,  3.62it/s]  
 97%|█████████▋| 358715/371472 [6:23:11<56:08,  3.79it/s]
 97%|█████████▋| 358716/371472 [6:23:12<58:26,  3.64it/s]
 97%|█████████▋| 358717/371472 [6:23:12<56:54,  3.74it/s]
 97%|█████████▋| 358718/371472 [6:23:12<56:24,  3.77it/s]
 97%|█████████▋| 358719/371472 [6:23:13<57:38,  3.69it/s]
 97%|█████████▋| 358720/371472 [6:23:13<57:01,  3.73it/s]
                                                         
{'loss': 2.7284, 'learning_rate': 1.309121075653429e-07, 'epoch': 15.45}
+
 97%|█████████▋| 358720/371472 [6:23:13<57:01,  3.73it/s]
 97%|█████████▋| 358721/371472 [6:23:13<58:44,  3.62it/s]
 97%|█████████▋| 358722/371472 [6:23:13<1:01:45,  3.44it/s]
 97%|█████████▋| 358723/371472 [6:23:14<1:00:54,  3.49it/s]
 97%|█████████▋| 358724/371472 [6:23:14<58:55,  3.61it/s]  
 97%|█████████▋| 358725/371472 [6:23:14<59:28,  3.57it/s]
 97%|█████████▋| 358726/371472 [6:23:15<58:01,  3.66it/s]
 97%|█████████▋| 358727/371472 [6:23:15<58:43,  3.62it/s]
 97%|█████████▋| 358728/371472 [6:23:15<58:41,  3.62it/s]
 97%|█████████▋| 358729/371472 [6:23:15<1:01:23,  3.46it/s]
 97%|█████████▋| 358730/371472 [6:23:16<59:44,  3.55it/s]  
 97%|█████████▋| 358731/371472 [6:23:16<59:00,  3.60it/s]
 97%|█████████▋| 358732/371472 [6:23:16<1:00:10,  3.53it/s]
 97%|█████████▋| 358733/371472 [6:23:16<57:01,  3.72it/s]  
 97%|█████████▋| 358734/371472 [6:23:17<55:54,  3.80it/s]
 97%|█████████▋| 358735/371472 [6:23:17<58:30,  3.63it/s]
 97%|█████████▋| 358736/371472 [6:23:17<1:02:33,  3.39it/s]
 97%|█████████▋| 358737/371472 [6:23:18<1:04:28,  3.29it/s]
 97%|█████████▋| 358738/371472 [6:23:18<1:02:36,  3.39it/s]
 97%|█████████▋| 358739/371472 [6:23:18<1:02:39,  3.39it/s]
 97%|█████████▋| 358740/371472 [6:23:19<1:02:35,  3.39it/s]
                                                           
{'loss': 2.5496, 'learning_rate': 1.3086362558986407e-07, 'epoch': 15.45}
+
 97%|█████████▋| 358740/371472 [6:23:19<1:02:35,  3.39it/s]
 97%|█████████▋| 358741/371472 [6:23:19<1:01:08,  3.47it/s]
 97%|█████████▋| 358742/371472 [6:23:19<58:04,  3.65it/s]  
 97%|█████████▋| 358743/371472 [6:23:19<58:08,  3.65it/s]
 97%|█████████▋| 358744/371472 [6:23:20<57:27,  3.69it/s]
 97%|█████████▋| 358745/371472 [6:23:20<55:00,  3.86it/s]
 97%|█████████▋| 358746/371472 [6:23:20<59:51,  3.54it/s]
 97%|█████████▋| 358747/371472 [6:23:20<1:01:48,  3.43it/s]
 97%|█████████▋| 358748/371472 [6:23:21<1:00:00,  3.53it/s]
 97%|█████████▋| 358749/371472 [6:23:21<1:00:49,  3.49it/s]
 97%|█████████▋| 358750/371472 [6:23:21<59:36,  3.56it/s]  
 97%|█████████▋| 358751/371472 [6:23:22<58:49,  3.60it/s]
 97%|█████████▋| 358752/371472 [6:23:22<57:08,  3.71it/s]
 97%|█████████▋| 358753/371472 [6:23:22<55:59,  3.79it/s]
 97%|█████████▋| 358754/371472 [6:23:22<1:00:30,  3.50it/s]
 97%|█████████▋| 358755/371472 [6:23:23<1:01:33,  3.44it/s]
 97%|█████████▋| 358756/371472 [6:23:23<1:01:45,  3.43it/s]
 97%|█████████▋| 358757/371472 [6:23:23<1:05:16,  3.25it/s]
 97%|█████████▋| 358758/371472 [6:23:24<1:01:27,  3.45it/s]
 97%|█████████▋| 358759/371472 [6:23:24<59:52,  3.54it/s]  
 97%|█████████▋| 358760/371472 [6:23:24<58:51,  3.60it/s]
                                                         
{'loss': 2.5931, 'learning_rate': 1.3081514361438512e-07, 'epoch': 15.45}
+
 97%|█████████▋| 358760/371472 [6:23:24<58:51,  3.60it/s]
 97%|█████████▋| 358761/371472 [6:23:24<57:52,  3.66it/s]
 97%|█████████▋| 358762/371472 [6:23:25<58:27,  3.62it/s]
 97%|█████████▋| 358763/371472 [6:23:25<59:05,  3.58it/s]
 97%|█████████▋| 358764/371472 [6:23:25<59:57,  3.53it/s]
 97%|█████████▋| 358765/371472 [6:23:26<1:12:08,  2.94it/s]
 97%|█████████▋| 358766/371472 [6:23:26<1:11:37,  2.96it/s]
 97%|█████████▋| 358767/371472 [6:23:26<1:06:35,  3.18it/s]
 97%|█████████▋| 358768/371472 [6:23:27<1:06:07,  3.20it/s]
 97%|█████████▋| 358769/371472 [6:23:27<1:01:54,  3.42it/s]
 97%|█████████▋| 358770/371472 [6:23:27<1:02:11,  3.40it/s]
 97%|█████████▋| 358771/371472 [6:23:28<1:09:29,  3.05it/s]
 97%|█████████▋| 358772/371472 [6:23:28<1:09:36,  3.04it/s]
 97%|█████████▋| 358773/371472 [6:23:28<1:07:59,  3.11it/s]
 97%|█████████▋| 358774/371472 [6:23:28<1:03:32,  3.33it/s]
 97%|█████████▋| 358775/371472 [6:23:29<1:00:30,  3.50it/s]
 97%|█████████▋| 358776/371472 [6:23:29<1:01:55,  3.42it/s]
 97%|█████████▋| 358777/371472 [6:23:29<1:04:36,  3.27it/s]
 97%|█████████▋| 358778/371472 [6:23:30<1:02:01,  3.41it/s]
 97%|█████████▋| 358779/371472 [6:23:30<59:09,  3.58it/s]  
 97%|█████████▋| 358780/371472 [6:23:30<56:45,  3.73it/s]
                                                         
{'loss': 2.6042, 'learning_rate': 1.3076666163890627e-07, 'epoch': 15.45}
+
 97%|█████████▋| 358780/371472 [6:23:30<56:45,  3.73it/s]
 97%|█████████▋| 358781/371472 [6:23:30<59:05,  3.58it/s]
 97%|█████████▋| 358782/371472 [6:23:31<59:25,  3.56it/s]
 97%|█████████▋| 358783/371472 [6:23:31<57:07,  3.70it/s]
 97%|█████████▋| 358784/371472 [6:23:31<56:50,  3.72it/s]
 97%|█████████▋| 358785/371472 [6:23:31<55:35,  3.80it/s]
 97%|█████████▋| 358786/371472 [6:23:32<55:25,  3.81it/s]
 97%|█████████▋| 358787/371472 [6:23:32<56:11,  3.76it/s]
 97%|█████████▋| 358788/371472 [6:23:32<57:27,  3.68it/s]
 97%|█████████▋| 358789/371472 [6:23:33<59:30,  3.55it/s]
 97%|█████████▋| 358790/371472 [6:23:33<58:36,  3.61it/s]
 97%|█████████▋| 358791/371472 [6:23:33<58:13,  3.63it/s]
 97%|█████████▋| 358792/371472 [6:23:33<1:01:51,  3.42it/s]
 97%|█████████▋| 358793/371472 [6:23:34<1:01:53,  3.41it/s]
 97%|█████████▋| 358794/371472 [6:23:34<59:01,  3.58it/s]  
 97%|█████████▋| 358795/371472 [6:23:34<57:39,  3.66it/s]
 97%|█████████▋| 358796/371472 [6:23:35<57:22,  3.68it/s]
 97%|█████████▋| 358797/371472 [6:23:35<55:20,  3.82it/s]
 97%|█████████▋| 358798/371472 [6:23:35<57:21,  3.68it/s]
 97%|█████████▋| 358799/371472 [6:23:35<57:44,  3.66it/s]
 97%|█████████▋| 358800/371472 [6:23:36<59:04,  3.57it/s]
                                                         
{'loss': 2.7924, 'learning_rate': 1.3071817966342734e-07, 'epoch': 15.45}
+
 97%|█████████▋| 358800/371472 [6:23:36<59:04,  3.57it/s]
 97%|█████████▋| 358801/371472 [6:23:36<58:29,  3.61it/s]
 97%|█████████▋| 358802/371472 [6:23:36<1:00:15,  3.50it/s]
 97%|█████████▋| 358803/371472 [6:23:36<59:18,  3.56it/s]  
 97%|█████████▋| 358804/371472 [6:23:37<58:49,  3.59it/s]
 97%|█████████▋| 358805/371472 [6:23:37<58:21,  3.62it/s]
 97%|█████████▋| 358806/371472 [6:23:37<1:00:41,  3.48it/s]
 97%|█████████▋| 358807/371472 [6:23:38<57:22,  3.68it/s]  
 97%|█████████▋| 358808/371472 [6:23:38<58:25,  3.61it/s]
 97%|█████████▋| 358809/371472 [6:23:38<56:13,  3.75it/s]
 97%|█████████▋| 358810/371472 [6:23:38<59:36,  3.54it/s]
 97%|█████████▋| 358811/371472 [6:23:39<58:25,  3.61it/s]
 97%|█████████▋| 358812/371472 [6:23:39<58:59,  3.58it/s]
 97%|█████████▋| 358813/371472 [6:23:39<57:20,  3.68it/s]
 97%|█████████▋| 358814/371472 [6:23:40<58:35,  3.60it/s]
 97%|█████████▋| 358815/371472 [6:23:40<59:36,  3.54it/s]
 97%|█████████▋| 358816/371472 [6:23:40<59:40,  3.53it/s]
 97%|█████████▋| 358817/371472 [6:23:40<1:02:46,  3.36it/s]
 97%|█████████▋| 358818/371472 [6:23:41<1:02:26,  3.38it/s]
 97%|█████████▋| 358819/371472 [6:23:41<1:00:49,  3.47it/s]
 97%|█████████▋| 358820/371472 [6:23:41<1:00:11,  3.50it/s]
                                                           
{'loss': 2.5375, 'learning_rate': 1.306696976879485e-07, 'epoch': 15.46}
+
 97%|█████████▋| 358820/371472 [6:23:41<1:00:11,  3.50it/s]
 97%|█████████▋| 358821/371472 [6:23:42<1:03:28,  3.32it/s]
 97%|█████████▋| 358822/371472 [6:23:42<1:01:41,  3.42it/s]
 97%|█████████▋| 358823/371472 [6:23:42<58:59,  3.57it/s]  
 97%|█████████▋| 358824/371472 [6:23:42<58:53,  3.58it/s]
 97%|█████████▋| 358825/371472 [6:23:43<59:39,  3.53it/s]
 97%|█████████▋| 358826/371472 [6:23:43<1:02:34,  3.37it/s]
 97%|█████████▋| 358827/371472 [6:23:43<1:00:56,  3.46it/s]
 97%|█████████▋| 358828/371472 [6:23:44<57:43,  3.65it/s]  
 97%|█████████▋| 358829/371472 [6:23:44<1:02:45,  3.36it/s]
 97%|█████████▋| 358830/371472 [6:23:44<1:02:17,  3.38it/s]
 97%|█████████▋| 358831/371472 [6:23:44<59:53,  3.52it/s]  
 97%|█████████▋| 358832/371472 [6:23:45<58:55,  3.58it/s]
 97%|█████████▋| 358833/371472 [6:23:45<57:03,  3.69it/s]
 97%|█████████▋| 358834/371472 [6:23:45<58:29,  3.60it/s]
 97%|█████████▋| 358835/371472 [6:23:46<59:25,  3.54it/s]
 97%|█████████▋| 358836/371472 [6:23:46<57:44,  3.65it/s]
 97%|█████████▋| 358837/371472 [6:23:46<59:15,  3.55it/s]
 97%|█████████▋| 358838/371472 [6:23:46<59:55,  3.51it/s]
 97%|█████████▋| 358839/371472 [6:23:47<58:06,  3.62it/s]
 97%|█████████▋| 358840/371472 [6:23:47<56:48,  3.71it/s]
                                                         
{'loss': 2.5327, 'learning_rate': 1.3062121571246954e-07, 'epoch': 15.46}
+
 97%|█████████▋| 358840/371472 [6:23:47<56:48,  3.71it/s]
 97%|█████████▋| 358841/371472 [6:23:47<56:34,  3.72it/s]
 97%|█████████▋| 358842/371472 [6:23:47<54:41,  3.85it/s]
 97%|█████████▋| 358843/371472 [6:23:48<53:17,  3.95it/s]
 97%|█████████▋| 358844/371472 [6:23:48<52:54,  3.98it/s]
 97%|█████████▋| 358845/371472 [6:23:48<53:39,  3.92it/s]
 97%|█████████▋| 358846/371472 [6:23:48<56:14,  3.74it/s]
 97%|█████████▋| 358847/371472 [6:23:49<55:44,  3.78it/s]
 97%|█████████▋| 358848/371472 [6:23:49<54:47,  3.84it/s]
 97%|█████████▋| 358849/371472 [6:23:49<55:21,  3.80it/s]
 97%|█████████▋| 358850/371472 [6:23:49<54:38,  3.85it/s]
 97%|█████████▋| 358851/371472 [6:23:50<54:27,  3.86it/s]
 97%|█████████▋| 358852/371472 [6:23:50<53:23,  3.94it/s]
 97%|█████████▋| 358853/371472 [6:23:50<53:42,  3.92it/s]
 97%|█████████▋| 358854/371472 [6:23:51<55:19,  3.80it/s]
 97%|█████████▋| 358855/371472 [6:23:51<58:07,  3.62it/s]
 97%|█████████▋| 358856/371472 [6:23:51<57:16,  3.67it/s]
 97%|█████████▋| 358857/371472 [6:23:51<1:01:32,  3.42it/s]
 97%|█████████▋| 358858/371472 [6:23:52<1:03:36,  3.31it/s]
 97%|█████████▋| 358859/371472 [6:23:52<1:01:49,  3.40it/s]
 97%|█████████▋| 358860/371472 [6:23:52<1:06:19,  3.17it/s]
                                                           
{'loss': 2.5791, 'learning_rate': 1.305727337369907e-07, 'epoch': 15.46}
+
 97%|█████████▋| 358860/371472 [6:23:52<1:06:19,  3.17it/s]
 97%|█████████▋| 358861/371472 [6:23:53<1:02:19,  3.37it/s]
 97%|█████████▋| 358862/371472 [6:23:53<1:00:39,  3.46it/s]
 97%|█████████▋| 358863/371472 [6:23:53<59:36,  3.53it/s]  
 97%|█████████▋| 358864/371472 [6:23:54<1:02:12,  3.38it/s]
 97%|█████████▋| 358865/371472 [6:23:54<1:00:38,  3.47it/s]
 97%|█████████▋| 358866/371472 [6:23:54<1:01:25,  3.42it/s]
 97%|█████████▋| 358867/371472 [6:23:54<58:35,  3.59it/s]  
 97%|█████████▋| 358868/371472 [6:23:55<59:30,  3.53it/s]
 97%|█████████▋| 358869/371472 [6:23:55<59:15,  3.54it/s]
 97%|█████████▋| 358870/371472 [6:23:55<57:16,  3.67it/s]
 97%|█████████▋| 358871/371472 [6:23:55<56:04,  3.75it/s]
 97%|█████████▋| 358872/371472 [6:23:56<1:01:26,  3.42it/s]
 97%|█████████▋| 358873/371472 [6:23:56<1:03:41,  3.30it/s]
 97%|█████████▋| 358874/371472 [6:23:56<1:03:32,  3.30it/s]
 97%|█████████▋| 358875/371472 [6:23:57<1:03:56,  3.28it/s]
 97%|█████████▋| 358876/371472 [6:23:57<1:06:48,  3.14it/s]
 97%|█████████▋| 358877/371472 [6:23:57<1:08:18,  3.07it/s]
 97%|█████████▋| 358878/371472 [6:23:58<1:09:05,  3.04it/s]
 97%|█████████▋| 358879/371472 [6:23:58<1:05:30,  3.20it/s]
 97%|█████████▋| 358880/371472 [6:23:58<1:01:03,  3.44it/s]
                                                           
{'loss': 2.5158, 'learning_rate': 1.3052425176151173e-07, 'epoch': 15.46}
+
 97%|█████████▋| 358880/371472 [6:23:58<1:01:03,  3.44it/s]
 97%|█████████▋| 358881/371472 [6:23:59<1:05:05,  3.22it/s]
 97%|█████████▋| 358882/371472 [6:23:59<1:02:23,  3.36it/s]
 97%|█████████▋| 358883/371472 [6:23:59<1:00:37,  3.46it/s]
 97%|█████████▋| 358884/371472 [6:23:59<57:18,  3.66it/s]  
 97%|█████████▋| 358885/371472 [6:24:00<57:22,  3.66it/s]
 97%|█████████▋| 358886/371472 [6:24:00<56:06,  3.74it/s]
 97%|█████████▋| 358887/371472 [6:24:00<55:49,  3.76it/s]
 97%|█████████▋| 358888/371472 [6:24:00<54:34,  3.84it/s]
 97%|█████████▋| 358889/371472 [6:24:01<55:29,  3.78it/s]
 97%|█████████▋| 358890/371472 [6:24:01<57:17,  3.66it/s]
 97%|█████████▋| 358891/371472 [6:24:01<55:57,  3.75it/s]
 97%|█████████▋| 358892/371472 [6:24:02<55:38,  3.77it/s]
 97%|█████████▋| 358893/371472 [6:24:02<54:57,  3.82it/s]
 97%|█████████▋| 358894/371472 [6:24:02<54:01,  3.88it/s]
 97%|█████████▋| 358895/371472 [6:24:02<54:19,  3.86it/s]
 97%|█████████▋| 358896/371472 [6:24:03<58:23,  3.59it/s]
 97%|█████████▋| 358897/371472 [6:24:03<1:06:36,  3.15it/s]
 97%|█████████▋| 358898/371472 [6:24:03<1:05:45,  3.19it/s]
 97%|█████████▋| 358899/371472 [6:24:04<1:03:45,  3.29it/s]
 97%|█████████▋| 358900/371472 [6:24:04<1:02:28,  3.35it/s]
                                                           
{'loss': 2.5614, 'learning_rate': 1.304757697860329e-07, 'epoch': 15.46}
+
 97%|█████████▋| 358900/371472 [6:24:04<1:02:28,  3.35it/s]
 97%|█████████▋| 358901/371472 [6:24:04<59:12,  3.54it/s]  
 97%|█████████▋| 358902/371472 [6:24:04<1:01:25,  3.41it/s]
 97%|█████████▋| 358903/371472 [6:24:05<1:03:16,  3.31it/s]
 97%|█████████▋| 358904/371472 [6:24:05<1:02:04,  3.37it/s]
 97%|█████████▋| 358905/371472 [6:24:05<59:54,  3.50it/s]  
 97%|█████████▋| 358906/371472 [6:24:06<59:38,  3.51it/s]
 97%|█████████▋| 358907/371472 [6:24:06<57:55,  3.62it/s]
 97%|█████████▋| 358908/371472 [6:24:06<58:44,  3.56it/s]
 97%|█████████▋| 358909/371472 [6:24:06<58:26,  3.58it/s]
 97%|█████████▋| 358910/371472 [6:24:07<57:18,  3.65it/s]
 97%|█████████▋| 358911/371472 [6:24:07<55:44,  3.76it/s]
 97%|█████████▋| 358912/371472 [6:24:07<56:23,  3.71it/s]
 97%|█████████▋| 358913/371472 [6:24:07<55:47,  3.75it/s]
 97%|█████████▋| 358914/371472 [6:24:08<57:43,  3.63it/s]
 97%|█████████▋| 358915/371472 [6:24:08<59:23,  3.52it/s]
 97%|█████████▋| 358916/371472 [6:24:08<57:59,  3.61it/s]
 97%|█████████▋| 358917/371472 [6:24:09<56:15,  3.72it/s]
 97%|█████████▋| 358918/371472 [6:24:09<56:01,  3.73it/s]
 97%|█████████▋| 358919/371472 [6:24:09<58:18,  3.59it/s]
 97%|█████████▋| 358920/371472 [6:24:09<58:19,  3.59it/s]
                                                         
{'loss': 2.5917, 'learning_rate': 1.3042728781055398e-07, 'epoch': 15.46}
+
 97%|█████████▋| 358920/371472 [6:24:09<58:19,  3.59it/s]
 97%|█████████▋| 358921/371472 [6:24:10<55:25,  3.77it/s]
 97%|█████████▋| 358922/371472 [6:24:10<56:27,  3.70it/s]
 97%|█████████▋| 358923/371472 [6:24:10<59:31,  3.51it/s]
 97%|█████████▋| 358924/371472 [6:24:11<1:01:20,  3.41it/s]
 97%|█████████▋| 358925/371472 [6:24:11<59:01,  3.54it/s]  
 97%|█████████▋| 358926/371472 [6:24:11<56:32,  3.70it/s]
 97%|█████████▋| 358927/371472 [6:24:11<59:08,  3.54it/s]
 97%|█████████▋| 358928/371472 [6:24:12<59:14,  3.53it/s]
 97%|█████████▋| 358929/371472 [6:24:12<1:00:33,  3.45it/s]
 97%|█████████▋| 358930/371472 [6:24:12<59:44,  3.50it/s]  
 97%|█████████▋| 358931/371472 [6:24:13<1:01:55,  3.38it/s]
 97%|█████████▋| 358932/371472 [6:24:13<1:01:03,  3.42it/s]
 97%|█████████▋| 358933/371472 [6:24:13<58:57,  3.54it/s]  
 97%|█████████▋| 358934/371472 [6:24:13<57:59,  3.60it/s]
 97%|█████████▋| 358935/371472 [6:24:14<59:20,  3.52it/s]
 97%|█████████▋| 358936/371472 [6:24:14<57:39,  3.62it/s]
 97%|█████████▋| 358937/371472 [6:24:14<55:41,  3.75it/s]
 97%|█████████▋| 358938/371472 [6:24:14<55:28,  3.77it/s]
 97%|█████████▋| 358939/371472 [6:24:15<55:35,  3.76it/s]
 97%|█████████▋| 358940/371472 [6:24:15<58:47,  3.55it/s]
                                                         
{'loss': 2.6282, 'learning_rate': 1.3037880583507513e-07, 'epoch': 15.46}
+
 97%|█████████▋| 358940/371472 [6:24:15<58:47,  3.55it/s]
 97%|█████████▋| 358941/371472 [6:24:15<57:13,  3.65it/s]
 97%|█████████▋| 358942/371472 [6:24:16<58:11,  3.59it/s]
 97%|█████████▋| 358943/371472 [6:24:16<55:14,  3.78it/s]
 97%|█████████▋| 358944/371472 [6:24:16<55:42,  3.75it/s]
 97%|█████████▋| 358945/371472 [6:24:16<57:34,  3.63it/s]
 97%|█████████▋| 358946/371472 [6:24:17<1:02:51,  3.32it/s]
 97%|█████████▋| 358947/371472 [6:24:17<1:01:16,  3.41it/s]
 97%|█████████▋| 358948/371472 [6:24:17<59:00,  3.54it/s]  
 97%|█████████▋| 358949/371472 [6:24:18<1:16:12,  2.74it/s]
 97%|█████████▋| 358950/371472 [6:24:18<1:08:51,  3.03it/s]
 97%|█████████▋| 358951/371472 [6:24:18<1:06:58,  3.12it/s]
 97%|█████████▋| 358952/371472 [6:24:19<1:03:11,  3.30it/s]
 97%|█████████▋| 358953/371472 [6:24:19<59:12,  3.52it/s]  
 97%|█████████▋| 358954/371472 [6:24:19<57:55,  3.60it/s]
 97%|█████████▋| 358955/371472 [6:24:19<55:45,  3.74it/s]
 97%|█████████▋| 358956/371472 [6:24:20<57:18,  3.64it/s]
 97%|█████████▋| 358957/371472 [6:24:20<58:48,  3.55it/s]
 97%|█████████▋| 358958/371472 [6:24:20<58:06,  3.59it/s]
 97%|█████████▋| 358959/371472 [6:24:21<59:44,  3.49it/s]
 97%|█████████▋| 358960/371472 [6:24:21<57:10,  3.65it/s]
                                                         
{'loss': 2.5659, 'learning_rate': 1.3033032385959618e-07, 'epoch': 15.46}
+
 97%|█████████▋| 358960/371472 [6:24:21<57:10,  3.65it/s]
 97%|█████████▋| 358961/371472 [6:24:21<58:26,  3.57it/s]
 97%|█████████▋| 358962/371472 [6:24:21<59:12,  3.52it/s]
 97%|█████████▋| 358963/371472 [6:24:22<57:55,  3.60it/s]
 97%|█████████▋| 358964/371472 [6:24:22<57:10,  3.65it/s]
 97%|█████████▋| 358965/371472 [6:24:22<57:24,  3.63it/s]
 97%|█████████▋| 358966/371472 [6:24:22<56:30,  3.69it/s]
 97%|█████████▋| 358967/371472 [6:24:23<58:18,  3.57it/s]
 97%|█████████▋| 358968/371472 [6:24:23<58:21,  3.57it/s]
 97%|█████████▋| 358969/371472 [6:24:23<57:57,  3.60it/s]
 97%|█████████▋| 358970/371472 [6:24:24<56:37,  3.68it/s]
 97%|█████████▋| 358971/371472 [6:24:24<55:39,  3.74it/s]
 97%|█████████▋| 358972/371472 [6:24:24<55:20,  3.76it/s]
 97%|█████████▋| 358973/371472 [6:24:24<56:32,  3.68it/s]
 97%|█████████▋| 358974/371472 [6:24:25<55:29,  3.75it/s]
 97%|█████████▋| 358975/371472 [6:24:25<55:57,  3.72it/s]
 97%|█████████▋| 358976/371472 [6:24:25<59:13,  3.52it/s]
 97%|█████████▋| 358977/371472 [6:24:25<56:18,  3.70it/s]
 97%|█████████▋| 358978/371472 [6:24:26<58:06,  3.58it/s]
 97%|█████████▋| 358979/371472 [6:24:26<58:13,  3.58it/s]
 97%|█████████▋| 358980/371472 [6:24:26<1:00:15,  3.45it/s]
                                                           
{'loss': 2.6214, 'learning_rate': 1.3028184188411725e-07, 'epoch': 15.46}
+
 97%|█████████▋| 358980/371472 [6:24:26<1:00:15,  3.45it/s]
 97%|█████████▋| 358981/371472 [6:24:27<1:02:25,  3.33it/s]
 97%|█████████▋| 358982/371472 [6:24:27<1:02:10,  3.35it/s]
 97%|█████████▋| 358983/371472 [6:24:27<59:46,  3.48it/s]  
 97%|█████████▋| 358984/371472 [6:24:27<58:38,  3.55it/s]
 97%|█████████▋| 358985/371472 [6:24:28<57:56,  3.59it/s]
 97%|█████████▋| 358986/371472 [6:24:28<57:01,  3.65it/s]
 97%|█████████▋| 358987/371472 [6:24:28<56:29,  3.68it/s]
 97%|█████████▋| 358988/371472 [6:24:29<56:52,  3.66it/s]
 97%|█████████▋| 358989/371472 [6:24:29<54:57,  3.79it/s]
 97%|█████████▋| 358990/371472 [6:24:29<54:25,  3.82it/s]
 97%|█████████▋| 358991/371472 [6:24:29<53:38,  3.88it/s]
 97%|█████████▋| 358992/371472 [6:24:30<53:22,  3.90it/s]
 97%|█████████▋| 358993/371472 [6:24:30<55:06,  3.77it/s]
 97%|█████████▋| 358994/371472 [6:24:30<57:04,  3.64it/s]
 97%|█████████▋| 358995/371472 [6:24:30<55:43,  3.73it/s]
 97%|█████████▋| 358996/371472 [6:24:31<54:58,  3.78it/s]
 97%|█████████▋| 358997/371472 [6:24:31<56:00,  3.71it/s]
 97%|█████████▋| 358998/371472 [6:24:31<55:00,  3.78it/s]
 97%|█████████▋| 358999/371472 [6:24:31<55:38,  3.74it/s]
 97%|█████████▋| 359000/371472 [6:24:32<57:29,  3.62it/s]
                                                         
{'loss': 2.5694, 'learning_rate': 1.302333599086384e-07, 'epoch': 15.46}
+
 97%|█████████▋| 359000/371472 [6:24:32<57:29,  3.62it/s]
 97%|█████████▋| 359001/371472 [6:24:32<57:22,  3.62it/s]
 97%|█████████▋| 359002/371472 [6:24:32<58:50,  3.53it/s]
 97%|█████████▋| 359003/371472 [6:24:33<1:00:16,  3.45it/s]
 97%|█████████▋| 359004/371472 [6:24:33<58:45,  3.54it/s]  
 97%|█████████▋| 359005/371472 [6:24:33<56:08,  3.70it/s]
 97%|█████████▋| 359006/371472 [6:24:33<54:48,  3.79it/s]
 97%|█████████▋| 359007/371472 [6:24:34<56:26,  3.68it/s]
 97%|█████████▋| 359008/371472 [6:24:34<58:57,  3.52it/s]
 97%|█████████▋| 359009/371472 [6:24:34<58:45,  3.54it/s]
 97%|█████████▋| 359010/371472 [6:24:35<59:08,  3.51it/s]
 97%|█████████▋| 359011/371472 [6:24:35<57:58,  3.58it/s]
 97%|█████████▋| 359012/371472 [6:24:35<57:33,  3.61it/s]
 97%|█████████▋| 359013/371472 [6:24:35<56:42,  3.66it/s]
 97%|█████████▋| 359014/371472 [6:24:36<55:43,  3.73it/s]
 97%|█████████▋| 359015/371472 [6:24:36<57:48,  3.59it/s]
 97%|█████████▋| 359016/371472 [6:24:36<1:00:27,  3.43it/s]
 97%|█████████▋| 359017/371472 [6:24:37<59:45,  3.47it/s]  
 97%|█████████▋| 359018/371472 [6:24:37<58:45,  3.53it/s]
 97%|█████████▋| 359019/371472 [6:24:37<57:45,  3.59it/s]
 97%|█████████▋| 359020/371472 [6:24:37<57:13,  3.63it/s]
                                                         
{'loss': 2.5497, 'learning_rate': 1.3018487793315947e-07, 'epoch': 15.46}
+
 97%|█████████▋| 359020/371472 [6:24:37<57:13,  3.63it/s]
 97%|█████████▋| 359021/371472 [6:24:38<57:38,  3.60it/s]
 97%|█████████▋| 359022/371472 [6:24:38<56:14,  3.69it/s]
 97%|█████████▋| 359023/371472 [6:24:38<54:07,  3.83it/s]
 97%|█████████▋| 359024/371472 [6:24:38<53:47,  3.86it/s]
 97%|█████████▋| 359025/371472 [6:24:39<56:00,  3.70it/s]
 97%|█████████▋| 359026/371472 [6:24:39<55:07,  3.76it/s]
 97%|█████████▋| 359027/371472 [6:24:39<54:23,  3.81it/s]
 97%|█████████▋| 359028/371472 [6:24:39<53:06,  3.91it/s]
 97%|█████████▋| 359029/371472 [6:24:40<52:21,  3.96it/s]
 97%|█████████▋| 359030/371472 [6:24:40<52:14,  3.97it/s]
 97%|█████████▋| 359031/371472 [6:24:40<51:35,  4.02it/s]
 97%|█████████▋| 359032/371472 [6:24:40<54:33,  3.80it/s]
 97%|█████████▋| 359033/371472 [6:24:41<54:10,  3.83it/s]
 97%|█████████▋| 359034/371472 [6:24:41<54:45,  3.79it/s]
 97%|█████████▋| 359035/371472 [6:24:41<58:07,  3.57it/s]
 97%|█████████▋| 359036/371472 [6:24:42<56:21,  3.68it/s]
 97%|█████████▋| 359037/371472 [6:24:42<59:43,  3.47it/s]
 97%|█████████▋| 359038/371472 [6:24:42<57:15,  3.62it/s]
 97%|█████████▋| 359039/371472 [6:24:42<56:13,  3.69it/s]
 97%|█████████▋| 359040/371472 [6:24:43<55:32,  3.73it/s]
                                                         
{'loss': 2.5184, 'learning_rate': 1.3013639595768062e-07, 'epoch': 15.46}
+
 97%|█████████▋| 359040/371472 [6:24:43<55:32,  3.73it/s]
 97%|█████████▋| 359041/371472 [6:24:43<55:29,  3.73it/s]
 97%|█████████▋| 359042/371472 [6:24:43<56:14,  3.68it/s]
 97%|█████████▋| 359043/371472 [6:24:43<54:11,  3.82it/s]
 97%|█████████▋| 359044/371472 [6:24:44<1:00:24,  3.43it/s]
 97%|█████████▋| 359045/371472 [6:24:44<59:47,  3.46it/s]  
 97%|█████████▋| 359046/371472 [6:24:44<1:01:50,  3.35it/s]
 97%|█████████▋| 359047/371472 [6:24:45<1:00:52,  3.40it/s]
 97%|█████████▋| 359048/371472 [6:24:45<59:00,  3.51it/s]  
 97%|█████████▋| 359049/371472 [6:24:45<1:03:08,  3.28it/s]
 97%|█████████▋| 359050/371472 [6:24:46<1:02:58,  3.29it/s]
 97%|█████████▋| 359051/371472 [6:24:46<59:14,  3.49it/s]  
 97%|█████████▋| 359052/371472 [6:24:46<1:02:46,  3.30it/s]
 97%|█████████▋| 359053/371472 [6:24:46<1:00:55,  3.40it/s]
 97%|█████████▋| 359054/371472 [6:24:47<58:35,  3.53it/s]  
 97%|█████████▋| 359055/371472 [6:24:47<1:00:58,  3.39it/s]
 97%|█████████▋| 359056/371472 [6:24:47<1:00:13,  3.44it/s]
 97%|█████████▋| 359057/371472 [6:24:48<59:19,  3.49it/s]  
 97%|█████████▋| 359058/371472 [6:24:48<57:29,  3.60it/s]
 97%|█████████▋| 359059/371472 [6:24:48<56:46,  3.64it/s]
 97%|█████████▋| 359060/371472 [6:24:48<58:23,  3.54it/s]
                                                         
{'loss': 2.4884, 'learning_rate': 1.300879139822017e-07, 'epoch': 15.47}
+
 97%|█████████▋| 359060/371472 [6:24:48<58:23,  3.54it/s]
 97%|█████████▋| 359061/371472 [6:24:49<57:09,  3.62it/s]
 97%|█████████▋| 359062/371472 [6:24:49<56:42,  3.65it/s]
 97%|█████████▋| 359063/371472 [6:24:49<54:13,  3.81it/s]
 97%|█████████▋| 359064/371472 [6:24:49<56:36,  3.65it/s]
 97%|█████████▋| 359065/371472 [6:24:50<55:00,  3.76it/s]
 97%|█████████▋| 359066/371472 [6:24:50<58:31,  3.53it/s]
 97%|█████████▋| 359067/371472 [6:24:50<1:01:31,  3.36it/s]
 97%|█████████▋| 359068/371472 [6:24:51<1:01:18,  3.37it/s]
 97%|█████████▋| 359069/371472 [6:24:51<1:02:11,  3.32it/s]
 97%|█████████▋| 359070/371472 [6:24:51<1:05:07,  3.17it/s]
 97%|█████████▋| 359071/371472 [6:24:52<1:02:14,  3.32it/s]
 97%|█████████▋| 359072/371472 [6:24:52<59:27,  3.48it/s]  
 97%|█████████▋| 359073/371472 [6:24:52<1:00:28,  3.42it/s]
 97%|█████████▋| 359074/371472 [6:24:52<58:33,  3.53it/s]  
 97%|█████████▋| 359075/371472 [6:24:53<58:09,  3.55it/s]
 97%|█████████▋| 359076/371472 [6:24:53<56:36,  3.65it/s]
 97%|█████████▋| 359077/371472 [6:24:53<54:39,  3.78it/s]
 97%|█████████▋| 359078/371472 [6:24:53<53:59,  3.83it/s]
 97%|█████████▋| 359079/371472 [6:24:54<1:01:21,  3.37it/s]
 97%|█████████▋| 359080/371472 [6:24:54<57:31,  3.59it/s]  
                                                         
{'loss': 2.5107, 'learning_rate': 1.3003943200672284e-07, 'epoch': 15.47}
+
 97%|█████████▋| 359080/371472 [6:24:54<57:31,  3.59it/s]
 97%|█████████▋| 359081/371472 [6:24:54<55:52,  3.70it/s]
 97%|█████████▋| 359082/371472 [6:24:55<54:24,  3.80it/s]
 97%|█████████▋| 359083/371472 [6:24:55<54:53,  3.76it/s]
 97%|█████████▋| 359084/371472 [6:24:55<55:53,  3.69it/s]
 97%|█████████▋| 359085/371472 [6:24:55<56:42,  3.64it/s]
 97%|█████████▋| 359086/371472 [6:24:56<56:21,  3.66it/s]
 97%|█████████▋| 359087/371472 [6:24:56<1:03:35,  3.25it/s]
 97%|█████████▋| 359088/371472 [6:24:56<1:05:16,  3.16it/s]
 97%|█████████▋| 359089/371472 [6:24:57<1:02:07,  3.32it/s]
 97%|█████████▋| 359090/371472 [6:24:57<1:00:06,  3.43it/s]
 97%|█████████▋| 359091/371472 [6:24:57<1:02:09,  3.32it/s]
 97%|█████████▋| 359092/371472 [6:24:58<59:10,  3.49it/s]  
 97%|█████████▋| 359093/371472 [6:24:58<59:30,  3.47it/s]
 97%|█████████▋| 359094/371472 [6:24:58<57:09,  3.61it/s]
 97%|█████████▋| 359095/371472 [6:24:58<56:27,  3.65it/s]
 97%|█████████▋| 359096/371472 [6:24:59<58:30,  3.53it/s]
 97%|█████████▋| 359097/371472 [6:24:59<57:46,  3.57it/s]
 97%|█████████▋| 359098/371472 [6:24:59<57:21,  3.60it/s]
 97%|███████��█▋| 359099/371472 [6:24:59<55:59,  3.68it/s]
 97%|█████████▋| 359100/371472 [6:25:00<54:14,  3.80it/s]
                                                         
{'loss': 2.5366, 'learning_rate': 1.299909500312439e-07, 'epoch': 15.47}
+
 97%|█████████▋| 359100/371472 [6:25:00<54:14,  3.80it/s]
 97%|█████████▋| 359101/371472 [6:25:00<55:28,  3.72it/s]
 97%|█████████▋| 359102/371472 [6:25:00<56:33,  3.65it/s]
 97%|█████████▋| 359103/371472 [6:25:00<54:09,  3.81it/s]
 97%|█████████▋| 359104/371472 [6:25:01<55:19,  3.73it/s]
 97%|█████████▋| 359105/371472 [6:25:01<54:41,  3.77it/s]
 97%|█████████▋| 359106/371472 [6:25:01<1:01:41,  3.34it/s]
 97%|█████████▋| 359107/371472 [6:25:02<1:01:38,  3.34it/s]
 97%|█████████▋| 359108/371472 [6:25:02<59:22,  3.47it/s]  
 97%|█████████▋| 359109/371472 [6:25:02<56:43,  3.63it/s]
 97%|█████████▋| 359110/371472 [6:25:02<57:04,  3.61it/s]
 97%|█████████▋| 359111/371472 [6:25:03<55:14,  3.73it/s]
 97%|█████████▋| 359112/371472 [6:25:03<54:27,  3.78it/s]
 97%|█████████▋| 359113/371472 [6:25:03<58:11,  3.54it/s]
 97%|█████████▋| 359114/371472 [6:25:04<58:08,  3.54it/s]
 97%|█████████▋| 359115/371472 [6:25:04<58:38,  3.51it/s]
 97%|█████████▋| 359116/371472 [6:25:04<55:44,  3.69it/s]
 97%|█████████▋| 359117/371472 [6:25:04<54:06,  3.81it/s]
 97%|█████████▋| 359118/371472 [6:25:05<55:15,  3.73it/s]
 97%|█████████▋| 359119/371472 [6:25:05<54:55,  3.75it/s]
 97%|█████████▋| 359120/371472 [6:25:05<56:35,  3.64it/s]
                                                         
{'loss': 2.7086, 'learning_rate': 1.2994246805576507e-07, 'epoch': 15.47}
+
 97%|█████████▋| 359120/371472 [6:25:05<56:35,  3.64it/s]
 97%|█████████▋| 359121/371472 [6:25:06<59:28,  3.46it/s]
 97%|█████████▋| 359122/371472 [6:25:06<59:25,  3.46it/s]
 97%|█████████▋| 359123/371472 [6:25:06<58:22,  3.53it/s]
 97%|█████████▋| 359124/371472 [6:25:06<58:35,  3.51it/s]
 97%|█████████▋| 359125/371472 [6:25:07<56:44,  3.63it/s]
 97%|█████████▋| 359126/371472 [6:25:07<56:15,  3.66it/s]
 97%|█████████▋| 359127/371472 [6:25:07<57:28,  3.58it/s]
 97%|█████████▋| 359128/371472 [6:25:08<1:03:37,  3.23it/s]
 97%|█████████▋| 359129/371472 [6:25:08<1:02:27,  3.29it/s]
 97%|█████████▋| 359130/371472 [6:25:08<59:41,  3.45it/s]  
 97%|█████████▋| 359131/371472 [6:25:08<1:02:31,  3.29it/s]
 97%|█████████▋| 359132/371472 [6:25:09<58:30,  3.52it/s]  
 97%|█████████▋| 359133/371472 [6:25:09<1:01:42,  3.33it/s]
 97%|█████████▋| 359134/371472 [6:25:09<1:00:07,  3.42it/s]
 97%|█████████▋| 359135/371472 [6:25:10<57:07,  3.60it/s]  
 97%|█████████▋| 359136/371472 [6:25:10<56:07,  3.66it/s]
 97%|█████████▋| 359137/371472 [6:25:10<56:07,  3.66it/s]
 97%|█████████▋| 359138/371472 [6:25:10<54:54,  3.74it/s]
 97%|█████████▋| 359139/371472 [6:25:11<57:58,  3.55it/s]
 97%|█████████▋| 359140/371472 [6:25:11<58:20,  3.52it/s]
                                                         
{'loss': 2.4984, 'learning_rate': 1.298939860802861e-07, 'epoch': 15.47}
+
 97%|█████████▋| 359140/371472 [6:25:11<58:20,  3.52it/s]
 97%|█████████▋| 359141/371472 [6:25:11<57:06,  3.60it/s]
 97%|█████████▋| 359142/371472 [6:25:11<57:00,  3.60it/s]
 97%|█████████▋| 359143/371472 [6:25:12<55:16,  3.72it/s]
 97%|█████████▋| 359144/371472 [6:25:12<1:00:59,  3.37it/s]
 97%|█████████▋| 359145/371472 [6:25:12<1:00:25,  3.40it/s]
 97%|█████████▋| 359146/371472 [6:25:13<58:02,  3.54it/s]  
 97%|█████████▋| 359147/371472 [6:25:13<59:10,  3.47it/s]
 97%|█████████▋| 359148/371472 [6:25:13<1:01:06,  3.36it/s]
 97%|█████████▋| 359149/371472 [6:25:14<59:08,  3.47it/s]  
 97%|█████████▋| 359150/371472 [6:25:14<58:45,  3.50it/s]
 97%|█████████▋| 359151/371472 [6:25:14<57:25,  3.58it/s]
 97%|█████████▋| 359152/371472 [6:25:14<56:38,  3.62it/s]
 97%|█████████▋| 359153/371472 [6:25:15<56:27,  3.64it/s]
 97%|█████████▋| 359154/371472 [6:25:15<59:18,  3.46it/s]
 97%|█████████▋| 359155/371472 [6:25:15<1:02:34,  3.28it/s]
 97%|█████████��| 359156/371472 [6:25:16<59:08,  3.47it/s]  
 97%|█████████▋| 359157/371472 [6:25:16<58:24,  3.51it/s]
 97%|█████████▋| 359158/371472 [6:25:16<57:44,  3.55it/s]
 97%|█████████▋| 359159/371472 [6:25:16<1:00:59,  3.36it/s]
 97%|█████████▋| 359160/371472 [6:25:17<1:02:54,  3.26it/s]
                                                           
{'loss': 2.376, 'learning_rate': 1.2984550410480726e-07, 'epoch': 15.47}
+
 97%|█████████▋| 359160/371472 [6:25:17<1:02:54,  3.26it/s]
 97%|█████████▋| 359161/371472 [6:25:17<58:33,  3.50it/s]  
 97%|█████████▋| 359162/371472 [6:25:17<57:51,  3.55it/s]
 97%|█████████▋| 359163/371472 [6:25:18<58:31,  3.51it/s]
 97%|█████████▋| 359164/371472 [6:25:18<56:42,  3.62it/s]
 97%|█████████▋| 359165/371472 [6:25:18<55:46,  3.68it/s]
 97%|█████████▋| 359166/371472 [6:25:18<55:25,  3.70it/s]
 97%|█████████▋| 359167/371472 [6:25:19<53:50,  3.81it/s]
 97%|█████████▋| 359168/371472 [6:25:19<51:46,  3.96it/s]
 97%|█████████▋| 359169/371472 [6:25:19<54:38,  3.75it/s]
 97%|█████████▋| 359170/371472 [6:25:19<55:08,  3.72it/s]
 97%|█████████▋| 359171/371472 [6:25:20<54:41,  3.75it/s]
 97%|█████████▋| 359172/371472 [6:25:20<54:14,  3.78it/s]
 97%|█████████▋| 359173/371472 [6:25:20<53:37,  3.82it/s]
 97%|█████████▋| 359174/371472 [6:25:20<55:21,  3.70it/s]
 97%|█████████▋| 359175/371472 [6:25:21<53:41,  3.82it/s]
 97%|█████████▋| 359176/371472 [6:25:21<55:14,  3.71it/s]
 97%|█████████▋| 359177/371472 [6:25:21<1:01:15,  3.35it/s]
 97%|█████████▋| 359178/371472 [6:25:22<58:38,  3.49it/s]  
 97%|█████████▋| 359179/371472 [6:25:22<57:15,  3.58it/s]
 97%|█████████▋| 359180/371472 [6:25:22<59:47,  3.43it/s]
                                                         
{'loss': 2.6673, 'learning_rate': 1.2979702212932833e-07, 'epoch': 15.47}
+
 97%|█████████▋| 359180/371472 [6:25:22<59:47,  3.43it/s]
 97%|█████████▋| 359181/371472 [6:25:22<59:49,  3.42it/s]
 97%|█████████▋| 359182/371472 [6:25:23<57:58,  3.53it/s]
 97%|█████████▋| 359183/371472 [6:25:23<59:18,  3.45it/s]
 97%|█████████▋| 359184/371472 [6:25:23<57:33,  3.56it/s]
 97%|█████████▋| 359185/371472 [6:25:24<1:02:26,  3.28it/s]
 97%|█████████▋| 359186/371472 [6:25:24<1:00:21,  3.39it/s]
 97%|█████████▋| 359187/371472 [6:25:24<59:13,  3.46it/s]  
 97%|█████████▋| 359188/371472 [6:25:24<56:58,  3.59it/s]
 97%|█████████▋| 359189/371472 [6:25:25<1:00:09,  3.40it/s]
 97%|█████████▋| 359190/371472 [6:25:25<58:57,  3.47it/s]  
 97%|█████████▋| 359191/371472 [6:25:25<56:47,  3.60it/s]
 97%|█████████▋| 359192/371472 [6:25:26<57:25,  3.56it/s]
 97%|█████████▋| 359193/371472 [6:25:26<56:54,  3.60it/s]
 97%|█████████▋| 359194/371472 [6:25:26<56:11,  3.64it/s]
 97%|█████████▋| 359195/371472 [6:25:26<59:08,  3.46it/s]
 97%|█████████▋| 359196/371472 [6:25:27<57:39,  3.55it/s]
 97%|█████████▋| 359197/371472 [6:25:27<56:06,  3.65it/s]
 97%|█████████▋| 359198/371472 [6:25:27<55:42,  3.67it/s]
 97%|█████████▋| 359199/371472 [6:25:28<55:44,  3.67it/s]
 97%|█████████▋| 359200/371472 [6:25:28<55:32,  3.68it/s]
                                                         
{'loss': 2.5011, 'learning_rate': 1.2974854015384948e-07, 'epoch': 15.47}
+
 97%|█████████▋| 359200/371472 [6:25:28<55:32,  3.68it/s]
 97%|█████████▋| 359201/371472 [6:25:28<57:45,  3.54it/s]
 97%|█████████▋| 359202/371472 [6:25:28<1:00:24,  3.39it/s]
 97%|█████████▋| 359203/371472 [6:25:29<59:17,  3.45it/s]  
 97%|█████████▋| 359204/371472 [6:25:29<57:54,  3.53it/s]
 97%|█████████▋| 359205/371472 [6:25:29<59:02,  3.46it/s]
 97%|█████████▋| 359206/371472 [6:25:30<58:08,  3.52it/s]
 97%|█████████▋| 359207/371472 [6:25:30<56:38,  3.61it/s]
 97%|█████████▋| 359208/371472 [6:25:30<56:30,  3.62it/s]
 97%|█████████▋| 359209/371472 [6:25:30<59:16,  3.45it/s]
 97%|█████████▋| 359210/371472 [6:25:31<58:14,  3.51it/s]
 97%|█████████▋| 359211/371472 [6:25:31<1:00:14,  3.39it/s]
 97%|█████████▋| 359212/371472 [6:25:31<58:13,  3.51it/s]  
 97%|█████████▋| 359213/371472 [6:25:32<56:15,  3.63it/s]
 97%|█████████▋| 359214/371472 [6:25:32<54:29,  3.75it/s]
 97%|█████████▋| 359215/371472 [6:25:32<54:37,  3.74it/s]
 97%|█████████▋| 359216/371472 [6:25:32<54:44,  3.73it/s]
 97%|█████████▋| 359217/371472 [6:25:33<58:15,  3.51it/s]
 97%|█████████▋| 359218/371472 [6:25:33<1:00:39,  3.37it/s]
 97%|█████████▋| 359219/371472 [6:25:33<57:56,  3.52it/s]  
 97%|█████████▋| 359220/371472 [6:25:33<57:46,  3.53it/s]
                                                         
{'loss': 2.7215, 'learning_rate': 1.2970005817837053e-07, 'epoch': 15.47}
+
 97%|█████████▋| 359220/371472 [6:25:33<57:46,  3.53it/s]
 97%|█████████▋| 359221/371472 [6:25:34<57:09,  3.57it/s]
 97%|█████████▋| 359222/371472 [6:25:34<56:53,  3.59it/s]
 97%|█████████▋| 359223/371472 [6:25:34<57:35,  3.54it/s]
 97%|█████████▋| 359224/371472 [6:25:35<56:32,  3.61it/s]
 97%|█████████▋| 359225/371472 [6:25:35<54:58,  3.71it/s]
 97%|█████████▋| 359226/371472 [6:25:35<55:54,  3.65it/s]
 97%|█████████▋| 359227/371472 [6:25:35<54:02,  3.78it/s]
 97%|█████████▋| 359228/371472 [6:25:36<53:58,  3.78it/s]
 97%|█████████▋| 359229/371472 [6:25:36<55:30,  3.68it/s]
 97%|█████████▋| 359230/371472 [6:25:36<56:23,  3.62it/s]
 97%|█████████▋| 359231/371472 [6:25:36<57:02,  3.58it/s]
 97%|█████████▋| 359232/371472 [6:25:37<55:19,  3.69it/s]
 97%|█████████▋| 359233/371472 [6:25:37<56:00,  3.64it/s]
 97%|█████████▋| 359234/371472 [6:25:37<56:20,  3.62it/s]
 97%|█████████▋| 359235/371472 [6:25:38<53:53,  3.78it/s]
 97%|█████████▋| 359236/371472 [6:25:38<53:19,  3.82it/s]
 97%|█████████▋| 359237/371472 [6:25:38<55:43,  3.66it/s]
 97%|█████████▋| 359238/371472 [6:25:38<53:31,  3.81it/s]
 97%|█████████▋| 359239/371472 [6:25:39<54:12,  3.76it/s]
 97%|█████████▋| 359240/371472 [6:25:39<56:51,  3.59it/s]
                                                         
{'loss': 2.6141, 'learning_rate': 1.296515762028917e-07, 'epoch': 15.47}
+
 97%|█████████▋| 359240/371472 [6:25:39<56:51,  3.59it/s]
 97%|█████████▋| 359241/371472 [6:25:39<56:26,  3.61it/s]
 97%|█████████▋| 359242/371472 [6:25:39<55:47,  3.65it/s]
 97%|█████████▋| 359243/371472 [6:25:40<57:30,  3.54it/s]
 97%|█████████▋| 359244/371472 [6:25:40<59:35,  3.42it/s]
 97%|█████████▋| 359245/371472 [6:25:40<57:08,  3.57it/s]
 97%|█████████▋| 359246/371472 [6:25:41<54:30,  3.74it/s]
 97%|█████████▋| 359247/371472 [6:25:41<56:48,  3.59it/s]
 97%|█████████▋| 359248/371472 [6:25:41<56:14,  3.62it/s]
 97%|█████████▋| 359249/371472 [6:25:42<1:02:27,  3.26it/s]
 97%|█████████▋| 359250/371472 [6:25:42<1:00:34,  3.36it/s]
 97%|█████████▋| 359251/371472 [6:25:42<58:18,  3.49it/s]  
 97%|█████████▋| 359252/371472 [6:25:42<58:00,  3.51it/s]
 97%|█████████▋| 359253/371472 [6:25:43<56:57,  3.58it/s]
 97%|█████████▋| 359254/371472 [6:25:43<56:13,  3.62it/s]
 97%|█████████▋| 359255/371472 [6:25:43<54:50,  3.71it/s]
 97%|█████████▋| 359256/371472 [6:25:43<54:22,  3.74it/s]
 97%|█████████▋| 359257/371472 [6:25:44<56:00,  3.63it/s]
 97%|█████████▋| 359258/371472 [6:25:44<55:10,  3.69it/s]
 97%|█████████▋| 359259/371472 [6:25:44<55:05,  3.69it/s]
 97%|█████████▋| 359260/371472 [6:25:44<55:52,  3.64it/s]
                                                         
{'loss': 2.5615, 'learning_rate': 1.2960309422741275e-07, 'epoch': 15.47}
+
 97%|█████████▋| 359260/371472 [6:25:44<55:52,  3.64it/s]
 97%|█████████▋| 359261/371472 [6:25:45<53:46,  3.78it/s]
 97%|█████████▋| 359262/371472 [6:25:45<52:54,  3.85it/s]
 97%|█████████▋| 359263/371472 [6:25:45<59:15,  3.43it/s]
 97%|█████████▋| 359264/371472 [6:25:46<57:51,  3.52it/s]
 97%|█████████▋| 359265/371472 [6:25:46<55:47,  3.65it/s]
 97%|█████████▋| 359266/371472 [6:25:46<55:37,  3.66it/s]
 97%|█████████▋| 359267/371472 [6:25:46<59:22,  3.43it/s]
 97%|█████████▋| 359268/371472 [6:25:47<58:27,  3.48it/s]
 97%|█████████▋| 359269/371472 [6:25:47<57:09,  3.56it/s]
 97%|█████████▋| 359270/371472 [6:25:47<55:59,  3.63it/s]
 97%|█████████▋| 359271/371472 [6:25:48<1:00:20,  3.37it/s]
 97%|█████████▋| 359272/371472 [6:25:48<58:53,  3.45it/s]  
 97%|█████████▋| 359273/371472 [6:25:48<56:54,  3.57it/s]
 97%|█████████▋| 359274/371472 [6:25:48<55:40,  3.65it/s]
 97%|█████████▋| 359275/371472 [6:25:49<54:00,  3.76it/s]
 97%|█████████▋| 359276/371472 [6:25:49<53:05,  3.83it/s]
 97%|█████████▋| 359277/371472 [6:25:49<59:26,  3.42it/s]
 97%|█████████▋| 359278/371472 [6:25:50<1:01:22,  3.31it/s]
 97%|█████████▋| 359279/371472 [6:25:50<1:00:21,  3.37it/s]
 97%|█████████▋| 359280/371472 [6:25:50<59:44,  3.40it/s]  
                                                         
{'loss': 2.463, 'learning_rate': 1.295546122519339e-07, 'epoch': 15.47}
+
 97%|█████████▋| 359280/371472 [6:25:50<59:44,  3.40it/s]
 97%|█████████▋| 359281/371472 [6:25:50<57:54,  3.51it/s]
 97%|█████████▋| 359282/371472 [6:25:51<55:44,  3.64it/s]
 97%|█████████▋| 359283/371472 [6:25:51<58:28,  3.47it/s]
 97%|█████████▋| 359284/371472 [6:25:51<55:49,  3.64it/s]
 97%|█████████▋| 359285/371472 [6:25:52<54:31,  3.73it/s]
 97%|█████████▋| 359286/371472 [6:25:52<1:00:49,  3.34it/s]
 97%|█████████▋| 359287/371472 [6:25:52<59:59,  3.39it/s]  
 97%|█████████▋| 359288/371472 [6:25:52<57:34,  3.53it/s]
 97%|█████████▋| 359289/371472 [6:25:53<55:28,  3.66it/s]
 97%|█████████▋| 359290/371472 [6:25:53<59:48,  3.39it/s]
 97%|█████████▋| 359291/371472 [6:25:53<59:41,  3.40it/s]
 97%|█████████▋| 359292/371472 [6:25:54<56:44,  3.58it/s]
 97%|█████████▋| 359293/371472 [6:25:54<56:19,  3.60it/s]
 97%|█████████▋| 359294/371472 [6:25:54<58:24,  3.47it/s]
 97%|█████████▋| 359295/371472 [6:25:54<55:58,  3.63it/s]
 97%|█████████▋| 359296/371472 [6:25:55<58:15,  3.48it/s]
 97%|█████████▋| 359297/371472 [6:25:55<57:57,  3.50it/s]
 97%|█████████▋| 359298/371472 [6:25:55<1:01:25,  3.30it/s]
 97%|█████████▋| 359299/371472 [6:25:56<1:00:36,  3.35it/s]
 97%|█████████▋| 359300/371472 [6:25:56<59:53,  3.39it/s]  
                                                         
{'loss': 2.5259, 'learning_rate': 1.2950613027645497e-07, 'epoch': 15.48}
+
 97%|█████████▋| 359300/371472 [6:25:56<59:53,  3.39it/s]
 97%|█████████▋| 359301/371472 [6:25:56<1:00:37,  3.35it/s]
 97%|█████████▋| 359302/371472 [6:25:56<58:13,  3.48it/s]  
 97%|█████████▋| 359303/371472 [6:25:57<56:56,  3.56it/s]
 97%|█████████▋| 359304/371472 [6:25:57<57:01,  3.56it/s]
 97%|█████████▋| 359305/371472 [6:25:57<59:09,  3.43it/s]
 97%|█████████▋| 359306/371472 [6:25:58<58:25,  3.47it/s]
 97%|█████████▋| 359307/371472 [6:25:58<57:26,  3.53it/s]
 97%|█████████▋| 359308/371472 [6:25:58<55:54,  3.63it/s]
 97%|█████████▋| 359309/371472 [6:25:58<55:30,  3.65it/s]
 97%|█████████▋| 359310/371472 [6:25:59<59:39,  3.40it/s]
 97%|█████████▋| 359311/371472 [6:25:59<59:17,  3.42it/s]
 97%|█████████▋| 359312/371472 [6:25:59<1:00:52,  3.33it/s]
 97%|█████████▋| 359313/371472 [6:26:00<57:48,  3.51it/s]  
 97%|█████████▋| 359314/371472 [6:26:00<55:47,  3.63it/s]
 97%|█████████▋| 359315/371472 [6:26:00<55:54,  3.62it/s]
 97%|█████████▋| 359316/371472 [6:26:00<55:04,  3.68it/s]
 97%|█████████▋| 359317/371472 [6:26:01<56:09,  3.61it/s]
 97%|█████████▋| 359318/371472 [6:26:01<55:02,  3.68it/s]
 97%|█████████▋| 359319/371472 [6:26:01<57:23,  3.53it/s]
 97%|█████████▋| 359320/371472 [6:26:02<54:49,  3.69it/s]
                                                         
{'loss': 2.5497, 'learning_rate': 1.2945764830097613e-07, 'epoch': 15.48}
+
 97%|█████████▋| 359320/371472 [6:26:02<54:49,  3.69it/s]
 97%|█████████▋| 359321/371472 [6:26:02<1:01:31,  3.29it/s]
 97%|█████████▋| 359322/371472 [6:26:02<59:42,  3.39it/s]  
 97%|█████████▋| 359323/371472 [6:26:02<57:30,  3.52it/s]
 97%|█████████▋| 359324/371472 [6:26:03<55:56,  3.62it/s]
 97%|█████████▋| 359325/371472 [6:26:03<54:38,  3.70it/s]
 97%|█████████▋| 359326/371472 [6:26:03<54:01,  3.75it/s]
 97%|█████████▋| 359327/371472 [6:26:03<53:47,  3.76it/s]
 97%|█████████▋| 359328/371472 [6:26:04<53:51,  3.76it/s]
 97%|█████████▋| 359329/371472 [6:26:04<57:07,  3.54it/s]
 97%|█████████▋| 359330/371472 [6:26:04<56:42,  3.57it/s]
 97%|█████████▋| 359331/371472 [6:26:05<54:57,  3.68it/s]
 97%|█████████▋| 359332/371472 [6:26:05<56:14,  3.60it/s]
 97%|█████████▋| 359333/371472 [6:26:05<55:33,  3.64it/s]
 97%|█████████▋| 359334/371472 [6:26:05<54:19,  3.72it/s]
 97%|█████████▋| 359335/371472 [6:26:06<54:13,  3.73it/s]
 97%|█████████▋| 359336/371472 [6:26:06<55:45,  3.63it/s]
 97%|█████████▋| 359337/371472 [6:26:06<54:56,  3.68it/s]
 97%|█████████▋| 359338/371472 [6:26:06<55:51,  3.62it/s]
 97%|█████████▋| 359339/371472 [6:26:07<59:46,  3.38it/s]
 97%|█████████▋| 359340/371472 [6:26:07<1:00:32,  3.34it/s]
                                                           
{'loss': 2.5841, 'learning_rate': 1.294091663254972e-07, 'epoch': 15.48}
+
 97%|█████████▋| 359340/371472 [6:26:07<1:00:32,  3.34it/s]
 97%|█████████▋| 359341/371472 [6:26:07<59:53,  3.38it/s]  
 97%|█████████▋| 359342/371472 [6:26:08<58:25,  3.46it/s]
 97%|█████████▋| 359343/371472 [6:26:08<1:04:41,  3.13it/s]
 97%|█████████▋| 359344/371472 [6:26:08<1:02:29,  3.23it/s]
 97%|█████████▋| 359345/371472 [6:26:09<59:57,  3.37it/s]  
 97%|█████████▋| 359346/371472 [6:26:09<1:00:25,  3.35it/s]
 97%|█████████▋| 359347/371472 [6:26:09<57:48,  3.50it/s]  
 97%|█████████▋| 359348/371472 [6:26:09<56:59,  3.55it/s]
 97%|█████████▋| 359349/371472 [6:26:10<59:42,  3.38it/s]
 97%|█████████▋| 359350/371472 [6:26:10<1:03:28,  3.18it/s]
 97%|█████████▋| 359351/371472 [6:26:10<1:02:47,  3.22it/s]
 97%|█████████▋| 359352/371472 [6:26:11<1:00:08,  3.36it/s]
 97%|█████████▋| 359353/371472 [6:26:11<59:11,  3.41it/s]  
 97%|█████████▋| 359354/371472 [6:26:11<59:13,  3.41it/s]
 97%|█████████▋| 359355/371472 [6:26:12<56:29,  3.57it/s]
 97%|█████████▋| 359356/371472 [6:26:12<54:21,  3.71it/s]
 97%|█████████▋| 359357/371472 [6:26:12<56:19,  3.58it/s]
 97%|█████████▋| 359358/371472 [6:26:12<58:38,  3.44it/s]
 97%|█████████▋| 359359/371472 [6:26:13<58:32,  3.45it/s]
 97%|█████████▋| 359360/371472 [6:26:13<56:11,  3.59it/s]
                                                         
{'loss': 2.6298, 'learning_rate': 1.2936068435001835e-07, 'epoch': 15.48}
+
 97%|█████████▋| 359360/371472 [6:26:13<56:11,  3.59it/s]
 97%|█████████▋| 359361/371472 [6:26:13<54:14,  3.72it/s]
 97%|█████████▋| 359362/371472 [6:26:14<59:06,  3.42it/s]
 97%|█████████▋| 359363/371472 [6:26:14<59:06,  3.41it/s]
 97%|█████████▋| 359364/371472 [6:26:14<57:04,  3.54it/s]
 97%|█████████▋| 359365/371472 [6:26:14<55:24,  3.64it/s]
 97%|█████████▋| 359366/371472 [6:26:15<53:25,  3.78it/s]
 97%|█████████▋| 359367/371472 [6:26:15<52:33,  3.84it/s]
 97%|█████████▋| 359368/371472 [6:26:15<51:47,  3.90it/s]
 97%|█████████▋| 359369/371472 [6:26:15<52:51,  3.82it/s]
 97%|█████████▋| 359370/371472 [6:26:16<51:22,  3.93it/s]
 97%|█████████▋| 359371/371472 [6:26:16<51:54,  3.89it/s]
 97%|█████████▋| 359372/371472 [6:26:16<54:52,  3.67it/s]
 97%|█████████▋| 359373/371472 [6:26:16<57:35,  3.50it/s]
 97%|█████████▋| 359374/371472 [6:26:17<56:03,  3.60it/s]
 97%|█████████▋| 359375/371472 [6:26:17<1:00:34,  3.33it/s]
 97%|█████████▋| 359376/371472 [6:26:17<57:11,  3.52it/s]  
 97%|█████████▋| 359377/371472 [6:26:18<57:21,  3.51it/s]
 97%|█████████▋| 359378/371472 [6:26:18<55:57,  3.60it/s]
 97%|█████████▋| 359379/371472 [6:26:18<57:13,  3.52it/s]
 97%|█████████▋| 359380/371472 [6:26:19<1:02:21,  3.23it/s]
                                                           
{'loss': 2.632, 'learning_rate': 1.293122023745394e-07, 'epoch': 15.48}
+
 97%|█████████▋| 359380/371472 [6:26:19<1:02:21,  3.23it/s]
 97%|█████████▋| 359381/371472 [6:26:19<59:51,  3.37it/s]  
 97%|█████████▋| 359382/371472 [6:26:19<56:28,  3.57it/s]
 97%|█████████▋| 359383/371472 [6:26:19<58:40,  3.43it/s]
 97%|█████████▋| 359384/371472 [6:26:20<56:48,  3.55it/s]
 97%|█████████▋| 359385/371472 [6:26:20<56:37,  3.56it/s]
 97%|█████████▋| 359386/371472 [6:26:20<1:01:24,  3.28it/s]
 97%|█████████▋| 359387/371472 [6:26:21<58:43,  3.43it/s]  
 97%|█████████▋| 359388/371472 [6:26:21<1:06:54,  3.01it/s]
 97%|█████████▋| 359389/371472 [6:26:21<1:02:40,  3.21it/s]
 97%|█████████▋| 359390/371472 [6:26:22<59:06,  3.41it/s]  
 97%|█████████▋| 359391/371472 [6:26:22<56:20,  3.57it/s]
 97%|█████████▋| 359392/371472 [6:26:22<56:39,  3.55it/s]
 97%|█████████▋| 359393/371472 [6:26:22<57:05,  3.53it/s]
 97%|█████████▋| 359394/371472 [6:26:23<58:00,  3.47it/s]
 97%|█████████▋| 359395/371472 [6:26:23<57:46,  3.48it/s]
 97%|█████████▋| 359396/371472 [6:26:23<1:01:38,  3.27it/s]
 97%|█████████▋| 359397/371472 [6:26:24<1:01:06,  3.29it/s]
 97%|█████████▋| 359398/371472 [6:26:24<59:14,  3.40it/s]  
 97%|█████████▋| 359399/371472 [6:26:24<1:01:38,  3.26it/s]
 97%|█████████▋| 359400/371472 [6:26:24<58:57,  3.41it/s]  
                                                         
{'loss': 2.5546, 'learning_rate': 1.2926372039906057e-07, 'epoch': 15.48}
+
 97%|█████████▋| 359400/371472 [6:26:24<58:57,  3.41it/s]
 97%|█████████▋| 359401/371472 [6:26:25<58:24,  3.44it/s]
 97%|█████████▋| 359402/371472 [6:26:25<59:45,  3.37it/s]
 97%|█████████▋| 359403/371472 [6:26:25<59:58,  3.35it/s]
 97%|█████████▋| 359404/371472 [6:26:26<57:41,  3.49it/s]
 97%|█████████▋| 359405/371472 [6:26:26<56:20,  3.57it/s]
 97%|█████████▋| 359406/371472 [6:26:26<55:19,  3.63it/s]
 97%|█████████▋| 359407/371472 [6:26:26<53:41,  3.75it/s]
 97%|█████████▋| 359408/371472 [6:26:27<55:22,  3.63it/s]
 97%|█████████▋| 359409/371472 [6:26:27<55:36,  3.62it/s]
 97%|█████████▋| 359410/371472 [6:26:27<54:25,  3.69it/s]
 97%|█████████▋| 359411/371472 [6:26:27<54:41,  3.68it/s]
 97%|█████████▋| 359412/371472 [6:26:28<57:53,  3.47it/s]
 97%|█████████▋| 359413/371472 [6:26:28<55:44,  3.61it/s]
 97%|█████████▋| 359414/371472 [6:26:28<54:26,  3.69it/s]
 97%|█████████▋| 359415/371472 [6:26:29<58:59,  3.41it/s]
 97%|█████████▋| 359416/371472 [6:26:29<59:04,  3.40it/s]
 97%|█████████▋| 359417/371472 [6:26:29<56:58,  3.53it/s]
 97%|█████████▋| 359418/371472 [6:26:29<56:03,  3.58it/s]
 97%|█████████▋| 359419/371472 [6:26:30<58:26,  3.44it/s]
 97%|█████████▋| 359420/371472 [6:26:30<56:53,  3.53it/s]
                                                         
{'loss': 2.4961, 'learning_rate': 1.2921523842358161e-07, 'epoch': 15.48}
+
 97%|█████████▋| 359420/371472 [6:26:30<56:53,  3.53it/s]
 97%|█████████▋| 359421/371472 [6:26:30<55:14,  3.64it/s]
 97%|█████████▋| 359422/371472 [6:26:31<54:58,  3.65it/s]
 97%|█████████▋| 359423/371472 [6:26:31<56:23,  3.56it/s]
 97%|█████████▋| 359424/371472 [6:26:31<58:52,  3.41it/s]
 97%|█████████▋| 359425/371472 [6:26:31<57:17,  3.50it/s]
 97%|█████████▋| 359426/371472 [6:26:32<57:21,  3.50it/s]
 97%|█████████▋| 359427/371472 [6:26:32<59:19,  3.38it/s]
 97%|█████████▋| 359428/371472 [6:26:32<59:07,  3.39it/s]
 97%|█████████▋| 359429/371472 [6:26:33<57:45,  3.47it/s]
 97%|█████████▋| 359430/371472 [6:26:33<55:21,  3.63it/s]
 97%|█████████▋| 359431/371472 [6:26:33<56:03,  3.58it/s]
 97%|█████████▋| 359432/371472 [6:26:33<53:48,  3.73it/s]
 97%|█████████▋| 359433/371472 [6:26:34<52:19,  3.83it/s]
 97%|█████████▋| 359434/371472 [6:26:34<1:00:15,  3.33it/s]
 97%|█████████▋| 359435/371472 [6:26:34<56:59,  3.52it/s]  
 97%|█████████▋| 359436/371472 [6:26:35<55:18,  3.63it/s]
 97%|█████████▋| 359437/371472 [6:26:35<54:00,  3.71it/s]
 97%|█████████▋| 359438/371472 [6:26:35<54:32,  3.68it/s]
 97%|█████████▋| 359439/371472 [6:26:35<58:18,  3.44it/s]
 97%|█████████▋| 359440/371472 [6:26:36<56:33,  3.55it/s]
                                                         
{'loss': 2.5815, 'learning_rate': 1.2916675644810277e-07, 'epoch': 15.48}
+
 97%|█████████▋| 359440/371472 [6:26:36<56:33,  3.55it/s]
 97%|█████████▋| 359441/371472 [6:26:36<56:54,  3.52it/s]
 97%|███���█████▋| 359442/371472 [6:26:36<1:01:40,  3.25it/s]
 97%|█████████▋| 359443/371472 [6:26:37<59:06,  3.39it/s]  
 97%|█████████▋| 359444/371472 [6:26:37<58:27,  3.43it/s]
 97%|█████████▋| 359445/371472 [6:26:37<59:25,  3.37it/s]
 97%|█████████▋| 359446/371472 [6:26:37<59:00,  3.40it/s]
 97%|█████████▋| 359447/371472 [6:26:38<56:51,  3.52it/s]
 97%|█████████▋| 359448/371472 [6:26:38<55:51,  3.59it/s]
 97%|█████████▋| 359449/371472 [6:26:38<53:17,  3.76it/s]
 97%|█████████▋| 359450/371472 [6:26:39<54:07,  3.70it/s]
 97%|█████████▋| 359451/371472 [6:26:39<54:35,  3.67it/s]
 97%|█████████▋| 359452/371472 [6:26:39<53:19,  3.76it/s]
 97%|█████████▋| 359453/371472 [6:26:39<52:58,  3.78it/s]
 97%|█████████▋| 359454/371472 [6:26:40<56:03,  3.57it/s]
 97%|█████████▋| 359455/371472 [6:26:40<57:11,  3.50it/s]
 97%|█████████▋| 359456/371472 [6:26:40<55:40,  3.60it/s]
 97%|█████████▋| 359457/371472 [6:26:41<1:07:25,  2.97it/s]
 97%|█████████▋| 359458/371472 [6:26:41<1:03:02,  3.18it/s]
 97%|█████████▋| 359459/371472 [6:26:41<1:01:10,  3.27it/s]
 97%|█████████▋| 359460/371472 [6:26:42<1:00:44,  3.30it/s]
                                                           
{'loss': 2.7068, 'learning_rate': 1.2911827447262384e-07, 'epoch': 15.48}
+
 97%|█████████▋| 359460/371472 [6:26:42<1:00:44,  3.30it/s]
 97%|█████████▋| 359461/371472 [6:26:42<57:44,  3.47it/s]  
 97%|█████████▋| 359462/371472 [6:26:42<58:38,  3.41it/s]
 97%|█████████▋| 359463/371472 [6:26:42<57:00,  3.51it/s]
 97%|█████████▋| 359464/371472 [6:26:43<56:37,  3.53it/s]
 97%|█████████▋| 359465/371472 [6:26:43<55:49,  3.58it/s]
 97%|█████████▋| 359466/371472 [6:26:43<55:12,  3.62it/s]
 97%|█████████▋| 359467/371472 [6:26:43<55:05,  3.63it/s]
 97%|█████████▋| 359468/371472 [6:26:44<55:05,  3.63it/s]
 97%|█████████▋| 359469/371472 [6:26:44<53:36,  3.73it/s]
 97%|█████████▋| 359470/371472 [6:26:44<52:56,  3.78it/s]
 97%|█████████▋| 359471/371472 [6:26:44<54:12,  3.69it/s]
 97%|█████████▋| 359472/371472 [6:26:45<54:34,  3.66it/s]
 97%|█████████▋| 359473/371472 [6:26:45<59:28,  3.36it/s]
 97%|█████████▋| 359474/371472 [6:26:45<57:37,  3.47it/s]
 97%|█████████▋| 359475/371472 [6:26:46<57:13,  3.49it/s]
 97%|█████████▋| 359476/371472 [6:26:46<55:15,  3.62it/s]
 97%|█████████▋| 359477/371472 [6:26:46<55:44,  3.59it/s]
 97%|█████████▋| 359478/371472 [6:26:46<54:42,  3.65it/s]
 97%|█████████▋| 359479/371472 [6:26:47<57:16,  3.49it/s]
 97%|█████████▋| 359480/371472 [6:26:47<54:10,  3.69it/s]
                                                         
{'loss': 2.4765, 'learning_rate': 1.2906979249714496e-07, 'epoch': 15.48}
+
 97%|█████████▋| 359480/371472 [6:26:47<54:10,  3.69it/s]
 97%|█████████▋| 359481/371472 [6:26:47<56:33,  3.53it/s]
 97%|█████████▋| 359482/371472 [6:26:48<54:57,  3.64it/s]
 97%|█████████▋| 359483/371472 [6:26:48<54:02,  3.70it/s]
 97%|█████████▋| 359484/371472 [6:26:48<55:04,  3.63it/s]
 97%|█████████▋| 359485/371472 [6:26:48<55:23,  3.61it/s]
 97%|█████████▋| 359486/371472 [6:26:49<54:18,  3.68it/s]
 97%|█████████▋| 359487/371472 [6:26:49<54:54,  3.64it/s]
 97%|█████████▋| 359488/371472 [6:26:49<52:59,  3.77it/s]
 97%|█████████▋| 359489/371472 [6:26:49<53:15,  3.75it/s]
 97%|█████████▋| 359490/371472 [6:26:50<53:23,  3.74it/s]
 97%|█████████▋| 359491/371472 [6:26:50<55:41,  3.59it/s]
 97%|█████████▋| 359492/371472 [6:26:50<56:30,  3.53it/s]
 97%|█████████▋| 359493/371472 [6:26:51<55:03,  3.63it/s]
 97%|█████████▋| 359494/371472 [6:26:51<54:38,  3.65it/s]
 97%|█████████▋| 359495/371472 [6:26:51<54:50,  3.64it/s]
 97%|█████████▋| 359496/371472 [6:26:51<53:44,  3.71it/s]
 97%|█████████▋| 359497/371472 [6:26:52<54:49,  3.64it/s]
 97%|█████████▋| 359498/371472 [6:26:52<55:24,  3.60it/s]
 97%|█████████▋| 359499/371472 [6:26:52<53:21,  3.74it/s]
 97%|█████████▋| 359500/371472 [6:26:52<52:30,  3.80it/s]
                                                         
{'loss': 2.5755, 'learning_rate': 1.2902131052166606e-07, 'epoch': 15.48}
+
 97%|█████████▋| 359500/371472 [6:26:52<52:30,  3.80it/s]
 97%|█████████▋| 359501/371472 [6:26:53<56:51,  3.51it/s]
 97%|█████████▋| 359502/371472 [6:26:53<57:07,  3.49it/s]
 97%|█████████▋| 359503/371472 [6:26:53<59:35,  3.35it/s]
 97%|█████████▋| 359504/371472 [6:26:54<59:52,  3.33it/s]
 97%|█████████▋| 359505/371472 [6:26:54<55:56,  3.57it/s]
 97%|█████████▋| 359506/371472 [6:26:54<57:36,  3.46it/s]
 97%|█████████▋| 359507/371472 [6:26:55<56:56,  3.50it/s]
 97%|█████████▋| 359508/371472 [6:26:55<55:18,  3.61it/s]
 97%|█████████▋| 359509/371472 [6:26:55<56:34,  3.52it/s]
 97%|█████████▋| 359510/371472 [6:26:55<59:17,  3.36it/s]
 97%|█████████▋| 359511/371472 [6:26:56<57:24,  3.47it/s]
 97%|█████████▋| 359512/371472 [6:26:56<59:11,  3.37it/s]
 97%|█████████▋| 359513/371472 [6:26:56<59:07,  3.37it/s]
 97%|█████████▋| 359514/371472 [6:26:57<57:56,  3.44it/s]
 97%|█████████▋| 359515/371472 [6:26:57<58:11,  3.42it/s]
 97%|█████████▋| 359516/371472 [6:26:57<1:00:13,  3.31it/s]
 97%|█████████▋| 359517/371472 [6:26:57<59:06,  3.37it/s]  
 97%|█████████▋| 359518/371472 [6:26:58<56:45,  3.51it/s]
 97%|█████████▋| 359519/371472 [6:26:58<56:27,  3.53it/s]
 97%|█████████▋| 359520/371472 [6:26:58<1:02:57,  3.16it/s]
                                                           
{'loss': 2.509, 'learning_rate': 1.289728285461871e-07, 'epoch': 15.49}
+
 97%|█████████▋| 359520/371472 [6:26:58<1:02:57,  3.16it/s]
 97%|█████████▋| 359521/371472 [6:26:59<1:02:03,  3.21it/s]
 97%|█████████▋| 359522/371472 [6:26:59<58:49,  3.39it/s]  
 97%|█████████▋| 359523/371472 [6:26:59<55:36,  3.58it/s]
 97%|█████████▋| 359524/371472 [6:27:00<57:34,  3.46it/s]
 97%|█████████▋| 359525/371472 [6:27:00<58:02,  3.43it/s]
 97%|█████████▋| 359526/371472 [6:27:00<54:53,  3.63it/s]
 97%|█████████▋| 359527/371472 [6:27:00<58:19,  3.41it/s]
 97%|█████████▋| 359528/371472 [6:27:01<57:17,  3.47it/s]
 97%|█████████▋| 359529/371472 [6:27:01<57:46,  3.45it/s]
 97%|█████████▋| 359530/371472 [6:27:01<54:29,  3.65it/s]
 97%|█████████▋| 359531/371472 [6:27:02<58:25,  3.41it/s]
 97%|█████████▋| 359532/371472 [6:27:02<56:46,  3.51it/s]
 97%|█████████▋| 359533/371472 [6:27:02<1:01:37,  3.23it/s]
 97%|█████████▋| 359534/371472 [6:27:03<1:02:55,  3.16it/s]
 97%|█████████▋| 359535/371472 [6:27:03<58:46,  3.38it/s]  
 97%|█████████▋| 359536/371472 [6:27:03<59:10,  3.36it/s]
 97%|█████████▋| 359537/371472 [6:27:03<1:01:38,  3.23it/s]
 97%|█████████▋| 359538/371472 [6:27:04<58:19,  3.41it/s]  
 97%|█████████▋| 359539/371472 [6:27:04<58:58,  3.37it/s]
 97%|█████████▋| 359540/371472 [6:27:04<59:00,  3.37it/s]
                                                         
{'loss': 2.5718, 'learning_rate': 1.2892434657070826e-07, 'epoch': 15.49}
+
 97%|█████████▋| 359540/371472 [6:27:04<59:00,  3.37it/s]
 97%|█████████▋| 359541/371472 [6:27:05<56:59,  3.49it/s]
 97%|█████████▋| 359542/371472 [6:27:05<56:59,  3.49it/s]
 97%|█████████▋| 359543/371472 [6:27:05<55:50,  3.56it/s]
 97%|█████████▋| 359544/371472 [6:27:05<52:46,  3.77it/s]
 97%|█████████▋| 359545/371472 [6:27:06<51:37,  3.85it/s]
 97%|█████████▋| 359546/371472 [6:27:06<51:53,  3.83it/s]
 97%|█████████▋| 359547/371472 [6:27:06<53:41,  3.70it/s]
 97%|█████████▋| 359548/371472 [6:27:06<52:49,  3.76it/s]
 97%|█████████▋| 359549/371472 [6:27:07<51:56,  3.83it/s]
 97%|█████████▋| 359550/371472 [6:27:07<53:18,  3.73it/s]
 97%|█████████▋| 359551/371472 [6:27:07<52:02,  3.82it/s]
 97%|█████████▋| 359552/371472 [6:27:07<53:36,  3.71it/s]
 97%|█████████▋| 359553/371472 [6:27:08<55:12,  3.60it/s]
 97%|█████████▋| 359554/371472 [6:27:08<56:23,  3.52it/s]
 97%|█████████▋| 359555/371472 [6:27:08<55:20,  3.59it/s]
 97%|█████████▋| 359556/371472 [6:27:09<53:41,  3.70it/s]
 97%|█████████▋| 359557/371472 [6:27:09<52:32,  3.78it/s]
 97%|█████████▋| 359558/371472 [6:27:09<55:17,  3.59it/s]
 97%|█████████▋| 359559/371472 [6:27:09<55:41,  3.57it/s]
 97%|█████████▋| 359560/371472 [6:27:10<55:44,  3.56it/s]
                                                         
{'loss': 2.6312, 'learning_rate': 1.2887586459522933e-07, 'epoch': 15.49}
+
 97%|█████████▋| 359560/371472 [6:27:10<55:44,  3.56it/s]
 97%|█████████▋| 359561/371472 [6:27:10<54:01,  3.67it/s]
 97%|█████████▋| 359562/371472 [6:27:10<52:01,  3.82it/s]
 97%|█████████▋| 359563/371472 [6:27:10<53:07,  3.74it/s]
 97%|█████████▋| 359564/371472 [6:27:11<52:22,  3.79it/s]
 97%|█████████▋| 359565/371472 [6:27:11<58:02,  3.42it/s]
 97%|█████████▋| 359566/371472 [6:27:11<1:00:18,  3.29it/s]
 97%|█████████▋| 359567/371472 [6:27:12<57:00,  3.48it/s]  
 97%|█████████▋| 359568/371472 [6:27:12<1:00:10,  3.30it/s]
 97%|█████████▋| 359569/371472 [6:27:12<58:49,  3.37it/s]  
 97%|█████████▋| 359570/371472 [6:27:13<57:08,  3.47it/s]
 97%|█████████▋| 359571/371472 [6:27:13<55:42,  3.56it/s]
 97%|█████████▋| 359572/371472 [6:27:13<56:01,  3.54it/s]
 97%|█████████▋| 359573/371472 [6:27:13<55:32,  3.57it/s]
 97%|█████████▋| 359574/371472 [6:27:14<57:31,  3.45it/s]
 97%|█████████▋| 359575/371472 [6:27:14<57:44,  3.43it/s]
 97%|█████████▋| 359576/371472 [6:27:14<55:32,  3.57it/s]
 97%|█████████▋| 359577/371472 [6:27:14<55:43,  3.56it/s]
 97%|█████████▋| 359578/371472 [6:27:15<55:13,  3.59it/s]
 97%|█████████▋| 359579/371472 [6:27:15<52:55,  3.75it/s]
 97%|█████████▋| 359580/371472 [6:27:15<52:42,  3.76it/s]
                                                         
{'loss': 2.6278, 'learning_rate': 1.2882738261975048e-07, 'epoch': 15.49}
+
 97%|█████████▋| 359580/371472 [6:27:15<52:42,  3.76it/s]
 97%|█████████▋| 359581/371472 [6:27:16<53:19,  3.72it/s]
 97%|█████████▋| 359582/371472 [6:27:16<59:54,  3.31it/s]
 97%|█████████▋| 359583/371472 [6:27:16<59:02,  3.36it/s]
 97%|█████████▋| 359584/371472 [6:27:16<56:46,  3.49it/s]
 97%|█████████▋| 359585/371472 [6:27:17<57:44,  3.43it/s]
 97%|█████████▋| 359586/371472 [6:27:17<55:30,  3.57it/s]
 97%|█████████▋| 359587/371472 [6:27:17<55:49,  3.55it/s]
 97%|█████████▋| 359588/371472 [6:27:18<54:52,  3.61it/s]
 97%|█████████▋| 359589/371472 [6:27:18<53:20,  3.71it/s]
 97%|█████████▋| 359590/371472 [6:27:18<51:41,  3.83it/s]
 97%|█████████▋| 359591/371472 [6:27:18<50:26,  3.93it/s]
 97%|█████████▋| 359592/371472 [6:27:19<51:54,  3.81it/s]
 97%|█████████▋| 359593/371472 [6:27:19<51:33,  3.84it/s]
 97%|█████████▋| 359594/371472 [6:27:19<50:30,  3.92it/s]
 97%|█████████▋| 359595/371472 [6:27:19<51:50,  3.82it/s]
 97%|█████████▋| 359596/371472 [6:27:20<51:39,  3.83it/s]
 97%|█████████▋| 359597/371472 [6:27:20<51:31,  3.84it/s]
 97%|█████████▋| 359598/371472 [6:27:20<50:34,  3.91it/s]
 97%|█████████▋| 359599/371472 [6:27:20<50:04,  3.95it/s]
 97%|█████████▋| 359600/371472 [6:27:21<50:44,  3.90it/s]
                                                         
{'loss': 2.7416, 'learning_rate': 1.2877890064427152e-07, 'epoch': 15.49}
+
 97%|█████████▋| 359600/371472 [6:27:21<50:44,  3.90it/s]
 97%|█████████▋| 359601/371472 [6:27:21<51:10,  3.87it/s]
 97%|█████████▋| 359602/371472 [6:27:21<56:34,  3.50it/s]
 97%|█████████▋| 359603/371472 [6:27:22<56:30,  3.50it/s]
 97%|█████████▋| 359604/371472 [6:27:22<58:30,  3.38it/s]
 97%|█████████▋| 359605/371472 [6:27:22<1:00:49,  3.25it/s]
 97%|█████████▋| 359606/371472 [6:27:22<58:43,  3.37it/s]  
 97%|█████████▋| 359607/371472 [6:27:23<56:08,  3.52it/s]
 97%|█████████▋| 359608/371472 [6:27:23<54:59,  3.60it/s]
 97%|█████████▋| 359609/371472 [6:27:23<53:04,  3.72it/s]
 97%|█████████▋| 359610/371472 [6:27:23<51:52,  3.81it/s]
 97%|█████████▋| 359611/371472 [6:27:24<50:48,  3.89it/s]
 97%|█████████▋| 359612/371472 [6:27:24<52:16,  3.78it/s]
 97%|█████████▋| 359613/371472 [6:27:24<51:14,  3.86it/s]
 97%|█████████▋| 359614/371472 [6:27:25<53:57,  3.66it/s]
 97%|█████████▋| 359615/371472 [6:27:25<52:35,  3.76it/s]
 97%|█████████▋| 359616/371472 [6:27:25<52:15,  3.78it/s]
 97%|█████████▋| 359617/371472 [6:27:25<53:31,  3.69it/s]
 97%|█████████▋| 359618/371472 [6:27:26<52:56,  3.73it/s]
 97%|█████████▋| 359619/371472 [6:27:26<51:35,  3.83it/s]
 97%|█████████▋| 359620/371472 [6:27:26<54:01,  3.66it/s]
                                                         
{'loss': 2.599, 'learning_rate': 1.287304186687927e-07, 'epoch': 15.49}
+
 97%|█████████▋| 359620/371472 [6:27:26<54:01,  3.66it/s]
 97%|█████████▋| 359621/371472 [6:27:26<53:44,  3.68it/s]
 97%|█████████▋| 359622/371472 [6:27:27<53:57,  3.66it/s]
 97%|█████████▋| 359623/371472 [6:27:27<55:19,  3.57it/s]
 97%|█████████▋| 359624/371472 [6:27:27<55:15,  3.57it/s]
 97%|█████████▋| 359625/371472 [6:27:28<56:24,  3.50it/s]
 97%|█████████▋| 359626/371472 [6:27:28<57:15,  3.45it/s]
 97%|█████████▋| 359627/371472 [6:27:28<55:39,  3.55it/s]
 97%|█████████▋| 359628/371472 [6:27:28<53:33,  3.69it/s]
 97%|█████████▋| 359629/371472 [6:27:29<53:02,  3.72it/s]
 97%|█████████▋| 359630/371472 [6:27:29<51:56,  3.80it/s]
 97%|█████████▋| 359631/371472 [6:27:29<51:29,  3.83it/s]
 97%|█████████▋| 359632/371472 [6:27:29<53:38,  3.68it/s]
 97%|█████████▋| 359633/371472 [6:27:30<54:42,  3.61it/s]
 97%|█████████▋| 359634/371472 [6:27:30<56:43,  3.48it/s]
 97%|█████████▋| 359635/371472 [6:27:30<59:58,  3.29it/s]
 97%|█████████▋| 359636/371472 [6:27:31<57:43,  3.42it/s]
 97%|█████████▋| 359637/371472 [6:27:31<54:37,  3.61it/s]
 97%|█████████▋| 359638/371472 [6:27:31<52:56,  3.73it/s]
 97%|█████████▋| 359639/371472 [6:27:31<52:11,  3.78it/s]
 97%|█████████▋| 359640/371472 [6:27:32<53:17,  3.70it/s]
                                                         
{'loss': 2.8069, 'learning_rate': 1.2868193669331375e-07, 'epoch': 15.49}
+
 97%|█████████▋| 359640/371472 [6:27:32<53:17,  3.70it/s]
 97%|█████████▋| 359641/371472 [6:27:32<56:36,  3.48it/s]
 97%|█████████▋| 359642/371472 [6:27:32<55:22,  3.56it/s]
 97%|█████████▋| 359643/371472 [6:27:33<53:04,  3.71it/s]
 97%|█████████▋| 359644/371472 [6:27:33<52:21,  3.76it/s]
 97%|█████████▋| 359645/371472 [6:27:33<52:16,  3.77it/s]
 97%|█████████▋| 359646/371472 [6:27:33<54:05,  3.64it/s]
 97%|█████████▋| 359647/371472 [6:27:34<55:52,  3.53it/s]
 97%|█████████▋| 359648/371472 [6:27:34<53:54,  3.66it/s]
 97%|█████████▋| 359649/371472 [6:27:34<53:37,  3.67it/s]
 97%|█████████▋| 359650/371472 [6:27:34<52:23,  3.76it/s]
 97%|█████████▋| 359651/371472 [6:27:35<53:24,  3.69it/s]
 97%|█████████▋| 359652/371472 [6:27:35<52:08,  3.78it/s]
 97%|█████████▋| 359653/371472 [6:27:35<53:29,  3.68it/s]
 97%|█████████▋| 359654/371472 [6:27:36<53:02,  3.71it/s]
 97%|█████████▋| 359655/371472 [6:27:36<52:53,  3.72it/s]
 97%|█████████▋| 359656/371472 [6:27:36<52:45,  3.73it/s]
 97%|█████████▋| 359657/371472 [6:27:36<54:09,  3.64it/s]
 97%|█████████▋| 359658/371472 [6:27:37<53:06,  3.71it/s]
 97%|█████████▋| 359659/371472 [6:27:37<58:51,  3.35it/s]
 97%|█████████▋| 359660/371472 [6:27:37<56:14,  3.50it/s]
                                                         
{'loss': 2.5786, 'learning_rate': 1.286334547178349e-07, 'epoch': 15.49}
+
 97%|█████████▋| 359660/371472 [6:27:37<56:14,  3.50it/s]
 97%|█████████▋| 359661/371472 [6:27:37<53:48,  3.66it/s]
 97%|█████████▋| 359662/371472 [6:27:38<52:51,  3.72it/s]
 97%|█████████▋| 359663/371472 [6:27:38<54:39,  3.60it/s]
 97%|█████████▋| 359664/371472 [6:27:38<52:09,  3.77it/s]
 97%|█████████▋| 359665/371472 [6:27:39<54:03,  3.64it/s]
 97%|█████████▋| 359666/371472 [6:27:39<55:04,  3.57it/s]
 97%|█████████▋| 359667/371472 [6:27:39<55:40,  3.53it/s]
 97%|█████████▋| 359668/371472 [6:27:39<53:56,  3.65it/s]
 97%|█████████▋| 359669/371472 [6:27:40<55:11,  3.56it/s]
 97%|█████████▋| 359670/371472 [6:27:40<57:04,  3.45it/s]
 97%|█████████▋| 359671/371472 [6:27:40<55:13,  3.56it/s]
 97%|█████████▋| 359672/371472 [6:27:41<53:57,  3.64it/s]
 97%|█████████▋| 359673/371472 [6:27:41<53:21,  3.69it/s]
 97%|████���████▋| 359674/371472 [6:27:41<52:44,  3.73it/s]
 97%|█████████▋| 359675/371472 [6:27:41<53:22,  3.68it/s]
 97%|█████████▋| 359676/371472 [6:27:42<1:01:50,  3.18it/s]
 97%|█████████▋| 359677/371472 [6:27:42<59:13,  3.32it/s]  
 97%|█████████▋| 359678/371472 [6:27:42<58:01,  3.39it/s]
 97%|█████████▋| 359679/371472 [6:27:43<55:37,  3.53it/s]
 97%|█████████▋| 359680/371472 [6:27:43<59:13,  3.32it/s]
                                                         
{'loss': 2.4706, 'learning_rate': 1.2858497274235597e-07, 'epoch': 15.49}
+
 97%|█████████▋| 359680/371472 [6:27:43<59:13,  3.32it/s]
 97%|█████████▋| 359681/371472 [6:27:43<57:39,  3.41it/s]
 97%|█████████▋| 359682/371472 [6:27:43<59:35,  3.30it/s]
 97%|█████████▋| 359683/371472 [6:27:44<56:49,  3.46it/s]
 97%|█████████▋| 359684/371472 [6:27:44<1:00:03,  3.27it/s]
 97%|█████████▋| 359685/371472 [6:27:44<1:01:55,  3.17it/s]
 97%|█████████▋| 359686/371472 [6:27:45<59:11,  3.32it/s]  
 97%|█████████▋| 359687/371472 [6:27:45<58:24,  3.36it/s]
 97%|█████████▋| 359688/371472 [6:27:45<1:00:40,  3.24it/s]
 97%|█████████▋| 359689/371472 [6:27:46<57:38,  3.41it/s]  
 97%|█████████▋| 359690/371472 [6:27:46<57:58,  3.39it/s]
 97%|█████████▋| 359691/371472 [6:27:46<59:59,  3.27it/s]
 97%|█████████▋| 359692/371472 [6:27:46<58:02,  3.38it/s]
 97%|█████████▋| 359693/371472 [6:27:47<57:11,  3.43it/s]
 97%|█████████▋| 359694/371472 [6:27:47<55:59,  3.51it/s]
 97%|█████████▋| 359695/371472 [6:27:47<56:09,  3.50it/s]
 97%|█████████▋| 359696/371472 [6:27:48<1:00:22,  3.25it/s]
 97%|█████████▋| 359697/371472 [6:27:48<57:51,  3.39it/s]  
 97%|█████████▋| 359698/371472 [6:27:48<58:27,  3.36it/s]
 97%|█████████▋| 359699/371472 [6:27:49<58:57,  3.33it/s]
 97%|█████████▋| 359700/371472 [6:27:49<58:44,  3.34it/s]
                                                         
{'loss': 2.6211, 'learning_rate': 1.2853649076687712e-07, 'epoch': 15.49}
+
 97%|█████████▋| 359700/371472 [6:27:49<58:44,  3.34it/s]
 97%|█████████▋| 359701/371472 [6:27:49<57:44,  3.40it/s]
 97%|█████████▋| 359702/371472 [6:27:49<55:46,  3.52it/s]
 97%|█████████▋| 359703/371472 [6:27:50<56:20,  3.48it/s]
 97%|█████████▋| 359704/371472 [6:27:50<54:17,  3.61it/s]
 97%|█████████▋| 359705/371472 [6:27:50<55:38,  3.52it/s]
 97%|█████████▋| 359706/371472 [6:27:50<53:35,  3.66it/s]
 97%|█████████▋| 359707/371472 [6:27:51<52:40,  3.72it/s]
 97%|█████████▋| 359708/371472 [6:27:51<53:30,  3.66it/s]
 97%|█████████▋| 359709/371472 [6:27:51<55:22,  3.54it/s]
 97%|█████████▋| 359710/371472 [6:27:52<55:32,  3.53it/s]
 97%|█████████▋| 359711/371472 [6:27:52<59:02,  3.32it/s]
 97%|█████████▋| 359712/371472 [6:27:52<1:02:03,  3.16it/s]
 97%|█████████▋| 359713/371472 [6:27:53<1:00:40,  3.23it/s]
 97%|█████████▋| 359714/371472 [6:27:53<57:31,  3.41it/s]  
 97%|█████████▋| 359715/371472 [6:27:53<55:27,  3.53it/s]
 97%|█████████▋| 359716/371472 [6:27:53<54:15,  3.61it/s]
 97%|█████████▋| 359717/371472 [6:27:54<54:27,  3.60it/s]
 97%|█████████▋| 359718/371472 [6:27:54<56:28,  3.47it/s]
 97%|█████████▋| 359719/371472 [6:27:54<54:18,  3.61it/s]
 97%|█████████▋| 359720/371472 [6:27:55<55:15,  3.54it/s]
                                                         
{'loss': 2.6636, 'learning_rate': 1.284880087913982e-07, 'epoch': 15.49}
+
 97%|█████████▋| 359720/371472 [6:27:55<55:15,  3.54it/s]
 97%|█████████▋| 359721/371472 [6:27:55<55:08,  3.55it/s]
 97%|█████████▋| 359722/371472 [6:27:55<1:00:33,  3.23it/s]
 97%|█████████▋| 359723/371472 [6:27:56<1:04:05,  3.06it/s]
 97%|█████████▋| 359724/371472 [6:27:56<59:14,  3.30it/s]  
 97%|█████████▋| 359725/371472 [6:27:56<57:07,  3.43it/s]
 97%|█████████▋| 359726/371472 [6:27:56<55:56,  3.50it/s]
 97%|█████████▋| 359727/371472 [6:27:57<54:16,  3.61it/s]
 97%|█████████▋| 359728/371472 [6:27:57<55:52,  3.50it/s]
 97%|█████████▋| 359729/371472 [6:27:57<59:17,  3.30it/s]
 97%|█████████▋| 359730/371472 [6:27:57<57:14,  3.42it/s]
 97%|███████���█▋| 359731/371472 [6:27:58<56:19,  3.47it/s]
 97%|█████████▋| 359732/371472 [6:27:58<56:51,  3.44it/s]
 97%|█████████▋| 359733/371472 [6:27:58<57:14,  3.42it/s]
 97%|█████████▋| 359734/371472 [6:27:59<53:37,  3.65it/s]
 97%|█████████▋| 359735/371472 [6:27:59<52:22,  3.73it/s]
 97%|█████████▋| 359736/371472 [6:27:59<52:44,  3.71it/s]
 97%|█████████▋| 359737/371472 [6:27:59<56:03,  3.49it/s]
 97%|█████████▋| 359738/371472 [6:28:00<55:20,  3.53it/s]
 97%|█████████▋| 359739/371472 [6:28:00<53:16,  3.67it/s]
 97%|█████████▋| 359740/371472 [6:28:00<52:06,  3.75it/s]
                                                         
{'loss': 2.6967, 'learning_rate': 1.2843952681591934e-07, 'epoch': 15.49}
+
 97%|█████████▋| 359740/371472 [6:28:00<52:06,  3.75it/s]
 97%|█████████▋| 359741/371472 [6:28:00<51:43,  3.78it/s]
 97%|█████████▋| 359742/371472 [6:28:01<50:59,  3.83it/s]
 97%|█████████▋| 359743/371472 [6:28:01<52:13,  3.74it/s]
 97%|█████████▋| 359744/371472 [6:28:01<53:48,  3.63it/s]
 97%|█████████▋| 359745/371472 [6:28:02<53:26,  3.66it/s]
 97%|█████████▋| 359746/371472 [6:28:02<55:28,  3.52it/s]
 97%|█████████▋| 359747/371472 [6:28:02<58:08,  3.36it/s]
 97%|█████████▋| 359748/371472 [6:28:03<58:02,  3.37it/s]
 97%|█████████▋| 359749/371472 [6:28:03<53:48,  3.63it/s]
 97%|█████████▋| 359750/371472 [6:28:03<1:01:13,  3.19it/s]
 97%|█████████▋| 359751/371472 [6:28:03<57:34,  3.39it/s]  
 97%|█████████▋| 359752/371472 [6:28:04<1:00:05,  3.25it/s]
 97%|█████████▋| 359753/371472 [6:28:04<57:40,  3.39it/s]  
 97%|█████████▋| 359754/371472 [6:28:04<59:45,  3.27it/s]
 97%|█████████▋| 359755/371472 [6:28:05<59:54,  3.26it/s]
 97%|█████████▋| 359756/371472 [6:28:05<1:03:31,  3.07it/s]
 97%|█████████▋| 359757/371472 [6:28:05<59:12,  3.30it/s]  
 97%|█████████▋| 359758/371472 [6:28:06<58:52,  3.32it/s]
 97%|█████████▋| 359759/371472 [6:28:06<57:24,  3.40it/s]
 97%|█████████▋| 359760/371472 [6:28:06<57:32,  3.39it/s]
                                                         
{'loss': 2.547, 'learning_rate': 1.283910448404404e-07, 'epoch': 15.5}
+
 97%|█████████▋| 359760/371472 [6:28:06<57:32,  3.39it/s]
 97%|█████████▋| 359761/371472 [6:28:06<56:35,  3.45it/s]
 97%|█████████▋| 359762/371472 [6:28:07<58:38,  3.33it/s]
 97%|█████████▋| 359763/371472 [6:28:07<55:32,  3.51it/s]
 97%|█████████▋| 359764/371472 [6:28:07<58:57,  3.31it/s]
 97%|█████████▋| 359765/371472 [6:28:08<56:21,  3.46it/s]
 97%|█████████▋| 359766/371472 [6:28:08<53:56,  3.62it/s]
 97%|█████████▋| 359767/371472 [6:28:08<53:16,  3.66it/s]
 97%|█████████▋| 359768/371472 [6:28:08<54:04,  3.61it/s]
 97%|█████████▋| 359769/371472 [6:28:09<52:07,  3.74it/s]
 97%|█████████▋| 359770/371472 [6:28:09<52:56,  3.68it/s]
 97%|█████████▋| 359771/371472 [6:28:09<52:41,  3.70it/s]
 97%|█████████▋| 359772/371472 [6:28:09<50:12,  3.88it/s]
 97%|█████████▋| 359773/371472 [6:28:10<51:22,  3.80it/s]
 97%|█████████▋| 359774/371472 [6:28:10<49:41,  3.92it/s]
 97%|█████████▋| 359775/371472 [6:28:10<50:05,  3.89it/s]
 97%|█████████▋| 359776/371472 [6:28:10<53:04,  3.67it/s]
 97%|█████████▋| 359777/371472 [6:28:11<56:00,  3.48it/s]
 97%|█████████▋| 359778/371472 [6:28:11<59:35,  3.27it/s]
 97%|█████████▋| 359779/371472 [6:28:11<58:16,  3.34it/s]
 97%|█████████▋| 359780/371472 [6:28:12<54:56,  3.55it/s]
                                                         
{'loss': 2.6945, 'learning_rate': 1.2834256286496156e-07, 'epoch': 15.5}
+
 97%|█████████▋| 359780/371472 [6:28:12<54:56,  3.55it/s]
 97%|█████████▋| 359781/371472 [6:28:12<53:32,  3.64it/s]
 97%|█████████▋| 359782/371472 [6:28:12<52:03,  3.74it/s]
 97%|█████████▋| 359783/371472 [6:28:12<50:46,  3.84it/s]
 97%|█████████▋| 359784/371472 [6:28:13<50:01,  3.89it/s]
 97%|█████████▋| 359785/371472 [6:28:13<52:43,  3.69it/s]
 97%|█████████▋| 359786/371472 [6:28:13<52:01,  3.74it/s]
 97%|█████████▋| 359787/371472 [6:28:14<53:23,  3.65it/s]
 97%|█████████▋| 359788/371472 [6:28:14<54:05,  3.60it/s]
 97%|█████████▋| 359789/371472 [6:28:14<52:25,  3.71it/s]
 97%|█████████▋| 359790/371472 [6:28:14<52:05,  3.74it/s]
 97%|█████████▋| 359791/371472 [6:28:15<55:49,  3.49it/s]
 97%|█████████▋| 359792/371472 [6:28:15<53:40,  3.63it/s]
 97%|█████████▋| 359793/371472 [6:28:15<55:12,  3.53it/s]
 97%|█████████▋| 359794/371472 [6:28:16<56:51,  3.42it/s]
 97%|█████████▋| 359795/371472 [6:28:16<56:28,  3.45it/s]
 97%|█████████▋| 359796/371472 [6:28:16<1:02:00,  3.14it/s]
 97%|█████████▋| 359797/371472 [6:28:16<59:22,  3.28it/s]  
 97%|█████████▋| 359798/371472 [6:28:17<55:56,  3.48it/s]
 97%|█████████▋| 359799/371472 [6:28:17<53:39,  3.63it/s]
 97%|█████████▋| 359800/371472 [6:28:17<54:26,  3.57it/s]
                                                         
{'loss': 2.5918, 'learning_rate': 1.282940808894826e-07, 'epoch': 15.5}
+
 97%|█████████▋| 359800/371472 [6:28:17<54:26,  3.57it/s]
 97%|█████████▋| 359801/371472 [6:28:18<53:11,  3.66it/s]
 97%|█████████▋| 359802/371472 [6:28:18<53:24,  3.64it/s]
 97%|█████████▋| 359803/371472 [6:28:18<53:03,  3.66it/s]
 97%|█████████▋| 359804/371472 [6:28:18<52:33,  3.70it/s]
 97%|█████████▋| 359805/371472 [6:28:19<54:03,  3.60it/s]
 97%|█████████▋| 359806/371472 [6:28:19<54:42,  3.55it/s]
 97%|█████████▋| 359807/371472 [6:28:19<55:08,  3.53it/s]
 97%|█████████▋| 359808/371472 [6:28:19<54:10,  3.59it/s]
 97%|█████████▋| 359809/371472 [6:28:20<58:04,  3.35it/s]
 97%|█████████▋| 359810/371472 [6:28:20<55:47,  3.48it/s]
 97%|█████████▋| 359811/371472 [6:28:20<55:10,  3.52it/s]
 97%|█████████▋| 359812/371472 [6:28:21<54:06,  3.59it/s]
 97%|█████████▋| 359813/371472 [6:28:21<51:29,  3.77it/s]
 97%|█████████▋| 359814/371472 [6:28:21<53:21,  3.64it/s]
 97%|█████████▋| 359815/371472 [6:28:21<52:38,  3.69it/s]
 97%|█████████▋| 359816/371472 [6:28:22<55:08,  3.52it/s]
 97%|█████████▋| 359817/371472 [6:28:22<56:46,  3.42it/s]
 97%|█████████▋| 359818/371472 [6:28:22<56:08,  3.46it/s]
 97%|█████████▋| 359819/371472 [6:28:23<54:24,  3.57it/s]
 97%|█████████▋| 359820/371472 [6:28:23<53:03,  3.66it/s]
                                                         
{'loss': 2.6382, 'learning_rate': 1.282455989140038e-07, 'epoch': 15.5}
+
 97%|█████████▋| 359820/371472 [6:28:23<53:03,  3.66it/s]
 97%|█████████▋| 359821/371472 [6:28:23<56:50,  3.42it/s]
 97%|█████████▋| 359822/371472 [6:28:23<55:18,  3.51it/s]
 97%|█████████▋| 359823/371472 [6:28:24<58:28,  3.32it/s]
 97%|█████████▋| 359824/371472 [6:28:24<57:23,  3.38it/s]
 97%|█████████▋| 359825/371472 [6:28:24<56:03,  3.46it/s]
 97%|█████████▋| 359826/371472 [6:28:25<56:41,  3.42it/s]
 97%|█████████▋| 359827/371472 [6:28:25<54:38,  3.55it/s]
 97%|█████████▋| 359828/371472 [6:28:25<55:03,  3.52it/s]
 97%|█████████▋| 359829/371472 [6:28:25<53:17,  3.64it/s]
 97%|█████████▋| 359830/371472 [6:28:26<53:18,  3.64it/s]
 97%|█████████▋| 359831/371472 [6:28:26<50:42,  3.83it/s]
 97%|█████████▋| 359832/371472 [6:28:26<52:19,  3.71it/s]
 97%|█████████▋| 359833/371472 [6:28:26<51:25,  3.77it/s]
 97%|█████████▋| 359834/371472 [6:28:27<51:44,  3.75it/s]
 97%|█████████▋| 359835/371472 [6:28:27<54:38,  3.55it/s]
 97%|█████████▋| 359836/371472 [6:28:27<57:12,  3.39it/s]
 97%|█████████▋| 359837/371472 [6:28:28<54:36,  3.55it/s]
 97%|█████████▋| 359838/371472 [6:28:28<55:29,  3.49it/s]
 97%|█████████▋| 359839/371472 [6:28:28<54:57,  3.53it/s]
 97%|█████████▋| 359840/371472 [6:28:29<55:38,  3.48it/s]
                                                         
{'loss': 2.5138, 'learning_rate': 1.2819711693852483e-07, 'epoch': 15.5}
+
 97%|█████████▋| 359840/371472 [6:28:29<55:38,  3.48it/s]
 97%|█████████▋| 359841/371472 [6:28:29<54:47,  3.54it/s]
 97%|█████████▋| 359842/371472 [6:28:29<57:07,  3.39it/s]
 97%|█████████▋| 359843/371472 [6:28:29<54:30,  3.56it/s]
 97%|█████████▋| 359844/371472 [6:28:30<56:45,  3.41it/s]
 97%|█████████▋| 359845/371472 [6:28:30<54:16,  3.57it/s]
 97%|████���████▋| 359846/371472 [6:28:30<54:16,  3.57it/s]
 97%|█████████▋| 359847/371472 [6:28:30<54:00,  3.59it/s]
 97%|█████████▋| 359848/371472 [6:28:31<53:43,  3.61it/s]
 97%|█████████▋| 359849/371472 [6:28:31<52:13,  3.71it/s]
 97%|█████████▋| 359850/371472 [6:28:31<52:10,  3.71it/s]
 97%|█████████▋| 359851/371472 [6:28:32<55:29,  3.49it/s]
 97%|█████████▋| 359852/371472 [6:28:32<53:13,  3.64it/s]
 97%|█████████▋| 359853/371472 [6:28:32<53:41,  3.61it/s]
 97%|█████████▋| 359854/371472 [6:28:32<51:53,  3.73it/s]
 97%|█████████▋| 359855/371472 [6:28:33<49:50,  3.88it/s]
 97%|█████████▋| 359856/371472 [6:28:33<50:56,  3.80it/s]
 97%|█████████▋| 359857/371472 [6:28:33<50:02,  3.87it/s]
 97%|█████████▋| 359858/371472 [6:28:33<48:55,  3.96it/s]
 97%|█████████▋| 359859/371472 [6:28:34<48:42,  3.97it/s]
 97%|█████████▋| 359860/371472 [6:28:34<53:45,  3.60it/s]
                                                         
{'loss': 2.4515, 'learning_rate': 1.2814863496304598e-07, 'epoch': 15.5}
+
 97%|█████████▋| 359860/371472 [6:28:34<53:45,  3.60it/s]
 97%|█████████▋| 359861/371472 [6:28:34<55:46,  3.47it/s]
 97%|█████████▋| 359862/371472 [6:28:35<1:01:47,  3.13it/s]
 97%|█████████▋| 359863/371472 [6:28:35<57:17,  3.38it/s]  
 97%|█████████▋| 359864/371472 [6:28:35<54:17,  3.56it/s]
 97%|█████████▋| 359865/371472 [6:28:35<53:26,  3.62it/s]
 97%|█████████▋| 359866/371472 [6:28:36<53:41,  3.60it/s]
 97%|█████████▋| 359867/371472 [6:28:36<55:38,  3.48it/s]
 97%|█████████▋| 359868/371472 [6:28:36<1:03:34,  3.04it/s]
 97%|█████████▋| 359869/371472 [6:28:37<1:00:47,  3.18it/s]
 97%|█████████▋| 359870/371472 [6:28:37<59:25,  3.25it/s]  
 97%|█████████▋| 359871/371472 [6:28:37<56:05,  3.45it/s]
 97%|█████████▋| 359872/371472 [6:28:38<53:39,  3.60it/s]
 97%|█████████▋| 359873/371472 [6:28:38<51:25,  3.76it/s]
 97%|█████████▋| 359874/371472 [6:28:38<52:00,  3.72it/s]
 97%|█████████▋| 359875/371472 [6:28:38<52:25,  3.69it/s]
 97%|█████████▋| 359876/371472 [6:28:39<53:47,  3.59it/s]
 97%|█████████▋| 359877/371472 [6:28:39<58:01,  3.33it/s]
 97%|█████████▋| 359878/371472 [6:28:39<59:39,  3.24it/s]
 97%|█████████▋| 359879/371472 [6:28:40<1:01:09,  3.16it/s]
 97%|█████████▋| 359880/371472 [6:28:40<57:38,  3.35it/s]  
                                                         
{'loss': 2.4686, 'learning_rate': 1.2810015298756705e-07, 'epoch': 15.5}
+
 97%|█████████▋| 359880/371472 [6:28:40<57:38,  3.35it/s]
 97%|█████████▋| 359881/371472 [6:28:40<55:03,  3.51it/s]
 97%|█████████▋| 359882/371472 [6:28:40<53:51,  3.59it/s]
 97%|█████████▋| 359883/371472 [6:28:41<55:06,  3.51it/s]
 97%|█████████▋| 359884/371472 [6:28:41<54:34,  3.54it/s]
 97%|█████████▋| 359885/371472 [6:28:41<52:41,  3.66it/s]
 97%|█████████▋| 359886/371472 [6:28:42<54:59,  3.51it/s]
 97%|█████████▋| 359887/371472 [6:28:42<58:45,  3.29it/s]
 97%|█████████▋| 359888/371472 [6:28:42<55:41,  3.47it/s]
 97%|█████████▋| 359889/371472 [6:28:42<54:43,  3.53it/s]
 97%|█████████▋| 359890/371472 [6:28:43<54:21,  3.55it/s]
 97%|█████████▋| 359891/371472 [6:28:43<56:57,  3.39it/s]
 97%|█████████▋| 359892/371472 [6:28:43<55:14,  3.49it/s]
 97%|█████████▋| 359893/371472 [6:28:44<53:42,  3.59it/s]
 97%|█████████▋| 359894/371472 [6:28:44<56:59,  3.39it/s]
 97%|█████████▋| 359895/371472 [6:28:44<55:36,  3.47it/s]
 97%|█████████▋| 359896/371472 [6:28:44<53:11,  3.63it/s]
 97%|█████████▋| 359897/371472 [6:28:45<53:53,  3.58it/s]
 97%|█████████▋| 359898/371472 [6:28:45<57:01,  3.38it/s]
 97%|█████████▋| 359899/371472 [6:28:45<54:14,  3.56it/s]
 97%|█████████▋| 359900/371472 [6:28:46<52:35,  3.67it/s]
                                                         
{'loss': 2.4857, 'learning_rate': 1.280516710120882e-07, 'epoch': 15.5}
+
 97%|█████████▋| 359900/371472 [6:28:46<52:35,  3.67it/s]
 97%|█████████▋| 359901/371472 [6:28:46<51:40,  3.73it/s]
 97%|█████████▋| 359902/371472 [6:28:46<53:42,  3.59it/s]
 97%|█████████▋| 359903/371472 [6:28:46<57:35,  3.35it/s]
 97%|█████████▋| 359904/371472 [6:28:47<57:24,  3.36it/s]
 97%|█████████▋| 359905/371472 [6:28:47<56:44,  3.40it/s]
 97%|█████████▋| 359906/371472 [6:28:47<57:31,  3.35it/s]
 97%|█████████▋| 359907/371472 [6:28:48<59:29,  3.24it/s]
 97%|█████████▋| 359908/371472 [6:28:48<58:04,  3.32it/s]
 97%|█████████▋| 359909/371472 [6:28:48<54:47,  3.52it/s]
 97%|█████████▋| 359910/371472 [6:28:48<54:43,  3.52it/s]
 97%|█████████▋| 359911/371472 [6:28:49<54:14,  3.55it/s]
 97%|█████████▋| 359912/371472 [6:28:49<54:57,  3.51it/s]
 97%|█████████▋| 359913/371472 [6:28:49<51:53,  3.71it/s]
 97%|█████████▋| 359914/371472 [6:28:50<53:07,  3.63it/s]
 97%|█████████▋| 359915/371472 [6:28:50<54:51,  3.51it/s]
 97%|█████████▋| 359916/371472 [6:28:50<52:04,  3.70it/s]
 97%|█████████▋| 359917/371472 [6:28:50<55:25,  3.47it/s]
 97%|█████████▋| 359918/371472 [6:28:51<53:59,  3.57it/s]
 97%|█████████▋| 359919/371472 [6:28:51<53:54,  3.57it/s]
 97%|█████████▋| 359920/371472 [6:28:51<54:08,  3.56it/s]
                                                         
{'loss': 2.523, 'learning_rate': 1.2800318903660925e-07, 'epoch': 15.5}
+
 97%|█████████▋| 359920/371472 [6:28:51<54:08,  3.56it/s]
 97%|█████████▋| 359921/371472 [6:28:51<52:40,  3.65it/s]
 97%|█████████▋| 359922/371472 [6:28:52<52:08,  3.69it/s]
 97%|█████████▋| 359923/371472 [6:28:52<55:26,  3.47it/s]
 97%|█████████▋| 359924/371472 [6:28:52<1:00:36,  3.18it/s]
 97%|█████████▋| 359925/371472 [6:28:53<57:04,  3.37it/s]  
 97%|█████████▋| 359926/371472 [6:28:53<56:36,  3.40it/s]
 97%|█████████▋| 359927/371472 [6:28:53<59:55,  3.21it/s]
 97%|█████████▋| 359928/371472 [6:28:54<1:00:53,  3.16it/s]
 97%|█████████▋| 359929/371472 [6:28:54<1:01:08,  3.15it/s]
 97%|█████████▋| 359930/371472 [6:28:54<57:42,  3.33it/s]  
 97%|█████████▋| 359931/371472 [6:28:55<55:50,  3.44it/s]
 97%|█████████▋| 359932/371472 [6:28:55<56:49,  3.39it/s]
 97%|█████████▋| 359933/371472 [6:28:55<55:40,  3.45it/s]
 97%|█████████▋| 359934/371472 [6:28:55<59:51,  3.21it/s]
 97%|█████████▋| 359935/371472 [6:28:56<1:00:26,  3.18it/s]
 97%|█████████▋| 359936/371472 [6:28:56<58:11,  3.30it/s]  
 97%|█████████▋| 359937/371472 [6:28:56<58:50,  3.27it/s]
 97%|█████████▋| 359938/371472 [6:28:57<57:18,  3.35it/s]
 97%|█████████▋| 359939/371472 [6:28:57<1:00:54,  3.16it/s]
 97%|█████████▋| 359940/371472 [6:28:57<57:50,  3.32it/s]  
                                                         
{'loss': 2.5763, 'learning_rate': 1.2795470706113043e-07, 'epoch': 15.5}
+
 97%|█████████▋| 359940/371472 [6:28:57<57:50,  3.32it/s]
 97%|█████████▋| 359941/371472 [6:28:58<56:57,  3.37it/s]
 97%|█████████▋| 359942/371472 [6:28:58<55:14,  3.48it/s]
 97%|█████████▋| 359943/371472 [6:28:58<56:35,  3.40it/s]
 97%|█████████▋| 359944/371472 [6:28:58<56:53,  3.38it/s]
 97%|█████████▋| 359945/371472 [6:28:59<57:13,  3.36it/s]
 97%|█████████▋| 359946/371472 [6:28:59<1:00:06,  3.20it/s]
 97%|█████████▋| 359947/371472 [6:28:59<1:04:46,  2.97it/s]
 97%|█████████▋| 359948/371472 [6:29:00<1:00:27,  3.18it/s]
 97%|█████████▋| 359949/371472 [6:29:00<55:39,  3.45it/s]  
 97%|█████████▋| 359950/371472 [6:29:00<56:29,  3.40it/s]
 97%|█████████▋| 359951/371472 [6:29:01<53:36,  3.58it/s]
 97%|█████████▋| 359952/371472 [6:29:01<51:21,  3.74it/s]
 97%|█████████▋| 359953/371472 [6:29:01<51:44,  3.71it/s]
 97%|█████████▋| 359954/371472 [6:29:01<52:41,  3.64it/s]
 97%|█████████▋| 359955/371472 [6:29:02<51:39,  3.72it/s]
 97%|█████████▋| 359956/371472 [6:29:02<52:09,  3.68it/s]
 97%|█████████▋| 359957/371472 [6:29:02<51:23,  3.73it/s]
 97%|█████████▋| 359958/371472 [6:29:02<49:05,  3.91it/s]
 97%|█████████▋| 359959/371472 [6:29:03<48:54,  3.92it/s]
 97%|█████████▋| 359960/371472 [6:29:03<53:55,  3.56it/s]
                                                         
{'loss': 2.5861, 'learning_rate': 1.2790622508565147e-07, 'epoch': 15.5}
+
 97%|█████████▋| 359960/371472 [6:29:03<53:55,  3.56it/s]
 97%|█████████▋| 359961/371472 [6:29:03<57:57,  3.31it/s]
 97%|█████████▋| 359962/371472 [6:29:04<56:03,  3.42it/s]
 97%|█████████▋| 359963/371472 [6:29:04<53:58,  3.55it/s]
 97%|█████████▋| 359964/371472 [6:29:04<52:11,  3.67it/s]
 97%|█████████▋| 359965/371472 [6:29:04<58:35,  3.27it/s]
 97%|█████████▋| 359966/371472 [6:29:05<56:37,  3.39it/s]
 97%|█████████▋| 359967/371472 [6:29:05<54:19,  3.53it/s]
 97%|█████████▋| 359968/371472 [6:29:05<51:49,  3.70it/s]
 97%|█████████▋| 359969/371472 [6:29:05<51:27,  3.73it/s]
 97%|█████████▋| 359970/371472 [6:29:06<53:53,  3.56it/s]
 97%|█████████▋| 359971/371472 [6:29:06<56:48,  3.37it/s]
 97%|█████████▋| 359972/371472 [6:29:06<59:47,  3.21it/s]
 97%|█████████▋| 359973/371472 [6:29:07<57:12,  3.35it/s]
 97%|█████████▋| 359974/371472 [6:29:07<53:39,  3.57it/s]
 97%|█████████▋| 359975/371472 [6:29:07<52:52,  3.62it/s]
 97%|█████████▋| 359976/371472 [6:29:08<51:49,  3.70it/s]
 97%|█████████▋| 359977/371472 [6:29:08<52:32,  3.65it/s]
 97%|█████████▋| 359978/371472 [6:29:08<53:01,  3.61it/s]
 97%|█████████▋| 359979/371472 [6:29:09<1:02:56,  3.04it/s]
 97%|█████████▋| 359980/371472 [6:29:09<1:01:50,  3.10it/s]
                                                           
{'loss': 2.5417, 'learning_rate': 1.2785774311017262e-07, 'epoch': 15.51}
+
 97%|█████████▋| 359980/371472 [6:29:09<1:01:50,  3.10it/s]
 97%|█████████▋| 359981/371472 [6:29:09<1:01:23,  3.12it/s]
 97%|█████████▋| 359982/371472 [6:29:09<57:38,  3.32it/s]  
 97%|█████████▋| 359983/371472 [6:29:10<55:23,  3.46it/s]
 97%|█████████▋| 359984/371472 [6:29:10<59:28,  3.22it/s]
 97%|█████████▋| 359985/371472 [6:29:10<56:50,  3.37it/s]
 97%|█████████▋| 359986/371472 [6:29:11<55:39,  3.44it/s]
 97%|█████████▋| 359987/371472 [6:29:11<56:38,  3.38it/s]
 97%|█████████▋| 359988/371472 [6:29:11<58:09,  3.29it/s]
 97%|█████████▋| 359989/371472 [6:29:11<56:17,  3.40it/s]
 97%|█████████▋| 359990/371472 [6:29:12<54:27,  3.51it/s]
 97%|█████████▋| 359991/371472 [6:29:12<52:59,  3.61it/s]
 97%|█████████▋| 359992/371472 [6:29:12<58:03,  3.30it/s]
 97%|█████████▋| 359993/371472 [6:29:13<56:20,  3.40it/s]
 97%|█████████▋| 359994/371472 [6:29:13<58:46,  3.25it/s]
 97%|█████████▋| 359995/371472 [6:29:13<57:51,  3.31it/s]
 97%|█████████▋| 359996/371472 [6:29:14<58:01,  3.30it/s]
 97%|█████████▋| 359997/371472 [6:29:14<54:54,  3.48it/s]
 97%|█████████▋| 359998/371472 [6:29:14<55:04,  3.47it/s]
 97%|█████████▋| 359999/371472 [6:29:14<54:55,  3.48it/s]
 97%|█████████▋| 360000/371472 [6:29:15<58:23,  3.27it/s]
                                                         
{'loss': 2.5145, 'learning_rate': 1.278092611346937e-07, 'epoch': 15.51}
+
 97%|█████████▋| 360000/371472 [6:29:15<58:23,  3.27it/s]
 97%|█████████▋| 360001/371472 [6:29:15<57:40,  3.31it/s]
 97%|█████████▋| 360002/371472 [6:29:15<56:59,  3.35it/s]
 97%|█████████▋| 360003/371472 [6:29:16<56:03,  3.41it/s]
 97%|█████████▋| 360004/371472 [6:29:16<55:32,  3.44it/s]
 97%|█████████▋| 360005/371472 [6:29:16<55:23,  3.45it/s]
 97%|█████████▋| 360006/371472 [6:29:16<54:50,  3.48it/s]
 97%|█████████▋| 360007/371472 [6:29:17<54:48,  3.49it/s]
 97%|█████████▋| 360008/371472 [6:29:17<57:01,  3.35it/s]
 97%|█████████▋| 360009/371472 [6:29:17<55:19,  3.45it/s]
 97%|█████████▋| 360010/371472 [6:29:18<57:30,  3.32it/s]
 97%|█████████▋| 360011/371472 [6:29:18<54:51,  3.48it/s]
 97%|█████████▋| 360012/371472 [6:29:18<52:12,  3.66it/s]
 97%|█████████▋| 360013/371472 [6:29:18<52:26,  3.64it/s]
 97%|█████████▋| 360014/371472 [6:29:19<52:24,  3.64it/s]
 97%|█████████▋| 360015/371472 [6:29:19<52:34,  3.63it/s]
 97%|█████████▋| 360016/371472 [6:29:19<52:29,  3.64it/s]
 97%|█████████▋| 360017/371472 [6:29:20<52:01,  3.67it/s]
 97%|█████████▋| 360018/371472 [6:29:20<50:53,  3.75it/s]
 97%|█████████▋| 360019/371472 [6:29:20<54:24,  3.51it/s]
 97%|█████████▋| 360020/371472 [6:29:20<52:16,  3.65it/s]
                                                         
{'loss': 2.7408, 'learning_rate': 1.2776077915921484e-07, 'epoch': 15.51}
+
 97%|█████████▋| 360020/371472 [6:29:20<52:16,  3.65it/s]
 97%|█████████▋| 360021/371472 [6:29:21<51:53,  3.68it/s]
 97%|█████████▋| 360022/371472 [6:29:21<51:11,  3.73it/s]
 97%|█████████▋| 360023/371472 [6:29:21<52:40,  3.62it/s]
 97%|█████████▋| 360024/371472 [6:29:21<54:33,  3.50it/s]
 97%|█████████▋| 360025/371472 [6:29:22<56:19,  3.39it/s]
 97%|█████████▋| 360026/371472 [6:29:22<54:55,  3.47it/s]
 97%|█████████▋| 360027/371472 [6:29:22<52:39,  3.62it/s]
 97%|█████████▋| 360028/371472 [6:29:23<52:32,  3.63it/s]
 97%|█████████▋| 360029/371472 [6:29:23<52:57,  3.60it/s]
 97%|█████████▋| 360030/371472 [6:29:23<51:31,  3.70it/s]
 97%|█████████▋| 360031/371472 [6:29:23<50:27,  3.78it/s]
 97%|█████████▋| 360032/371472 [6:29:24<49:25,  3.86it/s]
 97%|█████████▋| 360033/371472 [6:29:24<49:57,  3.82it/s]
 97%|█████████▋| 360034/371472 [6:29:24<53:21,  3.57it/s]
 97%|█████████▋| 360035/371472 [6:29:24<52:24,  3.64it/s]
 97%|█████████▋| 360036/371472 [6:29:25<51:10,  3.72it/s]
 97%|█████████▋| 360037/371472 [6:29:25<50:24,  3.78it/s]
 97%|█████████▋| 360038/371472 [6:29:25<51:21,  3.71it/s]
 97%|█████████▋| 360039/371472 [6:29:26<52:52,  3.60it/s]
 97%|█████████▋| 360040/371472 [6:29:26<51:36,  3.69it/s]
                                                         
{'loss': 2.5205, 'learning_rate': 1.2771229718373592e-07, 'epoch': 15.51}
+
 97%|█████████▋| 360040/371472 [6:29:26<51:36,  3.69it/s]
 97%|█████████▋| 360041/371472 [6:29:26<53:19,  3.57it/s]
 97%|█████████▋| 360042/371472 [6:29:26<51:43,  3.68it/s]
 97%|█████████▋| 360043/371472 [6:29:27<53:07,  3.59it/s]
 97%|█████████▋| 360044/371472 [6:29:27<52:14,  3.65it/s]
 97%|█████████▋| 360045/371472 [6:29:27<50:31,  3.77it/s]
 97%|█████████▋| 360046/371472 [6:29:27<50:33,  3.77it/s]
 97%|█████████▋| 360047/371472 [6:29:28<54:04,  3.52it/s]
 97%|█████████▋| 360048/371472 [6:29:28<58:17,  3.27it/s]
 97%|█████████▋| 360049/371472 [6:29:28<58:52,  3.23it/s]
 97%|█████████▋| 360050/371472 [6:29:29<1:01:19,  3.10it/s]
 97%|█████████▋| 360051/371472 [6:29:29<1:05:29,  2.91it/s]
 97%|█████████▋| 360052/371472 [6:29:29<1:01:00,  3.12it/s]
 97%|█████████▋| 360053/371472 [6:29:30<57:35,  3.30it/s]  
 97%|█████████▋| 360054/371472 [6:29:30<1:06:26,  2.86it/s]
 97%|█████████▋| 360055/371472 [6:29:30<59:52,  3.18it/s]  
 97%|█████████▋| 360056/371472 [6:29:31<59:13,  3.21it/s]
 97%|█████████▋| 360057/371472 [6:29:31<56:23,  3.37it/s]
 97%|█████████▋| 360058/371472 [6:29:31<56:08,  3.39it/s]
 97%|█████████▋| 360059/371472 [6:29:32<54:42,  3.48it/s]
 97%|█████████▋| 360060/371472 [6:29:32<55:52,  3.40it/s]
                                                         
{'loss': 2.639, 'learning_rate': 1.27663815208257e-07, 'epoch': 15.51}
+
 97%|█████████▋| 360060/371472 [6:29:32<55:52,  3.40it/s]
 97%|█████████▋| 360061/371472 [6:29:32<53:29,  3.56it/s]
 97%|█████████▋| 360062/371472 [6:29:32<51:52,  3.67it/s]
 97%|█████████▋| 360063/371472 [6:29:33<51:13,  3.71it/s]
 97%|█████████▋| 360064/371472 [6:29:33<52:14,  3.64it/s]
 97%|█████████▋| 360065/371472 [6:29:33<52:14,  3.64it/s]
 97%|█████████▋| 360066/371472 [6:29:33<54:12,  3.51it/s]
 97%|█████████▋| 360067/371472 [6:29:34<54:07,  3.51it/s]
 97%|█████████▋| 360068/371472 [6:29:34<52:04,  3.65it/s]
 97%|█████████▋| 360069/371472 [6:29:34<52:36,  3.61it/s]
 97%|█████████▋| 360070/371472 [6:29:35<54:11,  3.51it/s]
 97%|█████████▋| 360071/371472 [6:29:35<52:45,  3.60it/s]
 97%|█████████▋| 360072/371472 [6:29:35<50:51,  3.74it/s]
 97%|█████████▋| 360073/371472 [6:29:35<50:29,  3.76it/s]
 97%|█████████▋| 360074/371472 [6:29:36<50:07,  3.79it/s]
 97%|█████████▋| 360075/371472 [6:29:36<50:31,  3.76it/s]
 97%|█████████▋| 360076/371472 [6:29:36<51:43,  3.67it/s]
 97%|█████████▋| 360077/371472 [6:29:36<51:30,  3.69it/s]
 97%|████���████▋| 360078/371472 [6:29:37<54:36,  3.48it/s]
 97%|█████████▋| 360079/371472 [6:29:37<52:05,  3.64it/s]
 97%|█████████▋| 360080/371472 [6:29:37<50:23,  3.77it/s]
                                                         
{'loss': 2.6369, 'learning_rate': 1.276153332327781e-07, 'epoch': 15.51}
+
 97%|█████████▋| 360080/371472 [6:29:37<50:23,  3.77it/s]
 97%|█████████▋| 360081/371472 [6:29:38<50:11,  3.78it/s]
 97%|█████████▋| 360082/371472 [6:29:38<51:51,  3.66it/s]
 97%|█████████▋| 360083/371472 [6:29:38<58:21,  3.25it/s]
 97%|█████████▋| 360084/371472 [6:29:39<57:06,  3.32it/s]
 97%|█████████▋| 360085/371472 [6:29:39<59:26,  3.19it/s]
 97%|█████████▋| 360086/371472 [6:29:39<56:02,  3.39it/s]
 97%|█████████▋| 360087/371472 [6:29:39<55:03,  3.45it/s]
 97%|█████████▋| 360088/371472 [6:29:40<56:51,  3.34it/s]
 97%|█████████▋| 360089/371472 [6:29:40<57:15,  3.31it/s]
 97%|█████████▋| 360090/371472 [6:29:40<56:34,  3.35it/s]
 97%|█████████▋| 360091/371472 [6:29:41<54:57,  3.45it/s]
 97%|█████████▋| 360092/371472 [6:29:41<52:01,  3.65it/s]
 97%|█████████▋| 360093/371472 [6:29:41<54:01,  3.51it/s]
 97%|█████████▋| 360094/371472 [6:29:41<56:07,  3.38it/s]
 97%|█████████▋| 360095/371472 [6:29:42<57:11,  3.32it/s]
 97%|█████████▋| 360096/371472 [6:29:42<55:00,  3.45it/s]
 97%|█████████▋| 360097/371472 [6:29:42<54:27,  3.48it/s]
 97%|█████████▋| 360098/371472 [6:29:43<51:36,  3.67it/s]
 97%|█████████▋| 360099/371472 [6:29:43<50:12,  3.77it/s]
 97%|█████████▋| 360100/371472 [6:29:43<49:38,  3.82it/s]
                                                         
{'loss': 2.4437, 'learning_rate': 1.2756685125729918e-07, 'epoch': 15.51}
+
 97%|█████████▋| 360100/371472 [6:29:43<49:38,  3.82it/s]
 97%|█████████▋| 360101/371472 [6:29:43<52:10,  3.63it/s]
 97%|█████████▋| 360102/371472 [6:29:44<50:07,  3.78it/s]
 97%|█████████▋| 360103/371472 [6:29:44<1:06:58,  2.83it/s]
 97%|█████████▋| 360104/371472 [6:29:44<1:03:17,  2.99it/s]
 97%|█████████▋| 360105/371472 [6:29:45<57:26,  3.30it/s]  
 97%|█████████▋| 360106/371472 [6:29:45<57:24,  3.30it/s]
 97%|█████████▋| 360107/371472 [6:29:45<54:16,  3.49it/s]
 97%|█████████▋| 360108/371472 [6:29:46<56:20,  3.36it/s]
 97%|█████████▋| 360109/371472 [6:29:46<55:18,  3.42it/s]
 97%|█████████▋| 360110/371472 [6:29:46<54:47,  3.46it/s]
 97%|█████████▋| 360111/371472 [6:29:46<55:34,  3.41it/s]
 97%|█████████▋| 360112/371472 [6:29:47<53:36,  3.53it/s]
 97%|█████████▋| 360113/371472 [6:29:47<52:12,  3.63it/s]
 97%|█████████▋| 360114/371472 [6:29:47<51:43,  3.66it/s]
 97%|█████████▋| 360115/371472 [6:29:47<50:25,  3.75it/s]
 97%|█████████▋| 360116/371472 [6:29:48<50:18,  3.76it/s]
 97%|█████████▋| 360117/371472 [6:29:48<50:47,  3.73it/s]
 97%|█████████▋| 360118/371472 [6:29:48<51:24,  3.68it/s]
 97%|█████████▋| 360119/371472 [6:29:49<51:57,  3.64it/s]
 97%|█████████▋| 360120/371472 [6:29:49<49:59,  3.78it/s]
                                                         
{'loss': 2.5743, 'learning_rate': 1.2751836928182033e-07, 'epoch': 15.51}
+
 97%|█████████▋| 360120/371472 [6:29:49<49:59,  3.78it/s]
 97%|█████████▋| 360121/371472 [6:29:49<49:38,  3.81it/s]
 97%|█████████▋| 360122/371472 [6:29:49<49:05,  3.85it/s]
 97%|█████████▋| 360123/371472 [6:29:50<49:52,  3.79it/s]
 97%|█████████▋| 360124/371472 [6:29:50<52:27,  3.61it/s]
 97%|█████████▋| 360125/371472 [6:29:50<54:07,  3.49it/s]
 97%|█████████▋| 360126/371472 [6:29:50<54:02,  3.50it/s]
 97%|█████████▋| 360127/371472 [6:29:51<56:26,  3.35it/s]
 97%|█████████▋| 360128/371472 [6:29:51<53:53,  3.51it/s]
 97%|█████████▋| 360129/371472 [6:29:51<52:59,  3.57it/s]
 97%|█████████▋| 360130/371472 [6:29:52<52:03,  3.63it/s]
 97%|█████████▋| 360131/371472 [6:29:52<52:03,  3.63it/s]
 97%|█████████▋| 360132/371472 [6:29:52<54:23,  3.47it/s]
 97%|█████████▋| 360133/371472 [6:29:52<51:53,  3.64it/s]
 97%|█████████▋| 360134/371472 [6:29:53<50:36,  3.73it/s]
 97%|█████████▋| 360135/371472 [6:29:53<48:42,  3.88it/s]
 97%|█████████▋| 360136/371472 [6:29:53<50:16,  3.76it/s]
 97%|█████████▋| 360137/371472 [6:29:54<53:03,  3.56it/s]
 97%|█████████▋| 360138/371472 [6:29:54<53:11,  3.55it/s]
 97%|█████████▋| 360139/371472 [6:29:54<56:51,  3.32it/s]
 97%|█████████▋| 360140/371472 [6:29:54<54:58,  3.44it/s]
                                                         
{'loss': 2.4876, 'learning_rate': 1.274698873063414e-07, 'epoch': 15.51}
+
 97%|█████████▋| 360140/371472 [6:29:54<54:58,  3.44it/s]
 97%|█████████▋| 360141/371472 [6:29:55<53:43,  3.52it/s]
 97%|█████████▋| 360142/371472 [6:29:55<54:38,  3.46it/s]
 97%|█████████▋| 360143/371472 [6:29:55<52:11,  3.62it/s]
 97%|█████████▋| 360144/371472 [6:29:55<50:25,  3.74it/s]
 97%|█████████▋| 360145/371472 [6:29:56<50:44,  3.72it/s]
 97%|█████████▋| 360146/371472 [6:29:56<51:52,  3.64it/s]
 97%|█████████▋| 360147/371472 [6:29:56<58:22,  3.23it/s]
 97%|█████████▋| 360148/371472 [6:29:57<56:49,  3.32it/s]
 97%|█████████▋| 360149/371472 [6:29:57<55:19,  3.41it/s]
 97%|█████████▋| 360150/371472 [6:29:57<56:37,  3.33it/s]
 97%|█████████▋| 360151/371472 [6:29:58<59:11,  3.19it/s]
 97%|█████████▋| 360152/371472 [6:29:58<57:30,  3.28it/s]
 97%|█████████▋| 360153/371472 [6:29:58<58:33,  3.22it/s]
 97%|█████████▋| 360154/371472 [6:29:59<56:56,  3.31it/s]
 97%|█████████▋| 360155/371472 [6:29:59<55:13,  3.42it/s]
 97%|█████████▋| 360156/371472 [6:29:59<53:52,  3.50it/s]
 97%|█████████▋| 360157/371472 [6:29:59<54:49,  3.44it/s]
 97%|█████████▋| 360158/371472 [6:30:00<55:38,  3.39it/s]
 97%|█████████▋| 360159/371472 [6:30:00<54:36,  3.45it/s]
 97%|█████████▋| 360160/371472 [6:30:00<56:52,  3.31it/s]
                                                         
{'loss': 2.5064, 'learning_rate': 1.2742140533086256e-07, 'epoch': 15.51}
+
 97%|█████████▋| 360160/371472 [6:30:00<56:52,  3.31it/s]
 97%|█████████▋| 360161/371472 [6:30:01<54:11,  3.48it/s]
 97%|█████████▋| 360162/371472 [6:30:01<59:11,  3.19it/s]
 97%|█████████▋| 360163/371472 [6:30:01<58:44,  3.21it/s]
 97%|█████████▋| 360164/371472 [6:30:02<59:13,  3.18it/s]
 97%|█████████▋| 360165/371472 [6:30:02<55:56,  3.37it/s]
 97%|█████████▋| 360166/371472 [6:30:02<53:32,  3.52it/s]
 97%|█████████▋| 360167/371472 [6:30:02<53:02,  3.55it/s]
 97%|█████████▋| 360168/371472 [6:30:03<53:19,  3.53it/s]
 97%|█████████▋| 360169/371472 [6:30:03<52:49,  3.57it/s]
 97%|█████████▋| 360170/371472 [6:30:03<51:02,  3.69it/s]
 97%|█████████▋| 360171/371472 [6:30:03<49:45,  3.79it/s]
 97%|█████████▋| 360172/371472 [6:30:04<57:26,  3.28it/s]
 97%|█████████▋| 360173/371472 [6:30:04<56:24,  3.34it/s]
 97%|█████████▋| 360174/371472 [6:30:04<53:31,  3.52it/s]
 97%|█████████▋| 360175/371472 [6:30:05<52:34,  3.58it/s]
 97%|█████████▋| 360176/371472 [6:30:05<53:05,  3.55it/s]
 97%|█████████▋| 360177/371472 [6:30:05<52:25,  3.59it/s]
 97%|█████████▋| 360178/371472 [6:30:05<50:44,  3.71it/s]
 97%|█████████▋| 360179/371472 [6:30:06<51:40,  3.64it/s]
 97%|█████████▋| 360180/371472 [6:30:06<49:21,  3.81it/s]
                                                         
{'loss': 2.6562, 'learning_rate': 1.273729233553836e-07, 'epoch': 15.51}
+
 97%|█████████▋| 360180/371472 [6:30:06<49:21,  3.81it/s]
 97%|█████████▋| 360181/371472 [6:30:06<50:01,  3.76it/s]
 97%|█████████▋| 360182/371472 [6:30:06<50:55,  3.70it/s]
 97%|█████████▋| 360183/371472 [6:30:07<52:28,  3.59it/s]
 97%|█████████▋| 360184/371472 [6:30:07<52:08,  3.61it/s]
 97%|█████████▋| 360185/371472 [6:30:07<55:28,  3.39it/s]
 97%|█████████▋| 360186/371472 [6:30:08<55:27,  3.39it/s]
 97%|█████████▋| 360187/371472 [6:30:08<54:42,  3.44it/s]
 97%|█████████▋| 360188/371472 [6:30:08<54:15,  3.47it/s]
 97%|█████████▋| 360189/371472 [6:30:08<52:41,  3.57it/s]
 97%|█████████▋| 360190/371472 [6:30:09<56:55,  3.30it/s]
 97%|█████████▋| 360191/371472 [6:30:09<54:10,  3.47it/s]
 97%|█████████▋| 360192/371472 [6:30:09<53:43,  3.50it/s]
 97%|██��██████▋| 360193/371472 [6:30:10<51:55,  3.62it/s]
 97%|█████████▋| 360194/371472 [6:30:10<51:30,  3.65it/s]
 97%|█████████▋| 360195/371472 [6:30:10<50:48,  3.70it/s]
 97%|█████████▋| 360196/371472 [6:30:10<52:50,  3.56it/s]
 97%|█████████▋| 360197/371472 [6:30:11<51:58,  3.62it/s]
 97%|█████████▋| 360198/371472 [6:30:11<52:09,  3.60it/s]
 97%|█████████▋| 360199/371472 [6:30:11<51:08,  3.67it/s]
 97%|█████████▋| 360200/371472 [6:30:12<49:05,  3.83it/s]
                                                         
{'loss': 2.6843, 'learning_rate': 1.2732444137990478e-07, 'epoch': 15.51}
+
 97%|█████████▋| 360200/371472 [6:30:12<49:05,  3.83it/s]
 97%|█████████▋| 360201/371472 [6:30:12<48:49,  3.85it/s]
 97%|█████████▋| 360202/371472 [6:30:12<48:05,  3.91it/s]
 97%|█████████▋| 360203/371472 [6:30:12<48:19,  3.89it/s]
 97%|█████████▋| 360204/371472 [6:30:13<53:29,  3.51it/s]
 97%|█████████▋| 360205/371472 [6:30:13<54:36,  3.44it/s]
 97%|█████████▋| 360206/371472 [6:30:13<53:28,  3.51it/s]
 97%|█████████▋| 360207/371472 [6:30:13<52:37,  3.57it/s]
 97%|█████████▋| 360208/371472 [6:30:14<52:54,  3.55it/s]
 97%|█████████▋| 360209/371472 [6:30:14<54:23,  3.45it/s]
 97%|█████████▋| 360210/371472 [6:30:14<52:58,  3.54it/s]
 97%|█████████▋| 360211/371472 [6:30:15<52:20,  3.59it/s]
 97%|█████████▋| 360212/371472 [6:30:15<50:20,  3.73it/s]
 97%|█████████▋| 360213/371472 [6:30:15<48:47,  3.85it/s]
 97%|█████████▋| 360214/371472 [6:30:15<48:42,  3.85it/s]
 97%|█████████▋| 360215/371472 [6:30:16<53:30,  3.51it/s]
 97%|█████████▋| 360216/371472 [6:30:16<55:01,  3.41it/s]
 97%|█████████▋| 360217/371472 [6:30:16<54:31,  3.44it/s]
 97%|█████████▋| 360218/371472 [6:30:17<54:58,  3.41it/s]
 97%|█████████▋| 360219/371472 [6:30:17<54:10,  3.46it/s]
 97%|█████████▋| 360220/371472 [6:30:17<54:35,  3.44it/s]
                                                         
{'loss': 2.4722, 'learning_rate': 1.2727595940442582e-07, 'epoch': 15.52}
+
 97%|█████████▋| 360220/371472 [6:30:17<54:35,  3.44it/s]
 97%|█████████▋| 360221/371472 [6:30:17<54:43,  3.43it/s]
 97%|█████████▋| 360222/371472 [6:30:18<53:10,  3.53it/s]
 97%|█████████▋| 360223/371472 [6:30:18<51:49,  3.62it/s]
 97%|█████████▋| 360224/371472 [6:30:18<51:41,  3.63it/s]
 97%|█████████▋| 360225/371472 [6:30:19<51:26,  3.64it/s]
 97%|█████████▋| 360226/371472 [6:30:19<50:29,  3.71it/s]
 97%|█████████▋| 360227/371472 [6:30:19<49:53,  3.76it/s]
 97%|█████████▋| 360228/371472 [6:30:19<48:09,  3.89it/s]
 97%|█████████▋| 360229/371472 [6:30:20<47:11,  3.97it/s]
 97%|█████████▋| 360230/371472 [6:30:20<48:48,  3.84it/s]
 97%|█████████▋| 360231/371472 [6:30:20<48:55,  3.83it/s]
 97%|█████████▋| 360232/371472 [6:30:20<49:41,  3.77it/s]
 97%|█████████▋| 360233/371472 [6:30:21<51:54,  3.61it/s]
 97%|█████████▋| 360234/371472 [6:30:21<55:52,  3.35it/s]
 97%|█████████▋| 360235/371472 [6:30:21<55:54,  3.35it/s]
 97%|█████████▋| 360236/371472 [6:30:22<54:04,  3.46it/s]
 97%|█████████▋| 360237/371472 [6:30:22<53:08,  3.52it/s]
 97%|█████████▋| 360238/371472 [6:30:22<52:38,  3.56it/s]
 97%|█████████▋| 360239/371472 [6:30:22<52:58,  3.53it/s]
 97%|█████████▋| 360240/371472 [6:30:23<52:50,  3.54it/s]
                                                         
{'loss': 2.669, 'learning_rate': 1.27227477428947e-07, 'epoch': 15.52}
+
 97%|█████████▋| 360240/371472 [6:30:23<52:50,  3.54it/s]
 97%|█████████▋| 360241/371472 [6:30:23<50:42,  3.69it/s]
 97%|█████████▋| 360242/371472 [6:30:23<56:46,  3.30it/s]
 97%|█████████▋| 360243/371472 [6:30:24<57:15,  3.27it/s]
 97%|█████████▋| 360244/371472 [6:30:24<58:07,  3.22it/s]
 97%|█████████▋| 360245/371472 [6:30:24<54:57,  3.40it/s]
 97%|█████████▋| 360246/371472 [6:30:24<53:03,  3.53it/s]
 97%|█████████▋| 360247/371472 [6:30:25<52:16,  3.58it/s]
 97%|█████████▋| 360248/371472 [6:30:25<57:42,  3.24it/s]
 97%|█████████▋| 360249/371472 [6:30:25<53:52,  3.47it/s]
 97%|█████████▋| 360250/371472 [6:30:26<53:04,  3.52it/s]
 97%|█████████▋| 360251/371472 [6:30:26<54:19,  3.44it/s]
 97%|█████████▋| 360252/371472 [6:30:26<1:01:01,  3.06it/s]
 97%|█████████▋| 360253/371472 [6:30:27<57:49,  3.23it/s]  
 97%|█████████▋| 360254/371472 [6:30:27<57:15,  3.26it/s]
 97%|█████████▋| 360255/371472 [6:30:27<58:32,  3.19it/s]
 97%|█████████▋| 360256/371472 [6:30:28<1:01:08,  3.06it/s]
 97%|█████████▋| 360257/371472 [6:30:28<58:09,  3.21it/s]  
 97%|█████████▋| 360258/371472 [6:30:28<56:25,  3.31it/s]
 97%|█████████▋| 360259/371472 [6:30:28<55:31,  3.37it/s]
 97%|█████████▋| 360260/371472 [6:30:29<53:06,  3.52it/s]
                                                         
{'loss': 2.6349, 'learning_rate': 1.2717899545346805e-07, 'epoch': 15.52}
+
 97%|█████████▋| 360260/371472 [6:30:29<53:06,  3.52it/s]
 97%|█████████▋| 360261/371472 [6:30:29<52:50,  3.54it/s]
 97%|█████████▋| 360262/371472 [6:30:29<53:34,  3.49it/s]
 97%|█████████▋| 360263/371472 [6:30:30<51:57,  3.60it/s]
 97%|█████████▋| 360264/371472 [6:30:30<55:24,  3.37it/s]
 97%|█████████▋| 360265/371472 [6:30:30<53:30,  3.49it/s]
 97%|█████████▋| 360266/371472 [6:30:30<53:02,  3.52it/s]
 97%|█████████▋| 360267/371472 [6:30:31<51:05,  3.66it/s]
 97%|█████████▋| 360268/371472 [6:30:31<56:39,  3.30it/s]
 97%|█████████▋| 360269/371472 [6:30:31<53:50,  3.47it/s]
 97%|█████████▋| 360270/371472 [6:30:32<53:28,  3.49it/s]
 97%|█████████▋| 360271/371472 [6:30:32<53:10,  3.51it/s]
 97%|█████████▋| 360272/371472 [6:30:32<52:04,  3.58it/s]
 97%|█████████▋| 360273/371472 [6:30:32<52:01,  3.59it/s]
 97%|█████████▋| 360274/371472 [6:30:33<53:34,  3.48it/s]
 97%|█████████▋| 360275/371472 [6:30:33<50:21,  3.71it/s]
 97%|█████████▋| 360276/371472 [6:30:33<49:41,  3.75it/s]
 97%|█████████▋| 360277/371472 [6:30:33<51:11,  3.64it/s]
 97%|█████████▋| 360278/371472 [6:30:34<51:18,  3.64it/s]
 97%|█████████▋| 360279/371472 [6:30:34<50:19,  3.71it/s]
 97%|█████████▋| 360280/371472 [6:30:34<54:07,  3.45it/s]
                                                         
{'loss': 2.6174, 'learning_rate': 1.271305134779892e-07, 'epoch': 15.52}
+
 97%|█████████▋| 360280/371472 [6:30:34<54:07,  3.45it/s]
 97%|█████████▋| 360281/371472 [6:30:35<58:33,  3.19it/s]
 97%|█████████▋| 360282/371472 [6:30:35<57:06,  3.27it/s]
 97%|█████████▋| 360283/371472 [6:30:35<55:08,  3.38it/s]
 97%|█████████▋| 360284/371472 [6:30:36<53:47,  3.47it/s]
 97%|█████████▋| 360285/371472 [6:30:36<53:02,  3.52it/s]
 97%|█████████▋| 360286/371472 [6:30:36<52:47,  3.53it/s]
 97%|█████████▋| 360287/371472 [6:30:36<50:47,  3.67it/s]
 97%|█████████▋| 360288/371472 [6:30:37<52:30,  3.55it/s]
 97%|█████████▋| 360289/371472 [6:30:37<51:26,  3.62it/s]
 97%|█████████▋| 360290/371472 [6:30:37<51:06,  3.65it/s]
 97%|█████████▋| 360291/371472 [6:30:37<49:29,  3.76it/s]
 97%|█████████▋| 360292/371472 [6:30:38<49:55,  3.73it/s]
 97%|█████████▋| 360293/371472 [6:30:38<51:31,  3.62it/s]
 97%|█████████▋| 360294/371472 [6:30:38<50:20,  3.70it/s]
 97%|█████████▋| 360295/371472 [6:30:39<52:57,  3.52it/s]
 97%|█████████▋| 360296/371472 [6:30:39<56:39,  3.29it/s]
 97%|█████████▋| 360297/371472 [6:30:39<54:45,  3.40it/s]
 97%|█████████▋| 360298/371472 [6:30:39<55:49,  3.34it/s]
 97%|█████████▋| 360299/371472 [6:30:40<54:56,  3.39it/s]
 97%|█████████▋| 360300/371472 [6:30:40<52:07,  3.57it/s]
                                                         
{'loss': 2.5786, 'learning_rate': 1.2708203150251024e-07, 'epoch': 15.52}
+
 97%|█████████▋| 360300/371472 [6:30:40<52:07,  3.57it/s]
 97%|█████████▋| 360301/371472 [6:30:40<54:06,  3.44it/s]
 97%|█████████▋| 360302/371472 [6:30:41<52:46,  3.53it/s]
 97%|█████████▋| 360303/371472 [6:30:41<54:19,  3.43it/s]
 97%|█████████▋| 360304/371472 [6:30:41<52:30,  3.55it/s]
 97%|█████████▋| 360305/371472 [6:30:41<51:14,  3.63it/s]
 97%|█████████▋| 360306/371472 [6:30:42<48:56,  3.80it/s]
 97%|█████████▋| 360307/371472 [6:30:42<51:11,  3.64it/s]
 97%|��████████▋| 360308/371472 [6:30:42<51:20,  3.62it/s]
 97%|█████████▋| 360309/371472 [6:30:42<49:58,  3.72it/s]
 97%|█████████▋| 360310/371472 [6:30:43<49:12,  3.78it/s]
 97%|█████████▋| 360311/371472 [6:30:43<52:14,  3.56it/s]
 97%|█████████▋| 360312/371472 [6:30:43<52:25,  3.55it/s]
 97%|█████████▋| 360313/371472 [6:30:44<52:32,  3.54it/s]
 97%|█████████▋| 360314/371472 [6:30:44<56:22,  3.30it/s]
 97%|█████████▋| 360315/371472 [6:30:44<53:03,  3.50it/s]
 97%|█████████▋| 360316/371472 [6:30:45<58:59,  3.15it/s]
 97%|█████████▋| 360317/371472 [6:30:45<57:58,  3.21it/s]
 97%|█████████▋| 360318/371472 [6:30:45<57:20,  3.24it/s]
 97%|█████████▋| 360319/371472 [6:30:45<53:53,  3.45it/s]
 97%|█████████▋| 360320/371472 [6:30:46<58:26,  3.18it/s]
                                                         
{'loss': 2.7205, 'learning_rate': 1.2703354952703142e-07, 'epoch': 15.52}
+
 97%|█████████▋| 360320/371472 [6:30:46<58:26,  3.18it/s]
 97%|█████████▋| 360321/371472 [6:30:46<54:28,  3.41it/s]
 97%|█████████▋| 360322/371472 [6:30:46<54:33,  3.41it/s]
 97%|█████████▋| 360323/371472 [6:30:47<53:33,  3.47it/s]
 97%|█████████▋| 360324/371472 [6:30:47<52:10,  3.56it/s]
 97%|█████████▋| 360325/371472 [6:30:47<53:21,  3.48it/s]
 97%|█████████▋| 360326/371472 [6:30:47<50:53,  3.65it/s]
 97%|█████████▋| 360327/371472 [6:30:48<50:42,  3.66it/s]
 97%|█████████▋| 360328/371472 [6:30:48<51:11,  3.63it/s]
 97%|█████████▋| 360329/371472 [6:30:48<50:04,  3.71it/s]
 97%|█████████▋| 360330/371472 [6:30:49<48:54,  3.80it/s]
 97%|█████████▋| 360331/371472 [6:30:49<51:15,  3.62it/s]
 97%|█████████▋| 360332/371472 [6:30:49<50:16,  3.69it/s]
 97%|█████████▋| 360333/371472 [6:30:49<52:27,  3.54it/s]
 97%|█████████▋| 360334/371472 [6:30:50<50:43,  3.66it/s]
 97%|█████████▋| 360335/371472 [6:30:50<51:32,  3.60it/s]
 97%|█████████▋| 360336/371472 [6:30:50<50:57,  3.64it/s]
 97%|█████████▋| 360337/371472 [6:30:51<53:53,  3.44it/s]
 97%|█████████▋| 360338/371472 [6:30:51<52:49,  3.51it/s]
 97%|█████████▋| 360339/371472 [6:30:51<53:10,  3.49it/s]
 97%|█████████▋| 360340/371472 [6:30:51<55:44,  3.33it/s]
                                                         
{'loss': 2.5017, 'learning_rate': 1.2698506755155247e-07, 'epoch': 15.52}
+
 97%|█████████▋| 360340/371472 [6:30:51<55:44,  3.33it/s]
 97%|█████████▋| 360341/371472 [6:30:52<57:32,  3.22it/s]
 97%|█████████▋| 360342/371472 [6:30:52<55:35,  3.34it/s]
 97%|█████████▋| 360343/371472 [6:30:52<55:19,  3.35it/s]
 97%|█████████▋| 360344/371472 [6:30:53<55:28,  3.34it/s]
 97%|█████████▋| 360345/371472 [6:30:53<57:27,  3.23it/s]
 97%|█████████▋| 360346/371472 [6:30:53<55:38,  3.33it/s]
 97%|█████████▋| 360347/371472 [6:30:54<53:47,  3.45it/s]
 97%|█████████▋| 360348/371472 [6:30:54<56:29,  3.28it/s]
 97%|█████████▋| 360349/371472 [6:30:54<54:32,  3.40it/s]
 97%|█████████▋| 360350/371472 [6:30:54<52:29,  3.53it/s]
 97%|█████████▋| 360351/371472 [6:30:55<57:37,  3.22it/s]
 97%|█████████▋| 360352/371472 [6:30:55<55:20,  3.35it/s]
 97%|█████████▋| 360353/371472 [6:30:55<54:38,  3.39it/s]
 97%|█████████▋| 360354/371472 [6:30:56<59:35,  3.11it/s]
 97%|█████████▋| 360355/371472 [6:30:56<55:39,  3.33it/s]
 97%|█████████▋| 360356/371472 [6:30:56<56:08,  3.30it/s]
 97%|█████████▋| 360357/371472 [6:30:57<57:37,  3.22it/s]
 97%|█████████▋| 360358/371472 [6:30:57<1:00:38,  3.05it/s]
 97%|█████████▋| 360359/371472 [6:30:57<57:51,  3.20it/s]  
 97%|█████████▋| 360360/371472 [6:30:57<55:54,  3.31it/s]
                                                         
{'loss': 2.5386, 'learning_rate': 1.2693658557607362e-07, 'epoch': 15.52}
+
 97%|█████████▋| 360360/371472 [6:30:57<55:54,  3.31it/s]
 97%|█████████▋| 360361/371472 [6:30:58<54:24,  3.40it/s]
 97%|█████████▋| 360362/371472 [6:30:58<53:14,  3.48it/s]
 97%|█████████▋| 360363/371472 [6:30:58<52:02,  3.56it/s]
 97%|█████████▋| 360364/371472 [6:30:59<51:14,  3.61it/s]
 97%|█████████▋| 360365/371472 [6:30:59<53:37,  3.45it/s]
 97%|█████████▋| 360366/371472 [6:30:59<53:10,  3.48it/s]
 97%|█████████▋| 360367/371472 [6:30:59<53:01,  3.49it/s]
 97%|█████████▋| 360368/371472 [6:31:00<52:35,  3.52it/s]
 97%|█████████▋| 360369/371472 [6:31:00<50:35,  3.66it/s]
 97%|█████████▋| 360370/371472 [6:31:00<49:57,  3.70it/s]
 97%|█████████▋| 360371/371472 [6:31:01<48:58,  3.78it/s]
 97%|█████████▋| 360372/371472 [6:31:01<50:34,  3.66it/s]
 97%|█████████▋| 360373/371472 [6:31:01<50:08,  3.69it/s]
 97%|█████████▋| 360374/371472 [6:31:01<56:32,  3.27it/s]
 97%|█████████▋| 360375/371472 [6:31:02<53:46,  3.44it/s]
 97%|█████████▋| 360376/371472 [6:31:02<57:01,  3.24it/s]
 97%|█████████▋| 360377/371472 [6:31:02<54:48,  3.37it/s]
 97%|█████████▋| 360378/371472 [6:31:03<57:16,  3.23it/s]
 97%|█████████▋| 360379/371472 [6:31:03<55:48,  3.31it/s]
 97%|█████████▋| 360380/371472 [6:31:03<53:16,  3.47it/s]
                                                         
{'loss': 2.5244, 'learning_rate': 1.268881036005947e-07, 'epoch': 15.52}
+
 97%|█████████▋| 360380/371472 [6:31:03<53:16,  3.47it/s]
 97%|█████████▋| 360381/371472 [6:31:04<54:09,  3.41it/s]
 97%|█████████▋| 360382/371472 [6:31:04<51:41,  3.58it/s]
 97%|█████████▋| 360383/371472 [6:31:04<51:03,  3.62it/s]
 97%|█████████▋| 360384/371472 [6:31:04<55:15,  3.34it/s]
 97%|█████████▋| 360385/371472 [6:31:05<54:17,  3.40it/s]
 97%|█████████▋| 360386/371472 [6:31:05<55:01,  3.36it/s]
 97%|█████████▋| 360387/371472 [6:31:05<54:05,  3.42it/s]
 97%|█████████▋| 360388/371472 [6:31:05<51:08,  3.61it/s]
 97%|█████████▋| 360389/371472 [6:31:06<51:15,  3.60it/s]
 97%|█████████▋| 360390/371472 [6:31:06<53:16,  3.47it/s]
 97%|█████████▋| 360391/371472 [6:31:06<54:07,  3.41it/s]
 97%|█████████▋| 360392/371472 [6:31:07<52:55,  3.49it/s]
 97%|█████████▋| 360393/371472 [6:31:07<51:13,  3.61it/s]
 97%|█████████▋| 360394/371472 [6:31:07<52:43,  3.50it/s]
 97%|█████████▋| 360395/371472 [6:31:08<1:00:25,  3.06it/s]
 97%|█████████▋| 360396/371472 [6:31:08<55:31,  3.32it/s]  
 97%|█████████▋| 360397/371472 [6:31:08<53:52,  3.43it/s]
 97%|█████████▋| 360398/371472 [6:31:08<53:46,  3.43it/s]
 97%|█████████▋| 360399/371472 [6:31:09<54:00,  3.42it/s]
 97%|█████████▋| 360400/371472 [6:31:09<53:07,  3.47it/s]
                                                         
{'loss': 2.4744, 'learning_rate': 1.2683962162511584e-07, 'epoch': 15.52}
+
 97%|█████████▋| 360400/371472 [6:31:09<53:07,  3.47it/s]
 97%|█████████▋| 360401/371472 [6:31:09<51:26,  3.59it/s]
 97%|█████████▋| 360402/371472 [6:31:10<53:47,  3.43it/s]
 97%|█████████▋| 360403/371472 [6:31:10<51:26,  3.59it/s]
 97%|█████████▋| 360404/371472 [6:31:10<54:39,  3.38it/s]
 97%|█████████▋| 360405/371472 [6:31:10<52:20,  3.52it/s]
 97%|█████████▋| 360406/371472 [6:31:11<55:01,  3.35it/s]
 97%|█████████▋| 360407/371472 [6:31:11<59:21,  3.11it/s]
 97%|█████████▋| 360408/371472 [6:31:11<54:53,  3.36it/s]
 97%|█████████▋| 360409/371472 [6:31:12<56:50,  3.24it/s]
 97%|█████████▋| 360410/371472 [6:31:12<55:02,  3.35it/s]
 97%|█████████▋| 360411/371472 [6:31:12<55:33,  3.32it/s]
 97%|█████████▋| 360412/371472 [6:31:13<52:17,  3.53it/s]
 97%|█████████▋| 360413/371472 [6:31:13<51:08,  3.60it/s]
 97%|█████████▋| 360414/371472 [6:31:13<50:38,  3.64it/s]
 97%|█████████▋| 360415/371472 [6:31:13<52:10,  3.53it/s]
 97%|█████████▋| 360416/371472 [6:31:14<49:45,  3.70it/s]
 97%|█████████▋| 360417/371472 [6:31:14<49:31,  3.72it/s]
 97%|█████████▋| 360418/371472 [6:31:14<49:58,  3.69it/s]
 97%|█████████▋| 360419/371472 [6:31:14<50:22,  3.66it/s]
 97%|█████████▋| 360420/371472 [6:31:15<51:06,  3.60it/s]
                                                         
{'loss': 2.528, 'learning_rate': 1.267911396496369e-07, 'epoch': 15.52}
+
 97%|█████████▋| 360420/371472 [6:31:15<51:06,  3.60it/s]
 97%|█████████▋| 360421/371472 [6:31:15<56:51,  3.24it/s]
 97%|█████████▋| 360422/371472 [6:31:15<55:38,  3.31it/s]
 97%|█████████▋| 360423/371472 [6:31:16<59:11,  3.11it/s]
 97%|█████████▋| 360424/371472 [6:31:16<57:12,  3.22it/s]
 97%|█████████▋| 360425/371472 [6:31:16<59:16,  3.11it/s]
 97%|█████████▋| 360426/371472 [6:31:17<58:22,  3.15it/s]
 97%|█████████▋| 360427/371472 [6:31:17<55:03,  3.34it/s]
 97%|█████████▋| 360428/371472 [6:31:17<54:34,  3.37it/s]
 97%|█████████▋| 360429/371472 [6:31:18<54:16,  3.39it/s]
 97%|█████████▋| 360430/371472 [6:31:18<53:17,  3.45it/s]
 97%|█████████▋| 360431/371472 [6:31:18<53:40,  3.43it/s]
 97%|█████████▋| 360432/371472 [6:31:18<51:14,  3.59it/s]
 97%|█████████▋| 360433/371472 [6:31:19<50:14,  3.66it/s]
 97%|█████████▋| 360434/371472 [6:31:19<49:48,  3.69it/s]
 97%|█████████▋| 360435/371472 [6:31:19<48:12,  3.82it/s]
 97%|█████████▋| 360436/371472 [6:31:19<51:27,  3.57it/s]
 97%|█████████▋| 360437/371472 [6:31:20<50:40,  3.63it/s]
 97%|█████████▋| 360438/371472 [6:31:20<51:04,  3.60it/s]
 97%|█████████▋| 360439/371472 [6:31:20<50:50,  3.62it/s]
 97%|█████████▋| 360440/371472 [6:31:21<49:55,  3.68it/s]
                                                         
{'loss': 2.5705, 'learning_rate': 1.2674265767415806e-07, 'epoch': 15.52}
+
 97%|█████████▋| 360440/371472 [6:31:21<49:55,  3.68it/s]
 97%|█████████▋| 360441/371472 [6:31:21<50:46,  3.62it/s]
 97%|█████████▋| 360442/371472 [6:31:21<53:42,  3.42it/s]
 97%|█████████▋| 360443/371472 [6:31:21<52:39,  3.49it/s]
 97%|█████████▋| 360444/371472 [6:31:22<58:04,  3.16it/s]
 97%|█████████▋| 360445/371472 [6:31:22<54:50,  3.35it/s]
 97%|█████████▋| 360446/371472 [6:31:22<53:01,  3.47it/s]
 97%|█████████▋| 360447/371472 [6:31:23<50:44,  3.62it/s]
 97%|█████████▋| 360448/371472 [6:31:23<50:13,  3.66it/s]
 97%|█████████▋| 360449/371472 [6:31:23<50:31,  3.64it/s]
 97%|█████████▋| 360450/371472 [6:31:23<49:37,  3.70it/s]
 97%|█████████▋| 360451/371472 [6:31:24<47:57,  3.83it/s]
 97%|█████████▋| 360452/371472 [6:31:24<49:09,  3.74it/s]
 97%|█████████▋| 360453/371472 [6:31:24<55:04,  3.33it/s]
 97%|█████████▋| 360454/371472 [6:31:25<53:23,  3.44it/s]
 97%|█████████▋| 360455/371472 [6:31:25<50:49,  3.61it/s]
 97%|█████████▋| 360456/371472 [6:31:25<54:02,  3.40it/s]
 97%|█████████▋| 360457/371472 [6:31:25<53:54,  3.41it/s]
 97%|█████████▋| 360458/371472 [6:31:26<53:36,  3.42it/s]
 97%|█████████▋| 360459/371472 [6:31:26<53:53,  3.41it/s]
 97%|█████████▋| 360460/371472 [6:31:26<53:12,  3.45it/s]
                                                         
{'loss': 2.517, 'learning_rate': 1.2669417569867913e-07, 'epoch': 15.53}
+
 97%|█████████▋| 360460/371472 [6:31:26<53:12,  3.45it/s]
 97%|█████████▋| 360461/371472 [6:31:27<52:13,  3.51it/s]
 97%|█████████▋| 360462/371472 [6:31:27<50:51,  3.61it/s]
 97%|█████████▋| 360463/371472 [6:31:27<50:03,  3.67it/s]
 97%|█████████▋| 360464/371472 [6:31:27<50:24,  3.64it/s]
 97%|█████████▋| 360465/371472 [6:31:28<52:43,  3.48it/s]
 97%|█████████▋| 360466/371472 [6:31:28<50:15,  3.65it/s]
 97%|█████████▋| 360467/371472 [6:31:28<49:28,  3.71it/s]
 97%|█████████▋| 360468/371472 [6:31:28<49:24,  3.71it/s]
 97%|█████████▋| 360469/371472 [6:31:29<51:22,  3.57it/s]
 97%|█████████▋| 360470/371472 [6:31:29<51:33,  3.56it/s]
 97%|█████████▋| 360471/371472 [6:31:29<50:35,  3.62it/s]
 97%|█████████▋| 360472/371472 [6:31:30<51:38,  3.55it/s]
 97%|█████████▋| 360473/371472 [6:31:30<52:57,  3.46it/s]
 97%|█████████▋| 360474/371472 [6:31:30<54:41,  3.35it/s]
 97%|█████████▋| 360475/371472 [6:31:31<56:36,  3.24it/s]
 97%|█████████▋| 360476/371472 [6:31:31<53:58,  3.40it/s]
 97%|█████████▋| 360477/371472 [6:31:31<52:34,  3.49it/s]
 97%|█████████▋| 360478/371472 [6:31:31<57:49,  3.17it/s]
 97%|█████████▋| 360479/371472 [6:31:32<55:12,  3.32it/s]
 97%|█████████▋| 360480/371472 [6:31:32<52:28,  3.49it/s]
                                                         
{'loss': 2.608, 'learning_rate': 1.2664569372320028e-07, 'epoch': 15.53}
+
 97%|█████████▋| 360480/371472 [6:31:32<52:28,  3.49it/s]
 97%|█████████▋| 360481/371472 [6:31:32<52:20,  3.50it/s]
 97%|█████████▋| 360482/371472 [6:31:33<54:38,  3.35it/s]
 97%|█████████▋| 360483/371472 [6:31:33<53:34,  3.42it/s]
 97%|█████████▋| 360484/371472 [6:31:33<51:24,  3.56it/s]
 97%|█████████▋| 360485/371472 [6:31:33<52:44,  3.47it/s]
 97%|█████████▋| 360486/371472 [6:31:34<50:46,  3.61it/s]
 97%|█████████▋| 360487/371472 [6:31:34<51:22,  3.56it/s]
 97%|█████████▋| 360488/371472 [6:31:34<50:35,  3.62it/s]
 97%|█████████▋| 360489/371472 [6:31:35<50:56,  3.59it/s]
 97%|█████████▋| 360490/371472 [6:31:35<49:44,  3.68it/s]
 97%|█████████▋| 360491/371472 [6:31:35<49:20,  3.71it/s]
 97%|█████████▋| 360492/371472 [6:31:35<49:58,  3.66it/s]
 97%|█████████▋| 360493/371472 [6:31:36<51:01,  3.59it/s]
 97%|█████████▋| 360494/371472 [6:31:36<50:17,  3.64it/s]
 97%|█████████▋| 360495/371472 [6:31:36<54:15,  3.37it/s]
 97%|█████████▋| 360496/371472 [6:31:36<52:21,  3.49it/s]
 97%|█████████▋| 360497/371472 [6:31:37<50:35,  3.62it/s]
 97%|█████████▋| 360498/371472 [6:31:37<53:10,  3.44it/s]
 97%|█████████▋| 360499/371472 [6:31:37<51:16,  3.57it/s]
 97%|█████████▋| 360500/371472 [6:31:38<51:42,  3.54it/s]
                                                         
{'loss': 2.499, 'learning_rate': 1.2659721174772133e-07, 'epoch': 15.53}
+
 97%|█████████▋| 360500/371472 [6:31:38<51:42,  3.54it/s]
 97%|█████████▋| 360501/371472 [6:31:38<52:05,  3.51it/s]
 97%|█████████▋| 360502/371472 [6:31:38<51:30,  3.55it/s]
 97%|█████████▋| 360503/371472 [6:31:38<48:58,  3.73it/s]
 97%|█████████▋| 360504/371472 [6:31:39<52:23,  3.49it/s]
 97%|█████████▋| 360505/371472 [6:31:39<50:06,  3.65it/s]
 97%|█████████▋| 360506/371472 [6:31:39<50:53,  3.59it/s]
 97%|█████████▋| 360507/371472 [6:31:40<53:20,  3.43it/s]
 97%|█████████▋| 360508/371472 [6:31:40<50:43,  3.60it/s]
 97%|█████████▋| 360509/371472 [6:31:40<57:05,  3.20it/s]
 97%|█████████▋| 360510/371472 [6:31:41<57:12,  3.19it/s]
 97%|█████████▋| 360511/371472 [6:31:41<54:04,  3.38it/s]
 97%|█████████▋| 360512/371472 [6:31:41<56:47,  3.22it/s]
 97%|█████████▋| 360513/371472 [6:31:41<53:57,  3.38it/s]
 97%|█████████▋| 360514/371472 [6:31:42<53:25,  3.42it/s]
 97%|█████████▋| 360515/371472 [6:31:42<56:55,  3.21it/s]
 97%|█████████▋| 360516/371472 [6:31:42<57:11,  3.19it/s]
 97%|█████████▋| 360517/371472 [6:31:43<53:19,  3.42it/s]
 97%|█████████▋| 360518/371472 [6:31:43<54:34,  3.35it/s]
 97%|█████████▋| 360519/371472 [6:31:43<57:28,  3.18it/s]
 97%|█████████▋| 360520/371472 [6:31:44<53:51,  3.39it/s]
                                                         
{'loss': 2.6176, 'learning_rate': 1.2654872977224248e-07, 'epoch': 15.53}
+
 97%|█████████▋| 360520/371472 [6:31:44<53:51,  3.39it/s]
 97%|█████████▋| 360521/371472 [6:31:44<51:31,  3.54it/s]
 97%|█████████▋| 360522/371472 [6:31:44<50:26,  3.62it/s]
 97%|█████████▋| 360523/371472 [6:31:44<50:26,  3.62it/s]
 97%|█████████▋| 360524/371472 [6:31:45<54:42,  3.34it/s]
 97%|█████████▋| 360525/371472 [6:31:45<51:13,  3.56it/s]
 97%|█████████▋| 360526/371472 [6:31:45<50:19,  3.62it/s]
 97%|█████████▋| 360527/371472 [6:31:45<50:40,  3.60it/s]
 97%|█████████▋| 360528/371472 [6:31:46<53:27,  3.41it/s]
 97%|█████████▋| 360529/371472 [6:31:46<56:13,  3.24it/s]
 97%|█████████▋| 360530/371472 [6:31:46<55:54,  3.26it/s]
 97%|█████████▋| 360531/371472 [6:31:47<53:07,  3.43it/s]
 97%|█████████▋| 360532/371472 [6:31:47<50:44,  3.59it/s]
 97%|█████████▋| 360533/371472 [6:31:47<48:16,  3.78it/s]
 97%|█████████▋| 360534/371472 [6:31:47<47:58,  3.80it/s]
 97%|█████████▋| 360535/371472 [6:31:48<54:23,  3.35it/s]
 97%|█████████▋| 360536/371472 [6:31:48<55:41,  3.27it/s]
 97%|█████████▋| 360537/371472 [6:31:48<54:38,  3.34it/s]
 97%|█████████▋| 360538/371472 [6:31:49<52:08,  3.49it/s]
 97%|█████████▋| 360539/371472 [6:31:49<51:19,  3.55it/s]
 97%|█████████▋| 360540/371472 [6:31:49<52:33,  3.47it/s]
                                                         
{'loss': 2.4552, 'learning_rate': 1.2650024779676355e-07, 'epoch': 15.53}
+
 97%|█████████▋| 360540/371472 [6:31:49<52:33,  3.47it/s]
 97%|█████████▋| 360541/371472 [6:31:50<51:48,  3.52it/s]
 97%|█████████▋| 360542/371472 [6:31:50<50:50,  3.58it/s]
 97%|█████████▋| 360543/371472 [6:31:50<49:28,  3.68it/s]
 97%|█████████▋| 360544/371472 [6:31:50<50:46,  3.59it/s]
 97%|█████████▋| 360545/371472 [6:31:51<48:39,  3.74it/s]
 97%|█████████▋| 360546/371472 [6:31:51<51:28,  3.54it/s]
 97%|█████████▋| 360547/371472 [6:31:51<49:13,  3.70it/s]
 97%|█████████▋| 360548/371472 [6:31:51<52:05,  3.49it/s]
 97%|█████████▋| 360549/371472 [6:31:52<49:20,  3.69it/s]
 97%|█████████▋| 360550/371472 [6:31:52<50:05,  3.63it/s]
 97%|█████████▋| 360551/371472 [6:31:52<49:57,  3.64it/s]
 97%|█████████▋| 360552/371472 [6:31:53<49:36,  3.67it/s]
 97%|█████████▋| 360553/371472 [6:31:53<51:29,  3.53it/s]
 97%|█████████▋| 360554/371472 [6:31:53<55:05,  3.30it/s]
 97%|█████████▋| 360555/371472 [6:31:53<52:40,  3.45it/s]
 97%|█████████▋| 360556/371472 [6:31:54<54:20,  3.35it/s]
 97%|█████████▋| 360557/371472 [6:31:54<54:47,  3.32it/s]
 97%|█████████▋| 360558/371472 [6:31:54<54:48,  3.32it/s]
 97%|█████████▋| 360559/371472 [6:31:55<53:46,  3.38it/s]
 97%|█████████▋| 360560/371472 [6:31:55<54:46,  3.32it/s]
                                                         
{'loss': 2.6149, 'learning_rate': 1.264517658212847e-07, 'epoch': 15.53}
+
 97%|█████████▋| 360560/371472 [6:31:55<54:46,  3.32it/s]
 97%|█████████▋| 360561/371472 [6:31:55<57:24,  3.17it/s]
 97%|█████████▋| 360562/371472 [6:31:56<58:01,  3.13it/s]
 97%|█████████▋| 360563/371472 [6:31:56<56:55,  3.19it/s]
 97%|█████████▋| 360564/371472 [6:31:56<55:41,  3.26it/s]
 97%|█████████▋| 360565/371472 [6:31:57<53:58,  3.37it/s]
 97%|█████████▋| 360566/371472 [6:31:57<52:57,  3.43it/s]
 97%|█████████▋| 360567/371472 [6:31:57<52:49,  3.44it/s]
 97%|█████████▋| 360568/371472 [6:31:57<52:39,  3.45it/s]
 97%|█████████▋| 360569/371472 [6:31:58<54:44,  3.32it/s]
 97%|█████████▋| 360570/371472 [6:31:58<55:13,  3.29it/s]
 97%|█████████▋| 360571/371472 [6:31:58<56:34,  3.21it/s]
 97%|█████████▋| 360572/371472 [6:31:59<56:00,  3.24it/s]
 97%|█████████▋| 360573/371472 [6:31:59<54:44,  3.32it/s]
 97%|█████████▋| 360574/371472 [6:31:59<51:31,  3.53it/s]
 97%|█████████▋| 360575/371472 [6:31:59<52:25,  3.46it/s]
 97%|█████████▋| 360576/371472 [6:32:00<51:24,  3.53it/s]
 97%|█████████▋| 360577/371472 [6:32:00<50:38,  3.59it/s]
 97%|█████████▋| 360578/371472 [6:32:00<49:18,  3.68it/s]
 97%|█████████▋| 360579/371472 [6:32:01<47:56,  3.79it/s]
 97%|█████████▋| 360580/371472 [6:32:01<46:39,  3.89it/s]
                                                         
{'loss': 2.5328, 'learning_rate': 1.2640328384580577e-07, 'epoch': 15.53}
+
 97%|█████████▋| 360580/371472 [6:32:01<46:39,  3.89it/s]
 97%|█████████▋| 360581/371472 [6:32:01<46:58,  3.86it/s]
 97%|█████████▋| 360582/371472 [6:32:01<47:22,  3.83it/s]
 97%|█████████▋| 360583/371472 [6:32:02<48:06,  3.77it/s]
 97%|█████████▋| 360584/371472 [6:32:02<55:40,  3.26it/s]
 97%|█████████▋| 360585/371472 [6:32:02<53:20,  3.40it/s]
 97%|█████████▋| 360586/371472 [6:32:03<53:03,  3.42it/s]
 97%|█████████▋| 360587/371472 [6:32:03<51:31,  3.52it/s]
 97%|█████████▋| 360588/371472 [6:32:03<52:05,  3.48it/s]
 97%|█████████▋| 360589/371472 [6:32:03<51:34,  3.52it/s]
 97%|█████████▋| 360590/371472 [6:32:04<53:01,  3.42it/s]
 97%|█████████▋| 360591/371472 [6:32:04<50:35,  3.58it/s]
 97%|█████████▋| 360592/371472 [6:32:04<48:43,  3.72it/s]
 97%|█████████▋| 360593/371472 [6:32:04<48:44,  3.72it/s]
 97%|█████████▋| 360594/371472 [6:32:05<48:58,  3.70it/s]
 97%|█████████▋| 360595/371472 [6:32:05<50:35,  3.58it/s]
 97%|█████████▋| 360596/371472 [6:32:05<49:11,  3.68it/s]
 97%|█████████▋| 360597/371472 [6:32:06<49:19,  3.67it/s]
 97%|████���████▋| 360598/371472 [6:32:06<49:43,  3.65it/s]
 97%|█████████▋| 360599/371472 [6:32:06<51:27,  3.52it/s]
 97%|█████████▋| 360600/371472 [6:32:06<49:23,  3.67it/s]
                                                         
{'loss': 2.5616, 'learning_rate': 1.2635480187032682e-07, 'epoch': 15.53}
+
 97%|█████████▋| 360600/371472 [6:32:06<49:23,  3.67it/s]
 97%|█████████▋| 360601/371472 [6:32:07<50:26,  3.59it/s]
 97%|█████████▋| 360602/371472 [6:32:07<50:39,  3.58it/s]
 97%|█████████▋| 360603/371472 [6:32:07<53:49,  3.37it/s]
 97%|█████████▋| 360604/371472 [6:32:08<55:01,  3.29it/s]
 97%|█████████▋| 360605/371472 [6:32:08<53:38,  3.38it/s]
 97%|█████████▋| 360606/371472 [6:32:08<52:13,  3.47it/s]
 97%|█████████▋| 360607/371472 [6:32:08<53:50,  3.36it/s]
 97%|█████████▋| 360608/371472 [6:32:09<52:17,  3.46it/s]
 97%|█████████▋| 360609/371472 [6:32:09<52:04,  3.48it/s]
 97%|█████████▋| 360610/371472 [6:32:09<50:41,  3.57it/s]
 97%|█████████▋| 360611/371472 [6:32:10<49:34,  3.65it/s]
 97%|█████████▋| 360612/371472 [6:32:10<49:50,  3.63it/s]
 97%|█████████▋| 360613/371472 [6:32:10<51:23,  3.52it/s]
 97%|█████████▋| 360614/371472 [6:32:10<52:19,  3.46it/s]
 97%|█████████▋| 360615/371472 [6:32:11<52:13,  3.46it/s]
 97%|█████████▋| 360616/371472 [6:32:11<50:59,  3.55it/s]
 97%|█████████▋| 360617/371472 [6:32:11<51:29,  3.51it/s]
 97%|█████████▋| 360618/371472 [6:32:12<51:35,  3.51it/s]
 97%|█████████▋| 360619/371472 [6:32:12<51:21,  3.52it/s]
 97%|█████████▋| 360620/371472 [6:32:12<50:36,  3.57it/s]
                                                         
{'loss': 2.5625, 'learning_rate': 1.2630631989484797e-07, 'epoch': 15.53}
+
 97%|█████████▋| 360620/371472 [6:32:12<50:36,  3.57it/s]
 97%|█████████▋| 360621/371472 [6:32:12<53:44,  3.37it/s]
 97%|█████████▋| 360622/371472 [6:32:13<52:06,  3.47it/s]
 97%|█████████▋| 360623/371472 [6:32:13<54:36,  3.31it/s]
 97%|█████████▋| 360624/371472 [6:32:13<52:36,  3.44it/s]
 97%|█████████▋| 360625/371472 [6:32:14<54:24,  3.32it/s]
 97%|█████████▋| 360626/371472 [6:32:14<53:23,  3.39it/s]
 97%|█████████▋| 360627/371472 [6:32:14<50:22,  3.59it/s]
 97%|█████████▋| 360628/371472 [6:32:14<49:59,  3.61it/s]
 97%|█████████▋| 360629/371472 [6:32:15<52:09,  3.46it/s]
 97%|█████████▋| 360630/371472 [6:32:15<51:46,  3.49it/s]
 97%|█████████▋| 360631/371472 [6:32:15<49:33,  3.65it/s]
 97%|█████████▋| 360632/371472 [6:32:16<48:21,  3.74it/s]
 97%|█████████▋| 360633/371472 [6:32:16<49:22,  3.66it/s]
 97%|█████████▋| 360634/371472 [6:32:16<52:05,  3.47it/s]
 97%|█████████▋| 360635/371472 [6:32:16<52:37,  3.43it/s]
 97%|█████████▋| 360636/371472 [6:32:17<51:39,  3.50it/s]
 97%|█████████▋| 360637/371472 [6:32:17<53:08,  3.40it/s]
 97%|█████████▋| 360638/371472 [6:32:17<51:59,  3.47it/s]
 97%|█████████▋| 360639/371472 [6:32:18<50:20,  3.59it/s]
 97%|█████████▋| 360640/371472 [6:32:18<49:20,  3.66it/s]
                                                         
{'loss': 2.5174, 'learning_rate': 1.2625783791936904e-07, 'epoch': 15.53}
+
 97%|█████████▋| 360640/371472 [6:32:18<49:20,  3.66it/s]
 97%|█████████▋| 360641/371472 [6:32:18<49:05,  3.68it/s]
 97%|█████████▋| 360642/371472 [6:32:18<51:05,  3.53it/s]
 97%|█████████▋| 360643/371472 [6:32:19<49:00,  3.68it/s]
 97%|█████████▋| 360644/371472 [6:32:19<50:20,  3.58it/s]
 97%|█████████▋| 360645/371472 [6:32:19<50:27,  3.58it/s]
 97%|█████████▋| 360646/371472 [6:32:19<50:14,  3.59it/s]
 97%|█████████▋| 360647/371472 [6:32:20<54:16,  3.32it/s]
 97%|█████████▋| 360648/371472 [6:32:20<58:55,  3.06it/s]
 97%|█████████▋| 360649/371472 [6:32:21<57:26,  3.14it/s]
 97%|█████████▋| 360650/371472 [6:32:21<55:15,  3.26it/s]
 97%|█████████▋| 360651/371472 [6:32:21<55:19,  3.26it/s]
 97%|█████████▋| 360652/371472 [6:32:21<53:51,  3.35it/s]
 97%|█████████▋| 360653/371472 [6:32:22<51:03,  3.53it/s]
 97%|█████████▋| 360654/371472 [6:32:22<52:05,  3.46it/s]
 97%|█████████▋| 360655/371472 [6:32:22<50:24,  3.58it/s]
 97%|█████████▋| 360656/371472 [6:32:23<52:48,  3.41it/s]
 97%|█████████▋| 360657/371472 [6:32:23<52:28,  3.43it/s]
 97%|█████████▋| 360658/371472 [6:32:23<54:50,  3.29it/s]
 97%|█████████▋| 360659/371472 [6:32:23<52:52,  3.41it/s]
 97%|█████████▋| 360660/371472 [6:32:24<50:03,  3.60it/s]
                                                         
{'loss': 2.4765, 'learning_rate': 1.2620935594389016e-07, 'epoch': 15.53}
+
 97%|█████████▋| 360660/371472 [6:32:24<50:03,  3.60it/s]
 97%|█████████▋| 360661/371472 [6:32:24<49:36,  3.63it/s]
 97%|█████████▋| 360662/371472 [6:32:24<48:51,  3.69it/s]
 97%|█████████▋| 360663/371472 [6:32:24<51:05,  3.53it/s]
 97%|█████████▋| 360664/371472 [6:32:25<50:23,  3.57it/s]
 97%|█████████▋| 360665/371472 [6:32:25<48:59,  3.68it/s]
 97%|█████████▋| 360666/371472 [6:32:25<47:01,  3.83it/s]
 97%|█████████▋| 360667/371472 [6:32:26<47:17,  3.81it/s]
 97%|█████████▋| 360668/371472 [6:32:26<47:06,  3.82it/s]
 97%|█████████▋| 360669/371472 [6:32:26<47:49,  3.76it/s]
 97%|█████████▋| 360670/371472 [6:32:26<48:38,  3.70it/s]
 97%|█████████▋| 360671/371472 [6:32:27<50:01,  3.60it/s]
 97%|█████████▋| 360672/371472 [6:32:27<50:40,  3.55it/s]
 97%|█████████▋| 360673/371472 [6:32:27<51:42,  3.48it/s]
 97%|█████████▋| 360674/371472 [6:32:28<55:59,  3.21it/s]
 97%|█████████▋| 360675/371472 [6:32:28<53:19,  3.37it/s]
 97%|█████████▋| 360676/371472 [6:32:28<53:10,  3.38it/s]
 97%|█████████▋| 360677/371472 [6:32:28<52:13,  3.45it/s]
 97%|█████████▋| 360678/371472 [6:32:29<50:04,  3.59it/s]
 97%|█████████▋| 360679/371472 [6:32:29<48:45,  3.69it/s]
 97%|█████████▋| 360680/371472 [6:32:29<48:29,  3.71it/s]
                                                         
{'loss': 2.5934, 'learning_rate': 1.2616087396841126e-07, 'epoch': 15.54}
+
 97%|█████████▋| 360680/371472 [6:32:29<48:29,  3.71it/s]
 97%|█████████▋| 360681/371472 [6:32:29<47:45,  3.77it/s]
 97%|█████████▋| 360682/371472 [6:32:30<47:16,  3.80it/s]
 97%|█████████▋| 360683/371472 [6:32:30<47:35,  3.78it/s]
 97%|█████████▋| 360684/371472 [6:32:30<49:52,  3.60it/s]
 97%|█████████▋| 360685/371472 [6:32:31<48:16,  3.72it/s]
 97%|█████████▋| 360686/371472 [6:32:31<53:08,  3.38it/s]
 97%|█████████▋| 360687/371472 [6:32:31<52:42,  3.41it/s]
 97%|█████████▋| 360688/371472 [6:32:31<53:05,  3.39it/s]
 97%|█████████▋| 360689/371472 [6:32:32<54:33,  3.29it/s]
 97%|█████████▋| 360690/371472 [6:32:32<51:42,  3.48it/s]
 97%|█████████▋| 360691/371472 [6:32:32<50:14,  3.58it/s]
 97%|█████████▋| 360692/371472 [6:32:33<51:42,  3.47it/s]
 97%|█████████▋| 360693/371472 [6:32:33<52:15,  3.44it/s]
 97%|█████████▋| 360694/371472 [6:32:33<49:19,  3.64it/s]
 97%|█████████▋| 360695/371472 [6:32:33<51:33,  3.48it/s]
 97%|█████████▋| 360696/371472 [6:32:34<50:55,  3.53it/s]
 97%|█████████▋| 360697/371472 [6:32:34<50:43,  3.54it/s]
 97%|█████████▋| 360698/371472 [6:32:34<53:28,  3.36it/s]
 97%|█████████▋| 360699/371472 [6:32:35<51:40,  3.47it/s]
 97%|█████████▋| 360700/371472 [6:32:35<49:45,  3.61it/s]
                                                         
{'loss': 2.5004, 'learning_rate': 1.2611239199293241e-07, 'epoch': 15.54}
+
 97%|█████████▋| 360700/371472 [6:32:35<49:45,  3.61it/s]
 97%|█████████▋| 360701/371472 [6:32:35<49:33,  3.62it/s]
 97%|█████████▋| 360702/371472 [6:32:35<48:59,  3.66it/s]
 97%|█████████▋| 360703/371472 [6:32:36<48:31,  3.70it/s]
 97%|█████████▋| 360704/371472 [6:32:36<52:14,  3.44it/s]
 97%|█████████▋| 360705/371472 [6:32:36<51:58,  3.45it/s]
 97%|█████████▋| 360706/371472 [6:32:37<59:04,  3.04it/s]
 97%|█████████▋| 360707/371472 [6:32:37<1:10:50,  2.53it/s]
 97%|█████████▋| 360708/371472 [6:32:38<1:05:34,  2.74it/s]
 97%|█████████▋| 360709/371472 [6:32:38<1:02:57,  2.85it/s]
 97%|█████████▋| 360710/371472 [6:32:38<1:03:09,  2.84it/s]
 97%|█████████▋| 360711/371472 [6:32:39<59:50,  3.00it/s]  
 97%|█████████▋| 360712/371472 [6:32:39<57:24,  3.12it/s]
 97%|���████████▋| 360713/371472 [6:32:39<57:19,  3.13it/s]
 97%|█████████▋| 360714/371472 [6:32:39<56:13,  3.19it/s]
 97%|█████████▋| 360715/371472 [6:32:40<52:56,  3.39it/s]
 97%|█████████▋| 360716/371472 [6:32:40<53:26,  3.35it/s]
 97%|█████████▋| 360717/371472 [6:32:40<52:00,  3.45it/s]
 97%|█████████▋| 360718/371472 [6:32:41<52:38,  3.40it/s]
 97%|█████████▋| 360719/371472 [6:32:41<53:06,  3.37it/s]
 97%|█████████▋| 360720/371472 [6:32:41<51:45,  3.46it/s]
                                                         
{'loss': 2.6951, 'learning_rate': 1.2606391001745346e-07, 'epoch': 15.54}
+
 97%|█████████▋| 360720/371472 [6:32:41<51:45,  3.46it/s]
 97%|█████████▋| 360721/371472 [6:32:41<50:36,  3.54it/s]
 97%|█████████▋| 360722/371472 [6:32:42<49:44,  3.60it/s]
 97%|█████████▋| 360723/371472 [6:32:42<49:21,  3.63it/s]
 97%|█████████▋| 360724/371472 [6:32:42<47:40,  3.76it/s]
 97%|█████████▋| 360725/371472 [6:32:42<47:03,  3.81it/s]
 97%|█████████▋| 360726/371472 [6:32:43<46:46,  3.83it/s]
 97%|█████████▋| 360727/371472 [6:32:43<47:31,  3.77it/s]
 97%|█████████▋| 360728/371472 [6:32:43<46:07,  3.88it/s]
 97%|█████████▋| 360729/371472 [6:32:43<47:56,  3.73it/s]
 97%|█████████▋| 360730/371472 [6:32:44<48:57,  3.66it/s]
 97%|█████████▋| 360731/371472 [6:32:44<51:28,  3.48it/s]
 97%|█████████▋| 360732/371472 [6:32:44<51:36,  3.47it/s]
 97%|█████████▋| 360733/371472 [6:32:45<49:41,  3.60it/s]
 97%|█████████▋| 360734/371472 [6:32:45<48:55,  3.66it/s]
 97%|█████████▋| 360735/371472 [6:32:45<47:07,  3.80it/s]
 97%|█████████▋| 360736/371472 [6:32:45<46:56,  3.81it/s]
 97%|█████████▋| 360737/371472 [6:32:46<51:08,  3.50it/s]
 97%|█████████▋| 360738/371472 [6:32:46<51:29,  3.47it/s]
 97%|█████████▋| 360739/371472 [6:32:46<51:21,  3.48it/s]
 97%|█████████▋| 360740/371472 [6:32:47<50:24,  3.55it/s]
                                                         
{'loss': 2.3781, 'learning_rate': 1.2601542804197464e-07, 'epoch': 15.54}
+
 97%|█████████▋| 360740/371472 [6:32:47<50:24,  3.55it/s]
 97%|█████████▋| 360741/371472 [6:32:47<50:12,  3.56it/s]
 97%|█████████▋| 360742/371472 [6:32:47<52:55,  3.38it/s]
 97%|█████████▋| 360743/371472 [6:32:47<52:26,  3.41it/s]
 97%|█████████▋| 360744/371472 [6:32:48<51:37,  3.46it/s]
 97%|█████████▋| 360745/371472 [6:32:48<51:07,  3.50it/s]
 97%|█████████▋| 360746/371472 [6:32:48<51:18,  3.48it/s]
 97%|█████████▋| 360747/371472 [6:32:49<57:36,  3.10it/s]
 97%|█████████▋| 360748/371472 [6:32:49<1:02:36,  2.85it/s]
 97%|█████████▋| 360749/371472 [6:32:49<58:32,  3.05it/s]  
 97%|█████████▋| 360750/371472 [6:32:50<54:38,  3.27it/s]
 97%|█████████▋| 360751/371472 [6:32:50<54:03,  3.30it/s]
 97%|█████████▋| 360752/371472 [6:32:50<51:05,  3.50it/s]
 97%|█████████▋| 360753/371472 [6:32:50<48:56,  3.65it/s]
 97%|█████████▋| 360754/371472 [6:32:51<47:54,  3.73it/s]
 97%|█████████▋| 360755/371472 [6:32:51<47:01,  3.80it/s]
 97%|█████████▋| 360756/371472 [6:32:51<46:26,  3.85it/s]
 97%|█████████▋| 360757/371472 [6:32:51<46:20,  3.85it/s]
 97%|█████████▋| 360758/371472 [6:32:52<48:13,  3.70it/s]
 97%|█████████▋| 360759/371472 [6:32:52<47:57,  3.72it/s]
 97%|█████████▋| 360760/371472 [6:32:52<48:02,  3.72it/s]
                                                         
{'loss': 2.4353, 'learning_rate': 1.2596694606649568e-07, 'epoch': 15.54}
+
 97%|█████████▋| 360760/371472 [6:32:52<48:02,  3.72it/s]
 97%|█████████▋| 360761/371472 [6:32:53<47:18,  3.77it/s]
 97%|█████████▋| 360762/371472 [6:32:53<49:18,  3.62it/s]
 97%|█████████▋| 360763/371472 [6:32:53<50:18,  3.55it/s]
 97%|█████████▋| 360764/371472 [6:32:53<49:50,  3.58it/s]
 97%|█████████▋| 360765/371472 [6:32:54<48:43,  3.66it/s]
 97%|█████████▋| 360766/371472 [6:32:54<48:36,  3.67it/s]
 97%|█████████▋| 360767/371472 [6:32:54<47:20,  3.77it/s]
 97%|█████████▋| 360768/371472 [6:32:55<49:29,  3.61it/s]
 97%|█████████▋| 360769/371472 [6:32:55<48:41,  3.66it/s]
 97%|█████████▋| 360770/371472 [6:32:55<47:47,  3.73it/s]
 97%|█████████▋| 360771/371472 [6:32:55<47:17,  3.77it/s]
 97%|█████████▋| 360772/371472 [6:32:56<51:04,  3.49it/s]
 97%|█████████▋| 360773/371472 [6:32:56<49:40,  3.59it/s]
 97%|█████████▋| 360774/371472 [6:32:56<58:54,  3.03it/s]
 97%|█████████▋| 360775/371472 [6:32:57<58:26,  3.05it/s]
 97%|█████████▋| 360776/371472 [6:32:57<55:20,  3.22it/s]
 97%|█████████▋| 360777/371472 [6:32:57<56:10,  3.17it/s]
 97%|█████████▋| 360778/371472 [6:32:58<54:40,  3.26it/s]
 97%|█████████▋| 360779/371472 [6:32:58<53:10,  3.35it/s]
 97%|█████████▋| 360780/371472 [6:32:58<52:44,  3.38it/s]
                                                         
{'loss': 2.5696, 'learning_rate': 1.2591846409101683e-07, 'epoch': 15.54}
+
 97%|█████████▋| 360780/371472 [6:32:58<52:44,  3.38it/s]
 97%|█████████▋| 360781/371472 [6:32:58<51:40,  3.45it/s]
 97%|█████████▋| 360782/371472 [6:32:59<50:27,  3.53it/s]
 97%|█████████▋| 360783/371472 [6:32:59<53:16,  3.34it/s]
 97%|█████████▋| 360784/371472 [6:32:59<52:37,  3.39it/s]
 97%|█████████▋| 360785/371472 [6:33:00<51:46,  3.44it/s]
 97%|█████████▋| 360786/371472 [6:33:00<50:00,  3.56it/s]
 97%|█████████▋| 360787/371472 [6:33:00<49:02,  3.63it/s]
 97%|█████████▋| 360788/371472 [6:33:00<51:53,  3.43it/s]
 97%|█████████▋| 360789/371472 [6:33:01<51:43,  3.44it/s]
 97%|█████████▋| 360790/371472 [6:33:01<53:20,  3.34it/s]
 97%|█████████▋| 360791/371472 [6:33:01<51:14,  3.47it/s]
 97%|█████████▋| 360792/371472 [6:33:02<52:44,  3.38it/s]
 97%|█████████▋| 360793/371472 [6:33:02<52:29,  3.39it/s]
 97%|█████████▋| 360794/371472 [6:33:02<52:57,  3.36it/s]
 97%|█████████▋| 360795/371472 [6:33:02<52:20,  3.40it/s]
 97%|█████████▋| 360796/371472 [6:33:03<52:12,  3.41it/s]
 97%|█████████▋| 360797/371472 [6:33:03<50:06,  3.55it/s]
 97%|█████████▋| 360798/371472 [6:33:03<50:12,  3.54it/s]
 97%|█████████▋| 360799/371472 [6:33:04<49:02,  3.63it/s]
 97%|█████████▋| 360800/371472 [6:33:04<48:56,  3.63it/s]
                                                         
{'loss': 2.4297, 'learning_rate': 1.258699821155379e-07, 'epoch': 15.54}
+
 97%|█████████▋| 360800/371472 [6:33:04<48:56,  3.63it/s]
 97%|█████████▋| 360801/371472 [6:33:04<47:11,  3.77it/s]
 97%|█████████▋| 360802/371472 [6:33:04<49:36,  3.58it/s]
 97%|█████████▋| 360803/371472 [6:33:05<52:39,  3.38it/s]
 97%|█████████▋| 360804/371472 [6:33:05<51:04,  3.48it/s]
 97%|█████████▋| 360805/371472 [6:33:05<49:22,  3.60it/s]
 97%|█████████▋| 360806/371472 [6:33:06<47:29,  3.74it/s]
 97%|█████████▋| 360807/371472 [6:33:06<55:36,  3.20it/s]
 97%|█████████▋| 360808/371472 [6:33:06<53:01,  3.35it/s]
 97%|█████████▋| 360809/371472 [6:33:06<50:45,  3.50it/s]
 97%|█████████▋| 360810/371472 [6:33:07<49:20,  3.60it/s]
 97%|█████████▋| 360811/371472 [6:33:07<48:28,  3.67it/s]
 97%|█████████▋| 360812/371472 [6:33:07<48:28,  3.66it/s]
 97%|█████████▋| 360813/371472 [6:33:08<48:08,  3.69it/s]
 97%|█████████▋| 360814/371472 [6:33:08<48:00,  3.70it/s]
 97%|█████████▋| 360815/371472 [6:33:08<51:22,  3.46it/s]
 97%|█████████▋| 360816/371472 [6:33:08<49:56,  3.56it/s]
 97%|█████████▋| 360817/371472 [6:33:09<48:13,  3.68it/s]
 97%|█████████▋| 360818/371472 [6:33:09<49:09,  3.61it/s]
 97%|█████████▋| 360819/371472 [6:33:09<48:49,  3.64it/s]
 97%|█████████▋| 360820/371472 [6:33:09<49:37,  3.58it/s]
                                                         
{'loss': 2.5256, 'learning_rate': 1.2582150014005905e-07, 'epoch': 15.54}
+
 97%|█████████▋| 360820/371472 [6:33:09<49:37,  3.58it/s]
 97%|█████████▋| 360821/371472 [6:33:10<49:30,  3.59it/s]
 97%|█████████▋| 360822/371472 [6:33:10<49:16,  3.60it/s]
 97%|█████████▋| 360823/371472 [6:33:10<47:44,  3.72it/s]
 97%|█████████▋| 360824/371472 [6:33:11<49:09,  3.61it/s]
 97%|█████████▋| 360825/371472 [6:33:11<49:25,  3.59it/s]
 97%|█████████▋| 360826/371472 [6:33:11<49:27,  3.59it/s]
 97%|█████████▋| 360827/371472 [6:33:11<49:12,  3.61it/s]
 97%|█████████▋| 360828/371472 [6:33:12<48:54,  3.63it/s]
 97%|█████████▋| 360829/371472 [6:33:12<48:22,  3.67it/s]
 97%|█████████▋| 360830/371472 [6:33:12<46:54,  3.78it/s]
 97%|█████████▋| 360831/371472 [6:33:13<51:26,  3.45it/s]
 97%|█████████▋| 360832/371472 [6:33:13<52:38,  3.37it/s]
 97%|█████████▋| 360833/371472 [6:33:13<50:13,  3.53it/s]
 97%|█████████▋| 360834/371472 [6:33:13<51:20,  3.45it/s]
 97%|█████████▋| 360835/371472 [6:33:14<50:59,  3.48it/s]
 97%|█████████▋| 360836/371472 [6:33:14<56:45,  3.12it/s]
 97%|█████████▋| 360837/371472 [6:33:14<57:27,  3.08it/s]
 97%|█████████▋| 360838/371472 [6:33:15<58:13,  3.04it/s]
 97%|█████████▋| 360839/371472 [6:33:15<53:30,  3.31it/s]
 97%|█████████▋| 360840/371472 [6:33:15<50:59,  3.47it/s]
                                                         
{'loss': 2.4734, 'learning_rate': 1.2577301816458013e-07, 'epoch': 15.54}
+
 97%|█████████▋| 360840/371472 [6:33:15<50:59,  3.47it/s]
 97%|█████████▋| 360841/371472 [6:33:16<51:54,  3.41it/s]
 97%|█████████▋| 360842/371472 [6:33:16<50:30,  3.51it/s]
 97%|█████████▋| 360843/371472 [6:33:16<49:54,  3.55it/s]
 97%|█████████▋| 360844/371472 [6:33:16<49:23,  3.59it/s]
 97%|█████████▋| 360845/371472 [6:33:17<52:37,  3.37it/s]
 97%|█████████▋| 360846/371472 [6:33:17<50:24,  3.51it/s]
 97%|█████████▋| 360847/371472 [6:33:17<49:43,  3.56it/s]
 97%|█████████▋| 360848/371472 [6:33:18<49:52,  3.55it/s]
 97%|█████████▋| 360849/371472 [6:33:18<49:10,  3.60it/s]
 97%|█████████▋| 360850/371472 [6:33:18<48:53,  3.62it/s]
 97%|█████████▋| 360851/371472 [6:33:18<52:12,  3.39it/s]
 97%|█████████▋| 360852/371472 [6:33:19<52:04,  3.40it/s]
 97%|█████████▋| 360853/371472 [6:33:19<51:20,  3.45it/s]
 97%|█████████▋| 360854/371472 [6:33:19<52:54,  3.34it/s]
 97%|█████████▋| 360855/371472 [6:33:20<51:50,  3.41it/s]
 97%|█████████▋| 360856/371472 [6:33:20<51:23,  3.44it/s]
 97%|█████████▋| 360857/371472 [6:33:20<52:21,  3.38it/s]
 97%|█████████▋| 360858/371472 [6:33:20<53:14,  3.32it/s]
 97%|█████████▋| 360859/371472 [6:33:21<51:10,  3.46it/s]
 97%|█████████▋| 360860/371472 [6:33:21<51:03,  3.46it/s]
                                                         
{'loss': 2.7959, 'learning_rate': 1.2572453618910128e-07, 'epoch': 15.54}
+
 97%|█████████▋| 360860/371472 [6:33:21<51:03,  3.46it/s]
 97%|█████████▋| 360861/371472 [6:33:21<53:06,  3.33it/s]
 97%|█████████▋| 360862/371472 [6:33:22<52:54,  3.34it/s]
 97%|█████████▋| 360863/371472 [6:33:22<51:50,  3.41it/s]
 97%|█████████▋| 360864/371472 [6:33:22<51:48,  3.41it/s]
 97%|█████████▋| 360865/371472 [6:33:22<48:48,  3.62it/s]
 97%|█████████▋| 360866/371472 [6:33:23<53:20,  3.31it/s]
 97%|█████████▋| 360867/371472 [6:33:23<51:04,  3.46it/s]
 97%|█████████▋| 360868/371472 [6:33:23<53:02,  3.33it/s]
 97%|█████████▋| 360869/371472 [6:33:24<54:31,  3.24it/s]
 97%|█████████▋| 360870/371472 [6:33:24<52:06,  3.39it/s]
 97%|█████████▋| 360871/371472 [6:33:24<50:03,  3.53it/s]
 97%|█████████▋| 360872/371472 [6:33:25<50:26,  3.50it/s]
 97%|█████████▋| 360873/371472 [6:33:25<50:25,  3.50it/s]
 97%|█████████▋| 360874/371472 [6:33:25<49:07,  3.60it/s]
 97%|█████████▋| 360875/371472 [6:33:25<53:26,  3.30it/s]
 97%|█████████▋| 360876/371472 [6:33:26<50:32,  3.49it/s]
 97%|█████████▋| 360877/371472 [6:33:26<49:38,  3.56it/s]
 97%|█████████▋| 360878/371472 [6:33:26<50:37,  3.49it/s]
 97%|█████████▋| 360879/371472 [6:33:27<50:03,  3.53it/s]
 97%|█████████▋| 360880/371472 [6:33:27<49:14,  3.58it/s]
                                                         
{'loss': 2.6099, 'learning_rate': 1.2567605421362232e-07, 'epoch': 15.54}
+
 97%|█████████▋| 360880/371472 [6:33:27<49:14,  3.58it/s]
 97%|█████████▋| 360881/371472 [6:33:27<47:50,  3.69it/s]
 97%|█████████▋| 360882/371472 [6:33:27<47:56,  3.68it/s]
 97%|█████████▋| 360883/371472 [6:33:28<48:16,  3.66it/s]
 97%|█████████▋| 360884/371472 [6:33:28<48:42,  3.62it/s]
 97%|█████████▋| 360885/371472 [6:33:28<49:36,  3.56it/s]
 97%|█████████▋| 360886/371472 [6:33:28<49:33,  3.56it/s]
 97%|█████████▋| 360887/371472 [6:33:29<49:52,  3.54it/s]
 97%|█████████▋| 360888/371472 [6:33:29<1:00:02,  2.94it/s]
 97%|█████████▋| 360889/371472 [6:33:30<56:39,  3.11it/s]  
 97%|█████████▋| 360890/371472 [6:33:30<56:41,  3.11it/s]
 97%|█████████▋| 360891/371472 [6:33:30<56:22,  3.13it/s]
 97%|█████████▋| 360892/371472 [6:33:30<52:56,  3.33it/s]
 97%|█████████▋| 360893/371472 [6:33:31<51:18,  3.44it/s]
 97%|█████████▋| 360894/371472 [6:33:31<53:56,  3.27it/s]
 97%|█████████▋| 360895/371472 [6:33:31<52:16,  3.37it/s]
 97%|█████████▋| 360896/371472 [6:33:32<49:16,  3.58it/s]
 97%|█████████▋| 360897/371472 [6:33:32<49:06,  3.59it/s]
 97%|█████████▋| 360898/371472 [6:33:32<50:47,  3.47it/s]
 97%|█████████▋| 360899/371472 [6:33:32<48:51,  3.61it/s]
 97%|█████████▋| 360900/371472 [6:33:33<49:45,  3.54it/s]
                                                         
{'loss': 2.4343, 'learning_rate': 1.256275722381435e-07, 'epoch': 15.54}
+
 97%|█████████▋| 360900/371472 [6:33:33<49:45,  3.54it/s]
 97%|█████████▋| 360901/371472 [6:33:33<48:42,  3.62it/s]
 97%|█████████▋| 360902/371472 [6:33:33<50:23,  3.50it/s]
 97%|█████████▋| 360903/371472 [6:33:34<52:31,  3.35it/s]
 97%|█████████▋| 360904/371472 [6:33:34<49:37,  3.55it/s]
 97%|█████████▋| 360905/371472 [6:33:34<48:07,  3.66it/s]
 97%|█████████▋| 360906/371472 [6:33:34<49:51,  3.53it/s]
 97%|█████████▋| 360907/371472 [6:33:35<48:07,  3.66it/s]
 97%|█████████▋| 360908/371472 [6:33:35<46:45,  3.77it/s]
 97%|█████████▋| 360909/371472 [6:33:35<46:28,  3.79it/s]
 97%|█████████▋| 360910/371472 [6:33:35<46:32,  3.78it/s]
 97%|█████████▋| 360911/371472 [6:33:36<46:58,  3.75it/s]
 97%|█████████▋| 360912/371472 [6:33:36<46:54,  3.75it/s]
 97%|█████████▋| 360913/371472 [6:33:36<52:36,  3.35it/s]
 97%|█████████▋| 360914/371472 [6:33:37<50:31,  3.48it/s]
 97%|█████████▋| 360915/371472 [6:33:37<48:57,  3.59it/s]
 97%|█████████▋| 360916/371472 [6:33:37<48:15,  3.65it/s]
 97%|█████████▋| 360917/371472 [6:33:37<47:59,  3.67it/s]
 97%|█████████▋| 360918/371472 [6:33:38<49:02,  3.59it/s]
 97%|█████████▋| 360919/371472 [6:33:38<48:17,  3.64it/s]
 97%|█████████▋| 360920/371472 [6:33:38<48:19,  3.64it/s]
                                                         
{'loss': 2.5113, 'learning_rate': 1.2557909026266454e-07, 'epoch': 15.55}
+
 97%|█████████▋| 360920/371472 [6:33:38<48:19,  3.64it/s]
 97%|█████████▋| 360921/371472 [6:33:39<57:56,  3.03it/s]
 97%|█████████▋| 360922/371472 [6:33:39<54:35,  3.22it/s]
 97%|█████████▋| 360923/371472 [6:33:39<51:13,  3.43it/s]
 97%|█████████▋| 360924/371472 [6:33:39<50:46,  3.46it/s]
 97%|█████████▋| 360925/371472 [6:33:40<48:16,  3.64it/s]
 97%|█████████▋| 360926/371472 [6:33:40<48:20,  3.64it/s]
 97%|█████████▋| 360927/371472 [6:33:40<47:12,  3.72it/s]
 97%|█████████▋| 360928/371472 [6:33:41<52:34,  3.34it/s]
 97%|█████████▋| 360929/371472 [6:33:41<53:00,  3.32it/s]
 97%|█████████▋| 360930/371472 [6:33:41<55:30,  3.17it/s]
 97%|█████████▋| 360931/371472 [6:33:42<53:39,  3.27it/s]
 97%|█████████▋| 360932/371472 [6:33:42<51:22,  3.42it/s]
 97%|█████████▋| 360933/371472 [6:33:42<49:57,  3.52it/s]
 97%|█████████▋| 360934/371472 [6:33:42<50:29,  3.48it/s]
 97%|█████████▋| 360935/371472 [6:33:43<49:38,  3.54it/s]
 97%|█████████▋| 360936/371472 [6:33:43<48:02,  3.66it/s]
 97%|█████████▋| 360937/371472 [6:33:43<46:24,  3.78it/s]
 97%|█████████▋| 360938/371472 [6:33:43<45:49,  3.83it/s]
 97%|█████████▋| 360939/371472 [6:33:44<45:58,  3.82it/s]
 97%|█████████▋| 360940/371472 [6:33:44<46:52,  3.75it/s]
                                                         
{'loss': 2.5506, 'learning_rate': 1.255306082871857e-07, 'epoch': 15.55}
+
 97%|█████████▋| 360940/371472 [6:33:44<46:52,  3.75it/s]
 97%|█████████▋| 360941/371472 [6:33:44<45:50,  3.83it/s]
 97%|█████████▋| 360942/371472 [6:33:44<44:43,  3.92it/s]
 97%|█████████▋| 360943/371472 [6:33:45<47:52,  3.67it/s]
 97%|█████████▋| 360944/371472 [6:33:45<47:55,  3.66it/s]
 97%|█████████▋| 360945/371472 [6:33:45<48:24,  3.62it/s]
 97%|█████████▋| 360946/371472 [6:33:46<48:29,  3.62it/s]
 97%|█████████▋| 360947/371472 [6:33:46<47:07,  3.72it/s]
 97%|█████████▋| 360948/371472 [6:33:46<50:20,  3.48it/s]
 97%|█████████▋| 360949/371472 [6:33:46<50:08,  3.50it/s]
 97%|█████████▋| 360950/371472 [6:33:47<50:33,  3.47it/s]
 97%|█████████▋| 360951/371472 [6:33:47<57:15,  3.06it/s]
 97%|█████████▋| 360952/371472 [6:33:47<54:45,  3.20it/s]
 97%|█████████▋| 360953/371472 [6:33:48<51:27,  3.41it/s]
 97%|█████████▋| 360954/371472 [6:33:48<51:00,  3.44it/s]
 97%|█████████▋| 360955/371472 [6:33:48<48:32,  3.61it/s]
 97%|█████████▋| 360956/371472 [6:33:48<47:38,  3.68it/s]
 97%|█████████▋| 360957/371472 [6:33:49<46:24,  3.78it/s]
 97%|█████████▋| 360958/371472 [6:33:49<48:54,  3.58it/s]
 97%|█████████▋| 360959/371472 [6:33:49<48:27,  3.62it/s]
 97%|█████████▋| 360960/371472 [6:33:50<49:45,  3.52it/s]
                                                         
{'loss': 2.5927, 'learning_rate': 1.2548212631170677e-07, 'epoch': 15.55}
+
 97%|█████████▋| 360960/371472 [6:33:50<49:45,  3.52it/s]
 97%|█████████▋| 360961/371472 [6:33:50<48:52,  3.58it/s]
 97%|█████████▋| 360962/371472 [6:33:50<51:11,  3.42it/s]
 97%|█████████▋| 360963/371472 [6:33:50<48:56,  3.58it/s]
 97%|█████████▋| 360964/371472 [6:33:51<48:44,  3.59it/s]
 97%|█████████▋| 360965/371472 [6:33:51<48:48,  3.59it/s]
 97%|█████████▋| 360966/371472 [6:33:51<47:27,  3.69it/s]
 97%|█████████▋| 360967/371472 [6:33:51<46:23,  3.77it/s]
 97%|█████████▋| 360968/371472 [6:33:52<46:51,  3.74it/s]
 97%|█████████▋| 360969/371472 [6:33:52<48:37,  3.60it/s]
 97%|█████████▋| 360970/371472 [6:33:52<46:50,  3.74it/s]
 97%|█████████▋| 360971/371472 [6:33:53<45:00,  3.89it/s]
 97%|█████████▋| 360972/371472 [6:33:53<47:54,  3.65it/s]
 97%|█████████▋| 360973/371472 [6:33:53<46:35,  3.76it/s]
 97%|█████████▋| 360974/371472 [6:33:53<46:00,  3.80it/s]
 97%|█████████▋| 360975/371472 [6:33:54<49:46,  3.51it/s]
 97%|█████████▋| 360976/371472 [6:33:54<48:13,  3.63it/s]
 97%|█████████▋| 360977/371472 [6:33:54<46:49,  3.74it/s]
 97%|█████████▋| 360978/371472 [6:33:54<47:20,  3.69it/s]
 97%|█████████▋| 360979/371472 [6:33:55<46:54,  3.73it/s]
 97%|█████████▋| 360980/371472 [6:33:55<46:58,  3.72it/s]
                                                         
{'loss': 2.724, 'learning_rate': 1.2543364433622792e-07, 'epoch': 15.55}
+
 97%|█████████▋| 360980/371472 [6:33:55<46:58,  3.72it/s]
 97%|█████████▋| 360981/371472 [6:33:55<48:19,  3.62it/s]
 97%|█████████▋| 360982/371472 [6:33:56<50:00,  3.50it/s]
 97%|█████████▋| 360983/371472 [6:33:56<49:55,  3.50it/s]
 97%|█████████▋| 360984/371472 [6:33:56<47:53,  3.65it/s]
 97%|█████████▋| 360985/371472 [6:33:56<46:13,  3.78it/s]
 97%|█████████▋| 360986/371472 [6:33:57<46:44,  3.74it/s]
 97%|█████████▋| 360987/371472 [6:33:57<46:36,  3.75it/s]
 97%|█████████▋| 360988/371472 [6:33:57<47:49,  3.65it/s]
 97%|█████████▋| 360989/371472 [6:33:57<48:09,  3.63it/s]
 97%|█████████▋| 360990/371472 [6:33:58<48:01,  3.64it/s]
 97%|█████████▋| 360991/371472 [6:33:58<47:26,  3.68it/s]
 97%|█████████▋| 360992/371472 [6:33:58<46:42,  3.74it/s]
 97%|█████████▋| 360993/371472 [6:33:59<47:50,  3.65it/s]
 97%|█████████▋| 360994/371472 [6:33:59<49:57,  3.50it/s]
 97%|█████████▋| 360995/371472 [6:33:59<47:50,  3.65it/s]
 97%|█████████▋| 360996/371472 [6:33:59<48:22,  3.61it/s]
 97%|█████████▋| 360997/371472 [6:34:00<50:32,  3.45it/s]
 97%|█████████▋| 360998/371472 [6:34:00<48:33,  3.60it/s]
 97%|█████████▋| 360999/371472 [6:34:00<48:46,  3.58it/s]
 97%|█████████▋| 361000/371472 [6:34:00<46:42,  3.74it/s]
                                                         
{'loss': 2.7193, 'learning_rate': 1.2538516236074896e-07, 'epoch': 15.55}
+
 97%|█████████▋| 361000/371472 [6:34:00<46:42,  3.74it/s]
 97%|█████████▋| 361001/371472 [6:34:01<48:51,  3.57it/s]
 97%|█████████▋| 361002/371472 [6:34:01<49:14,  3.54it/s]
 97%|█████████▋| 361003/371472 [6:34:01<48:58,  3.56it/s]
 97%|█████████▋| 361004/371472 [6:34:02<47:25,  3.68it/s]
 97%|█████████▋| 361005/371472 [6:34:02<46:37,  3.74it/s]
 97%|█████████▋| 361006/371472 [6:34:02<45:35,  3.83it/s]
 97%|█████████▋| 361007/371472 [6:34:02<45:14,  3.86it/s]
 97%|█████████▋| 361008/371472 [6:34:03<45:46,  3.81it/s]
 97%|█████████▋| 361009/371472 [6:34:03<45:43,  3.81it/s]
 97%|█████████▋| 361010/371472 [6:34:03<44:05,  3.95it/s]
 97%|█████████▋| 361011/371472 [6:34:03<45:10,  3.86it/s]
 97%|█████████▋| 361012/371472 [6:34:04<45:39,  3.82it/s]
 97%|█████████▋| 361013/371472 [6:34:04<45:02,  3.87it/s]
 97%|█████████▋| 361014/371472 [6:34:04<48:39,  3.58it/s]
 97%|█████████▋| 361015/371472 [6:34:05<49:29,  3.52it/s]
 97%|█████████▋| 361016/371472 [6:34:05<48:18,  3.61it/s]
 97%|█████████▋| 361017/371472 [6:34:05<46:43,  3.73it/s]
 97%|█████████▋| 361018/371472 [6:34:05<49:41,  3.51it/s]
 97%|█████████▋| 361019/371472 [6:34:06<49:17,  3.53it/s]
 97%|█████████▋| 361020/371472 [6:34:06<51:16,  3.40it/s]
                                                         
{'loss': 2.4747, 'learning_rate': 1.2533668038527014e-07, 'epoch': 15.55}
+
 97%|█████████▋| 361020/371472 [6:34:06<51:16,  3.40it/s]
 97%|█████████▋| 361021/371472 [6:34:06<51:05,  3.41it/s]
 97%|█████████▋| 361022/371472 [6:34:07<48:17,  3.61it/s]
 97%|█████████▋| 361023/371472 [6:34:07<47:45,  3.65it/s]
 97%|█████████▋| 361024/371472 [6:34:07<47:03,  3.70it/s]
 97%|█████████▋| 361025/371472 [6:34:07<46:52,  3.71it/s]
 97%|█████████▋| 361026/371472 [6:34:08<48:01,  3.63it/s]
 97%|█████████▋| 361027/371472 [6:34:08<50:04,  3.48it/s]
 97%|█████████▋| 361028/371472 [6:34:08<48:32,  3.59it/s]
 97%|█████████▋| 361029/371472 [6:34:08<47:00,  3.70it/s]
 97%|█████████▋| 361030/371472 [6:34:09<49:14,  3.53it/s]
 97%|█████████▋| 361031/371472 [6:34:09<49:31,  3.51it/s]
 97%|█████████▋| 361032/371472 [6:34:09<49:21,  3.52it/s]
 97%|█████████▋| 361033/371472 [6:34:10<55:41,  3.12it/s]
 97%|█████████▋| 361034/371472 [6:34:10<51:46,  3.36it/s]
 97%|█████████▋| 361035/371472 [6:34:10<50:44,  3.43it/s]
 97%|█████████▋| 361036/371472 [6:34:11<50:16,  3.46it/s]
 97%|█████████▋| 361037/371472 [6:34:11<50:03,  3.47it/s]
 97%|█████████▋| 361038/371472 [6:34:11<48:45,  3.57it/s]
 97%|█████████▋| 361039/371472 [6:34:11<47:08,  3.69it/s]
 97%|█████████▋| 361040/371472 [6:34:12<45:38,  3.81it/s]
                                                         
{'loss': 2.4665, 'learning_rate': 1.2528819840979119e-07, 'epoch': 15.55}
+
 97%|█████████▋| 361040/371472 [6:34:12<45:38,  3.81it/s]
 97%|█████████▋| 361041/371472 [6:34:12<46:09,  3.77it/s]
 97%|█████████▋| 361042/371472 [6:34:12<47:36,  3.65it/s]
 97%|█████████▋| 361043/371472 [6:34:12<47:03,  3.69it/s]
 97%|█████████▋| 361044/371472 [6:34:13<50:12,  3.46it/s]
 97%|█████████▋| 361045/371472 [6:34:13<50:10,  3.46it/s]
 97%|█████████▋| 361046/371472 [6:34:13<49:37,  3.50it/s]
 97%|█████████▋| 361047/371472 [6:34:14<49:45,  3.49it/s]
 97%|█████████▋| 361048/371472 [6:34:14<51:45,  3.36it/s]
 97%|█████████▋| 361049/371472 [6:34:14<48:43,  3.57it/s]
 97%|█████████▋| 361050/371472 [6:34:14<48:55,  3.55it/s]
 97%|█████████▋| 361051/371472 [6:34:15<46:55,  3.70it/s]
 97%|█████████▋| 361052/371472 [6:34:15<47:54,  3.62it/s]
 97%|█████████▋| 361053/371472 [6:34:15<46:57,  3.70it/s]
 97%|█████████▋| 361054/371472 [6:34:15<47:16,  3.67it/s]
 97%|█████████▋| 361055/371472 [6:34:16<46:46,  3.71it/s]
 97%|█████████▋| 361056/371472 [6:34:16<47:51,  3.63it/s]
 97%|█████████▋| 361057/371472 [6:34:16<48:49,  3.56it/s]
 97%|█████████▋| 361058/371472 [6:34:17<48:54,  3.55it/s]
 97%|█████████▋| 361059/371472 [6:34:17<48:50,  3.55it/s]
 97%|█████████▋| 361060/371472 [6:34:17<48:25,  3.58it/s]
                                                         
{'loss': 2.6839, 'learning_rate': 1.2523971643431234e-07, 'epoch': 15.55}
+
 97%|█████████▋| 361060/371472 [6:34:17<48:25,  3.58it/s]
 97%|█████████▋| 361061/371472 [6:34:17<46:22,  3.74it/s]
 97%|█████████▋| 361062/371472 [6:34:18<46:54,  3.70it/s]
 97%|█████████▋| 361063/371472 [6:34:18<46:49,  3.71it/s]
 97%|█████████▋| 361064/371472 [6:34:18<47:02,  3.69it/s]
 97%|█████████▋| 361065/371472 [6:34:19<48:22,  3.59it/s]
 97%|█████████▋| 361066/371472 [6:34:19<53:45,  3.23it/s]
 97%|█████████▋| 361067/371472 [6:34:19<51:52,  3.34it/s]
 97%|█████████▋| 361068/371472 [6:34:19<49:27,  3.51it/s]
 97%|█████████▋| 361069/371472 [6:34:20<59:29,  2.91it/s]
 97%|█████████▋| 361070/371472 [6:34:20<55:32,  3.12it/s]
 97%|█████████▋| 361071/371472 [6:34:21<55:38,  3.12it/s]
 97%|█████████▋| 361072/371472 [6:34:21<54:46,  3.16it/s]
 97%|█████████▋| 361073/371472 [6:34:21<52:03,  3.33it/s]
 97%|█████████▋| 361074/371472 [6:34:21<49:58,  3.47it/s]
 97%|█████████▋| 361075/371472 [6:34:22<48:08,  3.60it/s]
 97%|█████████▋| 361076/371472 [6:34:22<47:09,  3.67it/s]
 97%|█████████▋| 361077/371472 [6:34:22<47:36,  3.64it/s]
 97%|█████████▋| 361078/371472 [6:34:22<48:57,  3.54it/s]
 97%|█████████▋| 361079/371472 [6:34:23<47:30,  3.65it/s]
 97%|█████████▋| 361080/371472 [6:34:23<49:56,  3.47it/s]
                                                         
{'loss': 2.6732, 'learning_rate': 1.251912344588334e-07, 'epoch': 15.55}
+
 97%|█████████▋| 361080/371472 [6:34:23<49:56,  3.47it/s]
 97%|█████████▋| 361081/371472 [6:34:23<49:38,  3.49it/s]
 97%|█████████▋| 361082/371472 [6:34:24<47:13,  3.67it/s]
 97%|█████████▋| 361083/371472 [6:34:24<47:46,  3.62it/s]
 97%|█████████▋| 361084/371472 [6:34:24<49:51,  3.47it/s]
 97%|█████████▋| 361085/371472 [6:34:24<51:05,  3.39it/s]
 97%|█████████▋| 361086/371472 [6:34:25<49:48,  3.48it/s]
 97%|█████████▋| 361087/371472 [6:34:25<49:25,  3.50it/s]
 97%|█████████▋| 361088/371472 [6:34:25<48:48,  3.55it/s]
 97%|█████████▋| 361089/371472 [6:34:26<47:06,  3.67it/s]
 97%|█████████▋| 361090/371472 [6:34:26<48:58,  3.53it/s]
 97%|█████████▋| 361091/371472 [6:34:26<48:50,  3.54it/s]
 97%|█████████▋| 361092/371472 [6:34:26<48:44,  3.55it/s]
 97%|█████████▋| 361093/371472 [6:34:27<49:24,  3.50it/s]
 97%|█████████▋| 361094/371472 [6:34:27<49:51,  3.47it/s]
 97%|█████████▋| 361095/371472 [6:34:27<53:08,  3.25it/s]
 97%|█████████▋| 361096/371472 [6:34:28<50:50,  3.40it/s]
 97%|█████████▋| 361097/371472 [6:34:28<48:25,  3.57it/s]
 97%|█████████▋| 361098/371472 [6:34:28<47:41,  3.63it/s]
 97%|█████████▋| 361099/371472 [6:34:28<46:52,  3.69it/s]
 97%|█████████▋| 361100/371472 [6:34:29<47:42,  3.62it/s]
                                                         
{'loss': 2.6763, 'learning_rate': 1.2514275248335456e-07, 'epoch': 15.55}
+
 97%|█████████▋| 361100/371472 [6:34:29<47:42,  3.62it/s]
 97%|█████████▋| 361101/371472 [6:34:29<53:06,  3.25it/s]
 97%|█████████▋| 361102/371472 [6:34:29<53:22,  3.24it/s]
 97%|█████████▋| 361103/371472 [6:34:30<53:20,  3.24it/s]
 97%|█████████▋| 361104/371472 [6:34:30<55:59,  3.09it/s]
 97%|█████████▋| 361105/371472 [6:34:30<54:19,  3.18it/s]
 97%|█████████▋| 361106/371472 [6:34:31<54:30,  3.17it/s]
 97%|█████████▋| 361107/371472 [6:34:31<54:42,  3.16it/s]
 97%|█████████▋| 361108/371472 [6:34:31<50:47,  3.40it/s]
 97%|█████████▋| 361109/371472 [6:34:31<48:13,  3.58it/s]
 97%|█████████▋| 361110/371472 [6:34:32<47:32,  3.63it/s]
 97%|█████████▋| 361111/371472 [6:34:32<48:19,  3.57it/s]
 97%|█████████▋| 361112/371472 [6:34:32<54:36,  3.16it/s]
 97%|█████████▋| 361113/371472 [6:34:33<55:22,  3.12it/s]
 97%|█████████▋| 361114/371472 [6:34:33<55:26,  3.11it/s]
 97%|█████████▋| 361115/371472 [6:34:33<53:14,  3.24it/s]
 97%|█████████▋| 361116/371472 [6:34:34<51:40,  3.34it/s]
 97%|█████████▋| 361117/371472 [6:34:34<49:22,  3.49it/s]
 97%|██��██████▋| 361118/371472 [6:34:34<47:23,  3.64it/s]
 97%|█████████▋| 361119/371472 [6:34:34<45:42,  3.78it/s]
 97%|█████████▋| 361120/371472 [6:34:35<45:22,  3.80it/s]
                                                         
{'loss': 2.716, 'learning_rate': 1.2509427050787563e-07, 'epoch': 15.55}
+
 97%|█████████▋| 361120/371472 [6:34:35<45:22,  3.80it/s]
 97%|█████████▋| 361121/371472 [6:34:35<44:55,  3.84it/s]
 97%|█████████▋| 361122/371472 [6:34:35<45:38,  3.78it/s]
 97%|█████████▋| 361123/371472 [6:34:35<46:29,  3.71it/s]
 97%|█████████▋| 361124/371472 [6:34:36<50:30,  3.41it/s]
 97%|█████████▋| 361125/371472 [6:34:36<56:07,  3.07it/s]
 97%|█████████▋| 361126/371472 [6:34:36<53:20,  3.23it/s]
 97%|█████████▋| 361127/371472 [6:34:37<49:30,  3.48it/s]
 97%|█████████▋| 361128/371472 [6:34:37<49:02,  3.52it/s]
 97%|█████████▋| 361129/371472 [6:34:37<50:37,  3.40it/s]
 97%|█████████▋| 361130/371472 [6:34:38<49:54,  3.45it/s]
 97%|█████████▋| 361131/371472 [6:34:38<49:15,  3.50it/s]
 97%|█████████▋| 361132/371472 [6:34:38<51:24,  3.35it/s]
 97%|█████████▋| 361133/371472 [6:34:38<51:42,  3.33it/s]
 97%|█████████▋| 361134/371472 [6:34:39<52:43,  3.27it/s]
 97%|█████████▋| 361135/371472 [6:34:39<53:30,  3.22it/s]
 97%|█████████▋| 361136/371472 [6:34:39<50:43,  3.40it/s]
 97%|█████████▋| 361137/371472 [6:34:40<50:26,  3.41it/s]
 97%|█████████▋| 361138/371472 [6:34:40<48:59,  3.52it/s]
 97%|█████████▋| 361139/371472 [6:34:40<47:46,  3.60it/s]
 97%|█████████▋| 361140/371472 [6:34:40<47:21,  3.64it/s]
                                                         
{'loss': 2.4667, 'learning_rate': 1.2504578853239667e-07, 'epoch': 15.55}
+
 97%|█████████▋| 361140/371472 [6:34:40<47:21,  3.64it/s]
 97%|█████████▋| 361141/371472 [6:34:41<49:32,  3.48it/s]
 97%|█████████▋| 361142/371472 [6:34:41<47:44,  3.61it/s]
 97%|█████████▋| 361143/371472 [6:34:41<46:25,  3.71it/s]
 97%|█████████▋| 361144/371472 [6:34:42<48:22,  3.56it/s]
 97%|█████████▋| 361145/371472 [6:34:42<46:29,  3.70it/s]
 97%|█████████▋| 361146/371472 [6:34:42<45:10,  3.81it/s]
 97%|█████████▋| 361147/371472 [6:34:42<49:39,  3.47it/s]
 97%|█████████▋| 361148/371472 [6:34:43<48:07,  3.58it/s]
 97%|█████████▋| 361149/371472 [6:34:43<48:07,  3.57it/s]
 97%|█████████▋| 361150/371472 [6:34:43<49:45,  3.46it/s]
 97%|█████████▋| 361151/371472 [6:34:44<48:29,  3.55it/s]
 97%|█████████▋| 361152/371472 [6:34:44<46:21,  3.71it/s]
 97%|█████████▋| 361153/371472 [6:34:44<47:00,  3.66it/s]
 97%|█████████▋| 361154/371472 [6:34:44<46:48,  3.67it/s]
 97%|█████████▋| 361155/371472 [6:34:45<45:58,  3.74it/s]
 97%|█████████▋| 361156/371472 [6:34:45<48:14,  3.56it/s]
 97%|█████████▋| 361157/371472 [6:34:45<46:31,  3.70it/s]
 97%|█████████▋| 361158/371472 [6:34:45<45:11,  3.80it/s]
 97%|█████████▋| 361159/371472 [6:34:46<44:22,  3.87it/s]
 97%|█████████▋| 361160/371472 [6:34:46<44:44,  3.84it/s]
                                                         
{'loss': 2.6191, 'learning_rate': 1.2499730655691783e-07, 'epoch': 15.56}
+
 97%|█████████▋| 361160/371472 [6:34:46<44:44,  3.84it/s]
 97%|█████████▋| 361161/371472 [6:34:46<48:55,  3.51it/s]
 97%|█████████▋| 361162/371472 [6:34:46<47:47,  3.60it/s]
 97%|█████████▋| 361163/371472 [6:34:47<46:04,  3.73it/s]
 97%|█████████▋| 361164/371472 [6:34:47<48:39,  3.53it/s]
 97%|█████████▋| 361165/371472 [6:34:47<48:32,  3.54it/s]
 97%|█████████▋| 361166/371472 [6:34:48<50:37,  3.39it/s]
 97%|█████████▋| 361167/371472 [6:34:48<52:45,  3.26it/s]
 97%|█████████▋| 361168/371472 [6:34:48<49:46,  3.45it/s]
 97%|█████████▋| 361169/371472 [6:34:49<50:25,  3.41it/s]
 97%|█████████▋| 361170/371472 [6:34:49<48:38,  3.53it/s]
 97%|█████████▋| 361171/371472 [6:34:49<49:59,  3.43it/s]
 97%|█████████▋| 361172/371472 [6:34:49<49:28,  3.47it/s]
 97%|█████████▋| 361173/371472 [6:34:50<47:21,  3.63it/s]
 97%|█████████▋| 361174/371472 [6:34:50<48:52,  3.51it/s]
 97%|█████████▋| 361175/371472 [6:34:50<48:49,  3.52it/s]
 97%|█████████▋| 361176/371472 [6:34:51<48:43,  3.52it/s]
 97%|█████████▋| 361177/371472 [6:34:51<46:26,  3.69it/s]
 97%|█████████▋| 361178/371472 [6:34:51<47:22,  3.62it/s]
 97%|█████████▋| 361179/371472 [6:34:51<46:57,  3.65it/s]
 97%|█████████▋| 361180/371472 [6:34:52<46:11,  3.71it/s]
                                                         
{'loss': 2.4963, 'learning_rate': 1.249488245814389e-07, 'epoch': 15.56}
+
 97%|█████████▋| 361180/371472 [6:34:52<46:11,  3.71it/s]
 97%|█████████▋| 361181/371472 [6:34:52<46:22,  3.70it/s]
 97%|█████████▋| 361182/371472 [6:34:52<45:43,  3.75it/s]
 97%|█████████▋| 361183/371472 [6:34:52<49:00,  3.50it/s]
 97%|█████████▋| 361184/371472 [6:34:53<47:04,  3.64it/s]
 97%|█████████▋| 361185/371472 [6:34:53<48:28,  3.54it/s]
 97%|█████████▋| 361186/371472 [6:34:53<49:22,  3.47it/s]
 97%|█████████▋| 361187/371472 [6:34:54<49:33,  3.46it/s]
 97%|█████████▋| 361188/371472 [6:34:54<49:33,  3.46it/s]
 97%|█████████▋| 361189/371472 [6:34:54<47:42,  3.59it/s]
 97%|█████████▋| 361190/371472 [6:34:54<46:30,  3.69it/s]
 97%|█████████▋| 361191/371472 [6:34:55<44:58,  3.81it/s]
 97%|█████████▋| 361192/371472 [6:34:55<45:02,  3.80it/s]
 97%|█████████▋| 361193/371472 [6:34:55<44:52,  3.82it/s]
 97%|█████████▋| 361194/371472 [6:34:55<44:28,  3.85it/s]
 97%|█████████▋| 361195/371472 [6:34:56<42:54,  3.99it/s]
 97%|█████████▋| 361196/371472 [6:34:56<42:35,  4.02it/s]
 97%|█████████▋| 361197/371472 [6:34:56<44:15,  3.87it/s]
 97%|█████████▋| 361198/371472 [6:34:56<44:18,  3.86it/s]
 97%|█████████▋| 361199/371472 [6:34:57<43:30,  3.94it/s]
 97%|█████████▋| 361200/371472 [6:34:57<43:46,  3.91it/s]
                                                         
{'loss': 2.4325, 'learning_rate': 1.2490034260596005e-07, 'epoch': 15.56}
+
 97%|█████████▋| 361200/371472 [6:34:57<43:46,  3.91it/s]
 97%|█████████▋| 361201/371472 [6:34:57<46:35,  3.67it/s]
 97%|█████████▋| 361202/371472 [6:34:57<47:18,  3.62it/s]
 97%|█████████▋| 361203/371472 [6:34:58<48:42,  3.51it/s]
 97%|█████████▋| 361204/371472 [6:34:58<48:26,  3.53it/s]
 97%|█████████▋| 361205/371472 [6:34:58<47:23,  3.61it/s]
 97%|█████████▋| 361206/371472 [6:34:59<44:59,  3.80it/s]
 97%|█████████▋| 361207/371472 [6:34:59<44:13,  3.87it/s]
 97%|█████████▋| 361208/371472 [6:34:59<44:14,  3.87it/s]
 97%|█████████▋| 361209/371472 [6:34:59<46:01,  3.72it/s]
 97%|█████████▋| 361210/371472 [6:35:00<45:17,  3.78it/s]
 97%|█████████▋| 361211/371472 [6:35:00<44:02,  3.88it/s]
 97%|█████████▋| 361212/371472 [6:35:00<44:12,  3.87it/s]
 97%|█████████▋| 361213/371472 [6:35:00<46:05,  3.71it/s]
 97%|█████████▋| 361214/371472 [6:35:01<46:15,  3.70it/s]
 97%|█████████▋| 361215/371472 [6:35:01<46:22,  3.69it/s]
 97%|█████████▋| 361216/371472 [6:35:01<47:30,  3.60it/s]
 97%|█████████▋| 361217/371472 [6:35:02<48:17,  3.54it/s]
 97%|█████████▋| 361218/371472 [6:35:02<46:47,  3.65it/s]
 97%|█████████▋| 361219/371472 [6:35:02<46:00,  3.71it/s]
 97%|█████████▋| 361220/371472 [6:35:02<46:57,  3.64it/s]
                                                         
{'loss': 2.6134, 'learning_rate': 1.2485186063048112e-07, 'epoch': 15.56}
+
 97%|█████████▋| 361220/371472 [6:35:02<46:57,  3.64it/s]
 97%|█████████▋| 361221/371472 [6:35:03<46:06,  3.71it/s]
 97%|█████████▋| 361222/371472 [6:35:03<44:55,  3.80it/s]
 97%|█████████▋| 361223/371472 [6:35:03<48:12,  3.54it/s]
 97%|█████████▋| 361224/371472 [6:35:04<52:09,  3.27it/s]
 97%|█████████▋| 361225/371472 [6:35:04<54:50,  3.11it/s]
 97%|█████████▋| 361226/371472 [6:35:04<51:06,  3.34it/s]
 97%|█████████▋| 361227/371472 [6:35:04<49:29,  3.45it/s]
 97%|█████████▋| 361228/371472 [6:35:05<48:00,  3.56it/s]
 97%|█████████▋| 361229/371472 [6:35:05<45:49,  3.73it/s]
 97%|█████████▋| 361230/371472 [6:35:05<52:02,  3.28it/s]
 97%|█████████▋| 361231/371472 [6:35:06<51:20,  3.32it/s]
 97%|█████████▋| 361232/371472 [6:35:06<52:22,  3.26it/s]
 97%|██��██████▋| 361233/371472 [6:35:06<56:06,  3.04it/s]
 97%|█████████▋| 361234/371472 [6:35:07<51:44,  3.30it/s]
 97%|█████████▋| 361235/371472 [6:35:07<54:37,  3.12it/s]
 97%|█████████▋| 361236/371472 [6:35:07<52:39,  3.24it/s]
 97%|█████████▋| 361237/371472 [6:35:07<49:28,  3.45it/s]
 97%|█████████▋| 361238/371472 [6:35:08<49:24,  3.45it/s]
 97%|█████████▋| 361239/371472 [6:35:08<48:58,  3.48it/s]
 97%|█████████▋| 361240/371472 [6:35:08<48:45,  3.50it/s]
                                                         
{'loss': 2.5301, 'learning_rate': 1.2480337865500227e-07, 'epoch': 15.56}
+
 97%|█████████▋| 361240/371472 [6:35:08<48:45,  3.50it/s]
 97%|█████████▋| 361241/371472 [6:35:09<47:28,  3.59it/s]
 97%|█████████▋| 361242/371472 [6:35:09<46:05,  3.70it/s]
 97%|█████████▋| 361243/371472 [6:35:09<46:24,  3.67it/s]
 97%|█████████▋| 361244/371472 [6:35:09<45:24,  3.75it/s]
 97%|█████████▋| 361245/371472 [6:35:10<45:48,  3.72it/s]
 97%|█████████▋| 361246/371472 [6:35:10<44:40,  3.82it/s]
 97%|█████████▋| 361247/371472 [6:35:10<43:05,  3.96it/s]
 97%|█████████▋| 361248/371472 [6:35:10<44:40,  3.81it/s]
 97%|█████████▋| 361249/371472 [6:35:11<48:14,  3.53it/s]
 97%|█████████▋| 361250/371472 [6:35:11<46:59,  3.63it/s]
 97%|█████████▋| 361251/371472 [6:35:11<45:28,  3.75it/s]
 97%|█████████▋| 361252/371472 [6:35:11<46:11,  3.69it/s]
 97%|█████████▋| 361253/371472 [6:35:12<44:31,  3.83it/s]
 97%|█████████▋| 361254/371472 [6:35:12<46:21,  3.67it/s]
 97%|█████████▋| 361255/371472 [6:35:12<46:35,  3.65it/s]
 97%|█████████▋| 361256/371472 [6:35:13<45:50,  3.71it/s]
 97%|█████████▋| 361257/371472 [6:35:13<44:57,  3.79it/s]
 97%|█████████▋| 361258/371472 [6:35:13<45:37,  3.73it/s]
 97%|█████████▋| 361259/371472 [6:35:13<46:53,  3.63it/s]
 97%|█████████▋| 361260/371472 [6:35:14<46:57,  3.62it/s]
                                                         
{'loss': 2.5886, 'learning_rate': 1.2475489667952332e-07, 'epoch': 15.56}
+
 97%|█████████▋| 361260/371472 [6:35:14<46:57,  3.62it/s]
 97%|█████████▋| 361261/371472 [6:35:14<48:23,  3.52it/s]
 97%|█████████▋| 361262/371472 [6:35:14<47:00,  3.62it/s]
 97%|█████████▋| 361263/371472 [6:35:14<45:13,  3.76it/s]
 97%|█████████▋| 361264/371472 [6:35:15<45:47,  3.72it/s]
 97%|█████████▋| 361265/371472 [6:35:15<45:40,  3.72it/s]
 97%|█████████▋| 361266/371472 [6:35:15<44:47,  3.80it/s]
 97%|█████████▋| 361267/371472 [6:35:15<43:37,  3.90it/s]
 97%|█████████▋| 361268/371472 [6:35:16<45:14,  3.76it/s]
 97%|█████████▋| 361269/371472 [6:35:16<46:03,  3.69it/s]
 97%|█████████▋| 361270/371472 [6:35:16<46:53,  3.63it/s]
 97%|█████████▋| 361271/371472 [6:35:17<47:44,  3.56it/s]
 97%|█████████▋| 361272/371472 [6:35:17<47:18,  3.59it/s]
 97%|█████████▋| 361273/371472 [6:35:17<49:51,  3.41it/s]
 97%|█████████▋| 361274/371472 [6:35:17<47:18,  3.59it/s]
 97%|█████████▋| 361275/371472 [6:35:18<45:37,  3.72it/s]
 97%|█████████▋| 361276/371472 [6:35:18<46:46,  3.63it/s]
 97%|█████████▋| 361277/371472 [6:35:18<50:42,  3.35it/s]
 97%|█████████▋| 361278/371472 [6:35:19<52:30,  3.24it/s]
 97%|█████████▋| 361279/371472 [6:35:19<50:04,  3.39it/s]
 97%|█████████▋| 361280/371472 [6:35:19<57:35,  2.95it/s]
                                                         
{'loss': 2.6661, 'learning_rate': 1.247064147040445e-07, 'epoch': 15.56}
+
 97%|█████████▋| 361280/371472 [6:35:19<57:35,  2.95it/s]
 97%|█████████▋| 361281/371472 [6:35:20<54:39,  3.11it/s]
 97%|█████████▋| 361282/371472 [6:35:20<50:48,  3.34it/s]
 97%|█████████▋| 361283/371472 [6:35:20<47:59,  3.54it/s]
 97%|█████████▋| 361284/371472 [6:35:20<46:46,  3.63it/s]
 97%|█████████▋| 361285/371472 [6:35:21<46:57,  3.62it/s]
 97%|█████████▋| 361286/371472 [6:35:21<47:38,  3.56it/s]
 97%|█████████▋| 361287/371472 [6:35:21<47:45,  3.55it/s]
 97%|█████████▋| 361288/371472 [6:35:22<47:04,  3.61it/s]
 97%|█████████▋| 361289/371472 [6:35:22<48:37,  3.49it/s]
 97%|█████████▋| 361290/371472 [6:35:22<51:12,  3.31it/s]
 97%|█████████▋| 361291/371472 [6:35:22<50:04,  3.39it/s]
 97%|█████████▋| 361292/371472 [6:35:23<47:56,  3.54it/s]
 97%|█████████▋| 361293/371472 [6:35:23<48:10,  3.52it/s]
 97%|█████████▋| 361294/371472 [6:35:23<45:29,  3.73it/s]
 97%|█████████▋| 361295/371472 [6:35:24<45:01,  3.77it/s]
 97%|█████████▋| 361296/371472 [6:35:24<45:47,  3.70it/s]
 97%|█████████▋| 361297/371472 [6:35:24<48:01,  3.53it/s]
 97%|█████████▋| 361298/371472 [6:35:24<48:48,  3.47it/s]
 97%|█████████▋| 361299/371472 [6:35:25<50:36,  3.35it/s]
 97%|█████████▋| 361300/371472 [6:35:25<48:28,  3.50it/s]
                                                         
{'loss': 2.4483, 'learning_rate': 1.2465793272856554e-07, 'epoch': 15.56}
+
 97%|█████████▋| 361300/371472 [6:35:25<48:28,  3.50it/s]
 97%|█████████▋| 361301/371472 [6:35:25<47:54,  3.54it/s]
 97%|█████████▋| 361302/371472 [6:35:26<48:05,  3.53it/s]
 97%|█████████▋| 361303/371472 [6:35:26<46:13,  3.67it/s]
 97%|█████████▋| 361304/371472 [6:35:26<47:45,  3.55it/s]
 97%|█████████▋| 361305/371472 [6:35:26<47:44,  3.55it/s]
 97%|█████████▋| 361306/371472 [6:35:27<46:02,  3.68it/s]
 97%|█████████▋| 361307/371472 [6:35:27<47:45,  3.55it/s]
 97%|█████████▋| 361308/371472 [6:35:27<48:39,  3.48it/s]
 97%|█████████▋| 361309/371472 [6:35:27<46:41,  3.63it/s]
 97%|█████████▋| 361310/371472 [6:35:28<47:55,  3.53it/s]
 97%|█████████▋| 361311/371472 [6:35:28<45:33,  3.72it/s]
 97%|█████████▋| 361312/371472 [6:35:28<44:50,  3.78it/s]
 97%|█████████▋| 361313/371472 [6:35:29<44:36,  3.80it/s]
 97%|█████████▋| 361314/371472 [6:35:29<47:01,  3.60it/s]
 97%|█████████▋| 361315/371472 [6:35:29<48:11,  3.51it/s]
 97%|█████████▋| 361316/371472 [6:35:29<46:27,  3.64it/s]
 97%|█████████▋| 361317/371472 [6:35:30<45:25,  3.73it/s]
 97%|█████████▋| 361318/371472 [6:35:30<44:55,  3.77it/s]
 97%|█████████▋| 361319/371472 [6:35:30<44:44,  3.78it/s]
 97%|█████████▋| 361320/371472 [6:35:30<47:14,  3.58it/s]
                                                         
{'loss': 2.6183, 'learning_rate': 1.246094507530867e-07, 'epoch': 15.56}
+
 97%|█████████▋| 361320/371472 [6:35:30<47:14,  3.58it/s]
 97%|█████████▋| 361321/371472 [6:35:31<49:48,  3.40it/s]
 97%|█████████▋| 361322/371472 [6:35:31<54:06,  3.13it/s]
 97%|█████████▋| 361323/371472 [6:35:31<50:34,  3.34it/s]
 97%|█████████▋| 361324/371472 [6:35:32<52:44,  3.21it/s]
 97%|█████████▋| 361325/371472 [6:35:32<51:58,  3.25it/s]
 97%|█████████▋| 361326/371472 [6:35:32<53:35,  3.16it/s]
 97%|█████████▋| 361327/371472 [6:35:33<56:43,  2.98it/s]
 97%|█████████▋| 361328/371472 [6:35:33<55:11,  3.06it/s]
 97%|█████████▋| 361329/371472 [6:35:33<53:13,  3.18it/s]
 97%|█████████▋| 361330/371472 [6:35:34<50:23,  3.35it/s]
 97%|█████████▋| 361331/371472 [6:35:34<51:32,  3.28it/s]
 97%|█████████▋| 361332/371472 [6:35:34<50:08,  3.37it/s]
 97%|█████████▋| 361333/371472 [6:35:35<51:25,  3.29it/s]
 97%|█████████▋| 361334/371472 [6:35:35<51:45,  3.26it/s]
 97%|█████████▋| 361335/371472 [6:35:35<53:28,  3.16it/s]
 97%|█████████▋| 361336/371472 [6:35:36<56:32,  2.99it/s]
 97%|█████████▋| 361337/371472 [6:35:36<52:47,  3.20it/s]
 97%|█████████▋| 361338/371472 [6:35:36<53:34,  3.15it/s]
 97%|█████████▋| 361339/371472 [6:35:36<51:42,  3.27it/s]
 97%|█████████▋| 361340/371472 [6:35:37<51:21,  3.29it/s]
                                                         
{'loss': 2.4052, 'learning_rate': 1.2456096877760776e-07, 'epoch': 15.56}
+
 97%|█████████▋| 361340/371472 [6:35:37<51:21,  3.29it/s]
 97%|█████████▋| 361341/371472 [6:35:37<50:27,  3.35it/s]
 97%|█████████▋| 361342/371472 [6:35:37<48:38,  3.47it/s]
 97%|█████████▋| 361343/371472 [6:35:38<47:17,  3.57it/s]
 97%|█████████▋| 361344/371472 [6:35:38<46:39,  3.62it/s]
 97%|█████████▋| 361345/371472 [6:35:38<45:42,  3.69it/s]
 97%|█████████▋| 361346/371472 [6:35:38<45:42,  3.69it/s]
 97%|█████████▋| 361347/371472 [6:35:39<46:15,  3.65it/s]
 97%|██��██████▋| 361348/371472 [6:35:39<47:06,  3.58it/s]
 97%|█████████▋| 361349/371472 [6:35:39<48:27,  3.48it/s]
 97%|█████████▋| 361350/371472 [6:35:40<52:57,  3.19it/s]
 97%|█████████▋| 361351/371472 [6:35:40<49:58,  3.38it/s]
 97%|█████████▋| 361352/371472 [6:35:40<50:49,  3.32it/s]
 97%|█████████▋| 361353/371472 [6:35:40<49:07,  3.43it/s]
 97%|█████████▋| 361354/371472 [6:35:41<47:27,  3.55it/s]
 97%|█████████▋| 361355/371472 [6:35:41<48:03,  3.51it/s]
 97%|█████████▋| 361356/371472 [6:35:41<46:41,  3.61it/s]
 97%|█████████▋| 361357/371472 [6:35:42<46:10,  3.65it/s]
 97%|█████████▋| 361358/371472 [6:35:42<50:06,  3.36it/s]
 97%|█████████▋| 361359/371472 [6:35:42<50:29,  3.34it/s]
 97%|█████████▋| 361360/371472 [6:35:43<50:15,  3.35it/s]
                                                         
{'loss': 2.5579, 'learning_rate': 1.245124868021289e-07, 'epoch': 15.56}
+
 97%|█████████▋| 361360/371472 [6:35:43<50:15,  3.35it/s]
 97%|█████████▋| 361361/371472 [6:35:43<50:27,  3.34it/s]
 97%|█████████▋| 361362/371472 [6:35:43<52:01,  3.24it/s]
 97%|█████████▋| 361363/371472 [6:35:43<50:11,  3.36it/s]
 97%|█████████▋| 361364/371472 [6:35:44<48:39,  3.46it/s]
 97%|█████████▋| 361365/371472 [6:35:44<48:35,  3.47it/s]
 97%|█████████▋| 361366/371472 [6:35:44<48:48,  3.45it/s]
 97%|█████████▋| 361367/371472 [6:35:45<47:14,  3.57it/s]
 97%|█████████▋| 361368/371472 [6:35:45<46:42,  3.60it/s]
 97%|█████████▋| 361369/371472 [6:35:45<46:07,  3.65it/s]
 97%|█████████▋| 361370/371472 [6:35:45<49:56,  3.37it/s]
 97%|█████████▋| 361371/371472 [6:35:46<50:08,  3.36it/s]
 97%|█████████▋| 361372/371472 [6:35:46<48:16,  3.49it/s]
 97%|█████████▋| 361373/371472 [6:35:46<49:15,  3.42it/s]
 97%|█████████▋| 361374/371472 [6:35:47<50:40,  3.32it/s]
 97%|█████████▋| 361375/371472 [6:35:47<53:50,  3.13it/s]
 97%|█████████▋| 361376/371472 [6:35:47<51:00,  3.30it/s]
 97%|█████████▋| 361377/371472 [6:35:47<49:17,  3.41it/s]
 97%|█████████▋| 361378/371472 [6:35:48<49:20,  3.41it/s]
 97%|█████████▋| 361379/371472 [6:35:48<51:19,  3.28it/s]
 97%|█████████▋| 361380/371472 [6:35:48<52:19,  3.21it/s]
                                                         
{'loss': 2.5738, 'learning_rate': 1.2446400482664996e-07, 'epoch': 15.57}
+
 97%|█████████▋| 361380/371472 [6:35:48<52:19,  3.21it/s]
 97%|█████████▋| 361381/371472 [6:35:49<50:21,  3.34it/s]
 97%|█████████▋| 361382/371472 [6:35:49<48:36,  3.46it/s]
 97%|█████████▋| 361383/371472 [6:35:49<48:27,  3.47it/s]
 97%|█████████▋| 361384/371472 [6:35:50<50:53,  3.30it/s]
 97%|█████████▋| 361385/371472 [6:35:50<49:52,  3.37it/s]
 97%|█████████▋| 361386/371472 [6:35:50<48:18,  3.48it/s]
 97%|█████████▋| 361387/371472 [6:35:50<51:03,  3.29it/s]
 97%|█████████▋| 361388/371472 [6:35:51<51:54,  3.24it/s]
 97%|█████████▋| 361389/371472 [6:35:51<50:54,  3.30it/s]
 97%|█████████▋| 361390/371472 [6:35:52<56:30,  2.97it/s]
 97%|█████████▋| 361391/371472 [6:35:52<55:17,  3.04it/s]
 97%|█████████▋| 361392/371472 [6:35:52<52:55,  3.17it/s]
 97%|█████████▋| 361393/371472 [6:35:52<52:36,  3.19it/s]
 97%|█████████▋| 361394/371472 [6:35:53<50:03,  3.35it/s]
 97%|█████████▋| 361395/371472 [6:35:53<47:38,  3.53it/s]
 97%|█████████▋| 361396/371472 [6:35:53<45:39,  3.68it/s]
 97%|█████████▋| 361397/371472 [6:35:53<45:25,  3.70it/s]
 97%|█████████▋| 361398/371472 [6:35:54<51:12,  3.28it/s]
 97%|█████████▋| 361399/371472 [6:35:54<51:42,  3.25it/s]
 97%|█████████▋| 361400/371472 [6:35:54<48:56,  3.43it/s]
                                                         
{'loss': 2.612, 'learning_rate': 1.2441552285117113e-07, 'epoch': 15.57}
+
 97%|█████████▋| 361400/371472 [6:35:54<48:56,  3.43it/s]
 97%|█████████▋| 361401/371472 [6:35:55<51:25,  3.26it/s]
 97%|█████████▋| 361402/371472 [6:35:55<51:27,  3.26it/s]
 97%|█████████▋| 361403/371472 [6:35:55<49:41,  3.38it/s]
 97%|█████████▋| 361404/371472 [6:35:56<48:21,  3.47it/s]
 97%|█████████▋| 361405/371472 [6:35:56<46:18,  3.62it/s]
 97%|█████████▋| 361406/371472 [6:35:56<45:29,  3.69it/s]
 97%|█████████▋| 361407/371472 [6:35:56<46:07,  3.64it/s]
 97%|█████████▋| 361408/371472 [6:35:57<46:24,  3.61it/s]
 97%|█████████▋| 361409/371472 [6:35:57<45:55,  3.65it/s]
 97%|█████████▋| 361410/371472 [6:35:57<46:12,  3.63it/s]
 97%|█████████▋| 361411/371472 [6:35:58<48:48,  3.44it/s]
 97%|█████████▋| 361412/371472 [6:35:58<46:24,  3.61it/s]
 97%|█████████▋| 361413/371472 [6:35:58<45:39,  3.67it/s]
 97%|█████████▋| 361414/371472 [6:35:58<44:34,  3.76it/s]
 97%|█████████▋| 361415/371472 [6:35:59<45:57,  3.65it/s]
 97%|█████████▋| 361416/371472 [6:35:59<46:00,  3.64it/s]
 97%|█████████▋| 361417/371472 [6:35:59<46:23,  3.61it/s]
 97%|█████████▋| 361418/371472 [6:35:59<44:39,  3.75it/s]
 97%|█████████▋| 361419/371472 [6:36:00<47:28,  3.53it/s]
 97%|█████████▋| 361420/371472 [6:36:00<45:17,  3.70it/s]
                                                         
{'loss': 2.4066, 'learning_rate': 1.2436704087569218e-07, 'epoch': 15.57}
+
 97%|█████████▋| 361420/371472 [6:36:00<45:17,  3.70it/s]
 97%|█████████▋| 361421/371472 [6:36:00<48:08,  3.48it/s]
 97%|█████████▋| 361422/371472 [6:36:01<47:46,  3.51it/s]
 97%|█████████▋| 361423/371472 [6:36:01<46:23,  3.61it/s]
 97%|█████████▋| 361424/371472 [6:36:01<46:54,  3.57it/s]
 97%|█████████▋| 361425/371472 [6:36:01<47:46,  3.51it/s]
 97%|█████████▋| 361426/371472 [6:36:02<47:32,  3.52it/s]
 97%|█████████▋| 361427/371472 [6:36:02<49:14,  3.40it/s]
 97%|█████████▋| 361428/371472 [6:36:02<47:17,  3.54it/s]
 97%|█████████▋| 361429/371472 [6:36:03<46:15,  3.62it/s]
 97%|█████████▋| 361430/371472 [6:36:03<46:35,  3.59it/s]
 97%|█████████▋| 361431/371472 [6:36:03<50:39,  3.30it/s]
 97%|█████████▋| 361432/371472 [6:36:03<49:01,  3.41it/s]
 97%|█████████▋| 361433/371472 [6:36:04<47:50,  3.50it/s]
 97%|█████████▋| 361434/371472 [6:36:04<46:26,  3.60it/s]
 97%|█████████▋| 361435/371472 [6:36:04<45:17,  3.69it/s]
 97%|█████████▋| 361436/371472 [6:36:05<46:43,  3.58it/s]
 97%|█████████▋| 361437/371472 [6:36:05<45:41,  3.66it/s]
 97%|█████████▋| 361438/371472 [6:36:05<44:08,  3.79it/s]
 97%|█████████▋| 361439/371472 [6:36:05<43:28,  3.85it/s]
 97%|█████████▋| 361440/371472 [6:36:06<44:32,  3.75it/s]
                                                         
{'loss': 2.65, 'learning_rate': 1.2431855890021333e-07, 'epoch': 15.57}
+
 97%|█████████▋| 361440/371472 [6:36:06<44:32,  3.75it/s]
 97%|█████████▋| 361441/371472 [6:36:06<45:19,  3.69it/s]
 97%|█████████▋| 361442/371472 [6:36:06<44:52,  3.72it/s]
 97%|█████████▋| 361443/371472 [6:36:06<45:41,  3.66it/s]
 97%|█████████▋| 361444/371472 [6:36:07<47:16,  3.54it/s]
 97%|█████████▋| 361445/371472 [6:36:07<45:21,  3.68it/s]
 97%|█████████▋| 361446/371472 [6:36:07<46:00,  3.63it/s]
 97%|█████████▋| 361447/371472 [6:36:07<46:18,  3.61it/s]
 97%|█████████▋| 361448/371472 [6:36:08<45:32,  3.67it/s]
 97%|█████████▋| 361449/371472 [6:36:08<44:18,  3.77it/s]
 97%|█████████▋| 361450/371472 [6:36:08<46:42,  3.58it/s]
 97%|█████████▋| 361451/371472 [6:36:09<47:51,  3.49it/s]
 97%|█████████▋| 361452/371472 [6:36:09<50:11,  3.33it/s]
 97%|█████████▋| 361453/371472 [6:36:09<50:46,  3.29it/s]
 97%|█████████▋| 361454/371472 [6:36:09<47:48,  3.49it/s]
 97%|█████████▋| 361455/371472 [6:36:10<46:53,  3.56it/s]
 97%|█████████▋| 361456/371472 [6:36:10<46:45,  3.57it/s]
 97%|█████████▋| 361457/371472 [6:36:10<47:27,  3.52it/s]
 97%|█████████▋| 361458/371472 [6:36:11<46:14,  3.61it/s]
 97%|█████████▋| 361459/371472 [6:36:11<46:03,  3.62it/s]
 97%|█████████▋| 361460/371472 [6:36:11<45:34,  3.66it/s]
                                                         
{'loss': 2.5046, 'learning_rate': 1.242700769247344e-07, 'epoch': 15.57}
+
 97%|█████████▋| 361460/371472 [6:36:11<45:34,  3.66it/s]
 97%|█████████▋| 361461/371472 [6:36:11<44:59,  3.71it/s]
 97%|█████████▋| 361462/371472 [6:36:12<45:12,  3.69it/s]
 97%|███��█████▋| 361463/371472 [6:36:12<44:37,  3.74it/s]
 97%|█████████▋| 361464/371472 [6:36:12<45:06,  3.70it/s]
 97%|█████████▋| 361465/371472 [6:36:12<44:41,  3.73it/s]
 97%|█████████▋| 361466/371472 [6:36:13<45:19,  3.68it/s]
 97%|█████████▋| 361467/371472 [6:36:13<45:40,  3.65it/s]
 97%|█████████▋| 361468/371472 [6:36:13<45:58,  3.63it/s]
 97%|█████████▋| 361469/371472 [6:36:14<48:38,  3.43it/s]
 97%|█████████▋| 361470/371472 [6:36:14<47:46,  3.49it/s]
 97%|█████████▋| 361471/371472 [6:36:14<47:32,  3.51it/s]
 97%|█████████▋| 361472/371472 [6:36:14<47:07,  3.54it/s]
 97%|█████████▋| 361473/371472 [6:36:15<48:23,  3.44it/s]
 97%|█████████▋| 361474/371472 [6:36:15<49:23,  3.37it/s]
 97%|█████████▋| 361475/371472 [6:36:15<48:28,  3.44it/s]
 97%|█████████▋| 361476/371472 [6:36:16<50:22,  3.31it/s]
 97%|█████████▋| 361477/371472 [6:36:16<50:49,  3.28it/s]
 97%|█████████▋| 361478/371472 [6:36:16<49:30,  3.36it/s]
 97%|█████████▋| 361479/371472 [6:36:17<47:18,  3.52it/s]
 97%|█████████▋| 361480/371472 [6:36:17<48:04,  3.46it/s]
                                                         
{'loss': 2.7726, 'learning_rate': 1.2422159494925555e-07, 'epoch': 15.57}
+
 97%|█████████▋| 361480/371472 [6:36:17<48:04,  3.46it/s]
 97%|█████████▋| 361481/371472 [6:36:17<48:43,  3.42it/s]
 97%|█████████▋| 361482/371472 [6:36:17<47:41,  3.49it/s]
 97%|█████████▋| 361483/371472 [6:36:18<46:32,  3.58it/s]
 97%|█████████▋| 361484/371472 [6:36:18<47:35,  3.50it/s]
 97%|█████████▋| 361485/371472 [6:36:18<46:56,  3.55it/s]
 97%|█████████▋| 361486/371472 [6:36:19<46:41,  3.56it/s]
 97%|█████████▋| 361487/371472 [6:36:19<51:35,  3.23it/s]
 97%|█████████▋| 361488/371472 [6:36:19<55:02,  3.02it/s]
 97%|█████████▋| 361489/371472 [6:36:20<50:15,  3.31it/s]
 97%|█████████▋| 361490/371472 [6:36:20<51:52,  3.21it/s]
 97%|█████████▋| 361491/371472 [6:36:20<52:35,  3.16it/s]
 97%|█████████▋| 361492/371472 [6:36:20<50:48,  3.27it/s]
 97%|█████████▋| 361493/371472 [6:36:21<56:09,  2.96it/s]
 97%|█████████▋| 361494/371472 [6:36:21<1:01:40,  2.70it/s]
 97%|█████████▋| 361495/371472 [6:36:22<55:37,  2.99it/s]  
 97%|█████████▋| 361496/371472 [6:36:22<52:15,  3.18it/s]
 97%|█████████▋| 361497/371472 [6:36:22<51:36,  3.22it/s]
 97%|█████████▋| 361498/371472 [6:36:22<50:14,  3.31it/s]
 97%|█████████▋| 361499/371472 [6:36:23<47:37,  3.49it/s]
 97%|█████████▋| 361500/371472 [6:36:23<46:38,  3.56it/s]
                                                         
{'loss': 2.5325, 'learning_rate': 1.2417311297377662e-07, 'epoch': 15.57}
+
 97%|█████████▋| 361500/371472 [6:36:23<46:38,  3.56it/s]
 97%|█████████▋| 361501/371472 [6:36:23<52:03,  3.19it/s]
 97%|█████████▋| 361502/371472 [6:36:24<50:46,  3.27it/s]
 97%|█████████▋| 361503/371472 [6:36:24<49:44,  3.34it/s]
 97%|█████████▋| 361504/371472 [6:36:24<54:17,  3.06it/s]
 97%|█████████▋| 361505/371472 [6:36:25<54:51,  3.03it/s]
 97%|█████████▋| 361506/371472 [6:36:25<52:19,  3.17it/s]
 97%|█████████▋| 361507/371472 [6:36:25<50:15,  3.30it/s]
 97%|█████████▋| 361508/371472 [6:36:25<49:45,  3.34it/s]
 97%|█████████▋| 361509/371472 [6:36:26<51:08,  3.25it/s]
 97%|█████████▋| 361510/371472 [6:36:26<49:59,  3.32it/s]
 97%|█████████▋| 361511/371472 [6:36:26<51:23,  3.23it/s]
 97%|█████████▋| 361512/371472 [6:36:27<50:21,  3.30it/s]
 97%|█████████▋| 361513/371472 [6:36:27<50:04,  3.32it/s]
 97%|█████████▋| 361514/371472 [6:36:27<51:11,  3.24it/s]
 97%|█████████▋| 361515/371472 [6:36:28<49:48,  3.33it/s]
 97%|█████████▋| 361516/371472 [6:36:28<47:44,  3.48it/s]
 97%|█████████▋| 361517/371472 [6:36:28<45:47,  3.62it/s]
 97%|█████████▋| 361518/371472 [6:36:28<45:23,  3.66it/s]
 97%|█████████▋| 361519/371472 [6:36:29<45:09,  3.67it/s]
 97%|█████████▋| 361520/371472 [6:36:29<44:43,  3.71it/s]
                                                         
{'loss': 2.6197, 'learning_rate': 1.2412463099829777e-07, 'epoch': 15.57}
+
 97%|█████████▋| 361520/371472 [6:36:29<44:43,  3.71it/s]
 97%|█████████▋| 361521/371472 [6:36:29<44:51,  3.70it/s]
 97%|█████████▋| 361522/371472 [6:36:30<49:02,  3.38it/s]
 97%|█████████▋| 361523/371472 [6:36:30<47:00,  3.53it/s]
 97%|█████████▋| 361524/371472 [6:36:30<45:46,  3.62it/s]
 97%|█████████▋| 361525/371472 [6:36:30<45:16,  3.66it/s]
 97%|█████████▋| 361526/371472 [6:36:31<44:18,  3.74it/s]
 97%|█████████▋| 361527/371472 [6:36:31<45:10,  3.67it/s]
 97%|█████████▋| 361528/371472 [6:36:31<44:22,  3.74it/s]
 97%|█████████▋| 361529/371472 [6:36:31<45:51,  3.61it/s]
 97%|█████████▋| 361530/371472 [6:36:32<45:17,  3.66it/s]
 97%|█████████▋| 361531/371472 [6:36:32<45:06,  3.67it/s]
 97%|█████████▋| 361532/371472 [6:36:32<44:41,  3.71it/s]
 97%|█████████▋| 361533/371472 [6:36:32<43:40,  3.79it/s]
 97%|█████████▋| 361534/371472 [6:36:33<43:14,  3.83it/s]
 97%|█████████▋| 361535/371472 [6:36:33<45:09,  3.67it/s]
 97%|█████████▋| 361536/371472 [6:36:33<44:43,  3.70it/s]
 97%|█████████▋| 361537/371472 [6:36:34<45:42,  3.62it/s]
 97%|█████████▋| 361538/371472 [6:36:34<43:42,  3.79it/s]
 97%|█████████▋| 361539/371472 [6:36:34<44:05,  3.75it/s]
 97%|█████████▋| 361540/371472 [6:36:34<43:12,  3.83it/s]
                                                         
{'loss': 2.7923, 'learning_rate': 1.2407614902281885e-07, 'epoch': 15.57}
+
 97%|█████████▋| 361540/371472 [6:36:34<43:12,  3.83it/s]
 97%|█████████▋| 361541/371472 [6:36:35<43:16,  3.82it/s]
 97%|█████████▋| 361542/371472 [6:36:35<45:46,  3.61it/s]
 97%|█████████▋| 361543/371472 [6:36:35<46:41,  3.54it/s]
 97%|█████████▋| 361544/371472 [6:36:36<47:11,  3.51it/s]
 97%|█████████▋| 361545/371472 [6:36:36<45:28,  3.64it/s]
 97%|█████████▋| 361546/371472 [6:36:36<46:51,  3.53it/s]
 97%|█████████▋| 361547/371472 [6:36:36<46:16,  3.57it/s]
 97%|█████████▋| 361548/371472 [6:36:37<46:24,  3.56it/s]
 97%|█████████▋| 361549/371472 [6:36:37<44:45,  3.70it/s]
 97%|█████████▋| 361550/371472 [6:36:37<48:06,  3.44it/s]
 97%|█████████▋| 361551/371472 [6:36:37<47:13,  3.50it/s]
 97%|█████████▋| 361552/371472 [6:36:38<45:34,  3.63it/s]
 97%|█████████▋| 361553/371472 [6:36:38<44:57,  3.68it/s]
 97%|█████████▋| 361554/371472 [6:36:38<48:47,  3.39it/s]
 97%|█████████▋| 361555/371472 [6:36:39<48:01,  3.44it/s]
 97%|█████████▋| 361556/371472 [6:36:39<48:00,  3.44it/s]
 97%|█████████▋| 361557/371472 [6:36:39<48:03,  3.44it/s]
 97%|█████████▋| 361558/371472 [6:36:39<45:47,  3.61it/s]
 97%|█████████▋| 361559/371472 [6:36:40<44:33,  3.71it/s]
 97%|█████████▋| 361560/371472 [6:36:40<43:35,  3.79it/s]
                                                         
{'loss': 2.6133, 'learning_rate': 1.2402766704734e-07, 'epoch': 15.57}
+
 97%|█████████▋| 361560/371472 [6:36:40<43:35,  3.79it/s]
 97%|█████████▋| 361561/371472 [6:36:40<43:40,  3.78it/s]
 97%|█████████▋| 361562/371472 [6:36:40<43:39,  3.78it/s]
 97%|█████████▋| 361563/371472 [6:36:41<46:33,  3.55it/s]
 97%|█████████▋| 361564/371472 [6:36:41<47:25,  3.48it/s]
 97%|█████████▋| 361565/371472 [6:36:41<45:36,  3.62it/s]
 97%|█████████▋| 361566/371472 [6:36:42<44:53,  3.68it/s]
 97%|█████████▋| 361567/371472 [6:36:42<49:27,  3.34it/s]
 97%|█████████▋| 361568/371472 [6:36:42<46:35,  3.54it/s]
 97%|█████████▋| 361569/371472 [6:36:42<44:28,  3.71it/s]
 97%|█████████▋| 361570/371472 [6:36:43<48:32,  3.40it/s]
 97%|█████████▋| 361571/371472 [6:36:43<46:57,  3.51it/s]
 97%|█████████▋| 361572/371472 [6:36:43<45:10,  3.65it/s]
 97%|█████████▋| 361573/371472 [6:36:44<44:12,  3.73it/s]
 97%|█████████▋| 361574/371472 [6:36:44<42:45,  3.86it/s]
 97%|█████████▋| 361575/371472 [6:36:44<42:15,  3.90it/s]
 97%|█████████▋| 361576/371472 [6:36:44<44:11,  3.73it/s]
 97%|█████████▋| 361577/371472 [6:36:45<44:18,  3.72it/s]
 97%|█████████▋| 361578/371472 [6:36:45<43:57,  3.75it/s]
 97%|█████████▋| 361579/371472 [6:36:45<43:47,  3.77it/s]
 97%|█████████▋| 361580/371472 [6:36:46<50:45,  3.25it/s]
                                                         
{'loss': 2.674, 'learning_rate': 1.2397918507186104e-07, 'epoch': 15.57}
+
 97%|█████████▋| 361580/371472 [6:36:46<50:45,  3.25it/s]
 97%|█████████▋| 361581/371472 [6:36:46<49:06,  3.36it/s]
 97%|█████████▋| 361582/371472 [6:36:46<53:50,  3.06it/s]
 97%|█████████▋| 361583/371472 [6:36:47<51:25,  3.21it/s]
 97%|█████████▋| 361584/371472 [6:36:47<49:23,  3.34it/s]
 97%|█████████▋| 361585/371472 [6:36:47<47:30,  3.47it/s]
 97%|█████████▋| 361586/371472 [6:36:47<48:57,  3.37it/s]
 97%|█████████▋| 361587/371472 [6:36:48<47:47,  3.45it/s]
 97%|█████████▋| 361588/371472 [6:36:48<46:37,  3.53it/s]
 97%|█████████▋| 361589/371472 [6:36:48<45:26,  3.63it/s]
 97%|█████████▋| 361590/371472 [6:36:48<45:41,  3.60it/s]
 97%|█████████▋| 361591/371472 [6:36:49<44:15,  3.72it/s]
 97%|█████████▋| 361592/371472 [6:36:49<44:28,  3.70it/s]
 97%|█████████▋| 361593/371472 [6:36:49<44:30,  3.70it/s]
 97%|█████████▋| 361594/371472 [6:36:49<43:54,  3.75it/s]
 97%|█████████▋| 361595/371472 [6:36:50<43:34,  3.78it/s]
 97%|█████████▋| 361596/371472 [6:36:50<42:38,  3.86it/s]
 97%|█████████▋| 361597/371472 [6:36:50<42:55,  3.83it/s]
 97%|█████████▋| 361598/371472 [6:36:51<43:53,  3.75it/s]
 97%|█████████▋| 361599/371472 [6:36:51<43:47,  3.76it/s]
 97%|█████████▋| 361600/371472 [6:36:51<43:32,  3.78it/s]
                                                         
{'loss': 2.6656, 'learning_rate': 1.2393070309638222e-07, 'epoch': 15.57}
+
 97%|█████████▋| 361600/371472 [6:36:51<43:32,  3.78it/s]
 97%|█████████▋| 361601/371472 [6:36:51<42:28,  3.87it/s]
 97%|█████████▋| 361602/371472 [6:36:52<43:42,  3.76it/s]
 97%|█████████▋| 361603/371472 [6:36:52<45:50,  3.59it/s]
 97%|█████████▋| 361604/371472 [6:36:52<48:15,  3.41it/s]
 97%|█████████▋| 361605/371472 [6:36:53<52:22,  3.14it/s]
 97%|█████████▋| 361606/371472 [6:36:53<48:07,  3.42it/s]
 97%|█████████▋| 361607/371472 [6:36:53<46:27,  3.54it/s]
 97%|█████████▋| 361608/371472 [6:36:53<45:35,  3.61it/s]
 97%|█████████▋| 361609/371472 [6:36:54<44:47,  3.67it/s]
 97%|█████████▋| 361610/371472 [6:36:54<43:38,  3.77it/s]
 97%|█████████▋| 361611/371472 [6:36:54<43:05,  3.81it/s]
 97%|█████████▋| 361612/371472 [6:36:54<43:19,  3.79it/s]
 97%|█████████▋| 361613/371472 [6:36:55<45:19,  3.62it/s]
 97%|█████████▋| 361614/371472 [6:36:55<49:48,  3.30it/s]
 97%|█████████▋| 361615/371472 [6:36:55<47:00,  3.49it/s]
 97%|█████████▋| 361616/371472 [6:36:56<46:54,  3.50it/s]
 97%|█████████▋| 361617/371472 [6:36:56<49:02,  3.35it/s]
 97%|█████████▋| 361618/371472 [6:36:56<47:03,  3.49it/s]
 97%|█████████▋| 361619/371472 [6:36:56<45:27,  3.61it/s]
 97%|█████████▋| 361620/371472 [6:36:57<47:16,  3.47it/s]
                                                         
{'loss': 2.5405, 'learning_rate': 1.2388222112090326e-07, 'epoch': 15.58}
+
 97%|█████████▋| 361620/371472 [6:36:57<47:16,  3.47it/s]
 97%|█████████▋| 361621/371472 [6:36:57<48:43,  3.37it/s]
 97%|█████████▋| 361622/371472 [6:36:57<46:41,  3.52it/s]
 97%|█████████▋| 361623/371472 [6:36:58<46:51,  3.50it/s]
 97%|█████████▋| 361624/371472 [6:36:58<46:10,  3.55it/s]
 97%|█████████▋| 361625/371472 [6:36:58<45:27,  3.61it/s]
 97%|█████████▋| 361626/371472 [6:36:58<44:48,  3.66it/s]
 97%|█████████▋| 361627/371472 [6:36:59<43:41,  3.76it/s]
 97%|█████████▋| 361628/371472 [6:36:59<44:01,  3.73it/s]
 97%|█████████▋| 361629/371472 [6:36:59<45:21,  3.62it/s]
 97%|█████████▋| 361630/371472 [6:36:59<44:50,  3.66it/s]
 97%|█████████▋| 361631/371472 [6:37:00<47:06,  3.48it/s]
 97%|█████████▋| 361632/371472 [6:37:00<49:45,  3.30it/s]
 97%|█████████▋| 361633/371472 [6:37:00<47:26,  3.46it/s]
 97%|█████████▋| 361634/371472 [6:37:01<45:28,  3.61it/s]
 97%|█████████▋| 361635/371472 [6:37:01<45:24,  3.61it/s]
 97%|█████████▋| 361636/371472 [6:37:01<44:50,  3.66it/s]
 97%|█████████▋| 361637/371472 [6:37:01<44:18,  3.70it/s]
 97%|████████���▋| 361638/371472 [6:37:02<43:18,  3.78it/s]
 97%|█████████▋| 361639/371472 [6:37:02<47:17,  3.47it/s]
 97%|█████████▋| 361640/371472 [6:37:02<46:55,  3.49it/s]
                                                         
{'loss': 2.55, 'learning_rate': 1.2383373914542442e-07, 'epoch': 15.58}
+
 97%|█████████▋| 361640/371472 [6:37:02<46:55,  3.49it/s]
 97%|█████████▋| 361641/371472 [6:37:03<47:25,  3.46it/s]
 97%|█████████▋| 361642/371472 [6:37:03<48:39,  3.37it/s]
 97%|█████████▋| 361643/371472 [6:37:03<47:38,  3.44it/s]
 97%|█████████▋| 361644/371472 [6:37:03<45:23,  3.61it/s]
 97%|█████████▋| 361645/371472 [6:37:04<45:29,  3.60it/s]
 97%|█████████▋| 361646/371472 [6:37:04<45:06,  3.63it/s]
 97%|█████████▋| 361647/371472 [6:37:04<44:59,  3.64it/s]
 97%|█████████▋| 361648/371472 [6:37:05<46:11,  3.54it/s]
 97%|█████████▋| 361649/371472 [6:37:05<44:48,  3.65it/s]
 97%|█████████▋| 361650/371472 [6:37:05<45:08,  3.63it/s]
 97%|█████████▋| 361651/371472 [6:37:05<44:39,  3.66it/s]
 97%|█████████▋| 361652/371472 [6:37:06<57:13,  2.86it/s]
 97%|█████████▋| 361653/371472 [6:37:06<53:05,  3.08it/s]
 97%|█████████▋| 361654/371472 [6:37:06<50:34,  3.23it/s]
 97%|█████████▋| 361655/371472 [6:37:07<47:13,  3.47it/s]
 97%|█████████▋| 361656/371472 [6:37:07<45:53,  3.57it/s]
 97%|█████████▋| 361657/371472 [6:37:07<45:18,  3.61it/s]
 97%|█████████▋| 361658/371472 [6:37:08<45:19,  3.61it/s]
 97%|█████████▋| 361659/371472 [6:37:08<46:44,  3.50it/s]
 97%|█████████▋| 361660/371472 [6:37:08<45:22,  3.60it/s]
                                                         
{'loss': 2.5734, 'learning_rate': 1.2378525716994549e-07, 'epoch': 15.58}
+
 97%|█████████▋| 361660/371472 [6:37:08<45:22,  3.60it/s]
 97%|█████████▋| 361661/371472 [6:37:08<44:37,  3.66it/s]
 97%|█████████▋| 361662/371472 [6:37:09<45:44,  3.57it/s]
 97%|█████████▋| 361663/371472 [6:37:09<45:21,  3.60it/s]
 97%|█████████▋| 361664/371472 [6:37:09<49:51,  3.28it/s]
 97%|█████████▋| 361665/371472 [6:37:10<49:51,  3.28it/s]
 97%|█████████▋| 361666/371472 [6:37:10<48:42,  3.35it/s]
 97%|█████████▋| 361667/371472 [6:37:10<49:25,  3.31it/s]
 97%|█████████▋| 361668/371472 [6:37:11<57:26,  2.84it/s]
 97%|█████████▋| 361669/371472 [6:37:11<52:46,  3.10it/s]
 97%|█████████▋| 361670/371472 [6:37:11<50:02,  3.26it/s]
 97%|█████████▋| 361671/371472 [6:37:11<47:01,  3.47it/s]
 97%|█████████▋| 361672/371472 [6:37:12<46:53,  3.48it/s]
 97%|█████████▋| 361673/371472 [6:37:12<45:23,  3.60it/s]
 97%|█████████▋| 361674/371472 [6:37:12<45:38,  3.58it/s]
 97%|█████████▋| 361675/371472 [6:37:13<45:43,  3.57it/s]
 97%|█████████▋| 361676/371472 [6:37:13<47:35,  3.43it/s]
 97%|█████████▋| 361677/371472 [6:37:13<46:36,  3.50it/s]
 97%|█████████▋| 361678/371472 [6:37:13<46:49,  3.49it/s]
 97%|█████████▋| 361679/371472 [6:37:14<47:45,  3.42it/s]
 97%|█████████▋| 361680/371472 [6:37:14<50:20,  3.24it/s]
                                                         
{'loss': 2.462, 'learning_rate': 1.2373677519446653e-07, 'epoch': 15.58}
+
 97%|█████████▋| 361680/371472 [6:37:14<50:20,  3.24it/s]
 97%|█████████▋| 361681/371472 [6:37:14<48:49,  3.34it/s]
 97%|█████████▋| 361682/371472 [6:37:15<47:55,  3.40it/s]
 97%|█████████▋| 361683/371472 [6:37:15<45:54,  3.55it/s]
 97%|█████████▋| 361684/371472 [6:37:15<44:54,  3.63it/s]
 97%|█████████▋| 361685/371472 [6:37:15<47:14,  3.45it/s]
 97%|█████████▋| 361686/371472 [6:37:16<48:09,  3.39it/s]
 97%|█████████▋| 361687/371472 [6:37:16<47:02,  3.47it/s]
 97%|█████████▋| 361688/371472 [6:37:16<48:03,  3.39it/s]
 97%|█████████▋| 361689/371472 [6:37:17<45:52,  3.55it/s]
 97%|█████████▋| 361690/371472 [6:37:17<45:17,  3.60it/s]
 97%|█████████▋| 361691/371472 [6:37:17<46:57,  3.47it/s]
 97%|█████████▋| 361692/371472 [6:37:17<44:39,  3.65it/s]
 97%|█████████▋| 361693/371472 [6:37:18<43:22,  3.76it/s]
 97%|█████████▋| 361694/371472 [6:37:18<43:19,  3.76it/s]
 97%|█████████▋| 361695/371472 [6:37:18<42:05,  3.87it/s]
 97%|█████████▋| 361696/371472 [6:37:18<42:26,  3.84it/s]
 97%|█████████▋| 361697/371472 [6:37:19<41:51,  3.89it/s]
 97%|█████████▋| 361698/371472 [6:37:19<43:05,  3.78it/s]
 97%|█████████▋| 361699/371472 [6:37:19<43:04,  3.78it/s]
 97%|█████████▋| 361700/371472 [6:37:20<44:24,  3.67it/s]
                                                         
{'loss': 2.5638, 'learning_rate': 1.2368829321898768e-07, 'epoch': 15.58}
+
 97%|█████████▋| 361700/371472 [6:37:20<44:24,  3.67it/s]
 97%|█████████▋| 361701/371472 [6:37:20<45:16,  3.60it/s]
 97%|█████████▋| 361702/371472 [6:37:20<43:58,  3.70it/s]
 97%|█████████▋| 361703/371472 [6:37:20<43:51,  3.71it/s]
 97%|█████████▋| 361704/371472 [6:37:21<53:05,  3.07it/s]
 97%|█████████▋| 361705/371472 [6:37:21<54:52,  2.97it/s]
 97%|█████████▋| 361706/371472 [6:37:21<52:11,  3.12it/s]
 97%|█████████▋| 361707/371472 [6:37:22<52:41,  3.09it/s]
 97%|█████████▋| 361708/371472 [6:37:22<47:52,  3.40it/s]
 97%|█████████▋| 361709/371472 [6:37:22<48:44,  3.34it/s]
 97%|█████████▋| 361710/371472 [6:37:23<48:05,  3.38it/s]
 97%|█████████▋| 361711/371472 [6:37:23<46:28,  3.50it/s]
 97%|█████████▋| 361712/371472 [6:37:23<45:46,  3.55it/s]
 97%|█████████▋| 361713/371472 [6:37:23<46:07,  3.53it/s]
 97%|█████████▋| 361714/371472 [6:37:24<45:23,  3.58it/s]
 97%|█████████▋| 361715/371472 [6:37:24<44:23,  3.66it/s]
 97%|█████████▋| 361716/371472 [6:37:24<47:06,  3.45it/s]
 97%|█████████▋| 361717/371472 [6:37:24<44:22,  3.66it/s]
 97%|█████████▋| 361718/371472 [6:37:25<43:01,  3.78it/s]
 97%|█████████▋| 361719/371472 [6:37:25<43:49,  3.71it/s]
 97%|█████████▋| 361720/371472 [6:37:25<45:50,  3.55it/s]
                                                         
{'loss': 2.3672, 'learning_rate': 1.2363981124350875e-07, 'epoch': 15.58}
+
 97%|█████████▋| 361720/371472 [6:37:25<45:50,  3.55it/s]
 97%|█████████▋| 361721/371472 [6:37:26<51:00,  3.19it/s]
 97%|█████████▋| 361722/371472 [6:37:26<51:11,  3.17it/s]
 97%|█████████▋| 361723/371472 [6:37:26<51:03,  3.18it/s]
 97%|█████████▋| 361724/371472 [6:37:27<50:57,  3.19it/s]
 97%|█████████▋| 361725/371472 [6:37:27<53:45,  3.02it/s]
 97%|█████████▋| 361726/371472 [6:37:27<51:35,  3.15it/s]
 97%|█████████▋| 361727/371472 [6:37:28<51:38,  3.14it/s]
 97%|█████████▋| 361728/371472 [6:37:28<51:37,  3.15it/s]
 97%|█████████▋| 361729/371472 [6:37:28<51:09,  3.17it/s]
 97%|█████████▋| 361730/371472 [6:37:29<51:13,  3.17it/s]
 97%|█████████▋| 361731/371472 [6:37:29<48:55,  3.32it/s]
 97%|█████████▋| 361732/371472 [6:37:29<47:25,  3.42it/s]
 97%|█████████▋| 361733/371472 [6:37:29<47:36,  3.41it/s]
 97%|█████████▋| 361734/371472 [6:37:30<56:46,  2.86it/s]
 97%|█████████▋| 361735/371472 [6:37:30<52:23,  3.10it/s]
 97%|█████████▋| 361736/371472 [6:37:30<50:00,  3.24it/s]
 97%|█████████▋| 361737/371472 [6:37:31<51:14,  3.17it/s]
 97%|█████████▋| 361738/371472 [6:37:31<47:25,  3.42it/s]
 97%|█████████▋| 361739/371472 [6:37:31<46:25,  3.49it/s]
 97%|█████████▋| 361740/371472 [6:37:32<45:03,  3.60it/s]
                                                         
{'loss': 2.526, 'learning_rate': 1.235913292680299e-07, 'epoch': 15.58}
+
 97%|█████████▋| 361740/371472 [6:37:32<45:03,  3.60it/s]
 97%|█████████▋| 361741/371472 [6:37:32<45:55,  3.53it/s]
 97%|█████████▋| 361742/371472 [6:37:32<46:56,  3.46it/s]
 97%|█████████▋| 361743/371472 [6:37:32<46:43,  3.47it/s]
 97%|█████████▋| 361744/371472 [6:37:33<46:26,  3.49it/s]
 97%|█████████▋| 361745/371472 [6:37:33<44:37,  3.63it/s]
 97%|█████████▋| 361746/371472 [6:37:33<45:50,  3.54it/s]
 97%|█████████▋| 361747/371472 [6:37:34<47:01,  3.45it/s]
 97%|█████████▋| 361748/371472 [6:37:34<45:55,  3.53it/s]
 97%|█████████▋| 361749/371472 [6:37:34<48:50,  3.32it/s]
 97%|█████████▋| 361750/371472 [6:37:34<48:43,  3.33it/s]
 97%|█████████▋| 361751/371472 [6:37:35<48:35,  3.33it/s]
 97%|█████████▋| 361752/371472 [6:37:35<47:05,  3.44it/s]
 97%|█████████���| 361753/371472 [6:37:35<45:12,  3.58it/s]
 97%|█████████▋| 361754/371472 [6:37:36<44:40,  3.62it/s]
 97%|█████████▋| 361755/371472 [6:37:36<46:43,  3.47it/s]
 97%|█████████▋| 361756/371472 [6:37:36<46:33,  3.48it/s]
 97%|█████████▋| 361757/371472 [6:37:36<48:04,  3.37it/s]
 97%|█████████▋| 361758/371472 [6:37:37<49:40,  3.26it/s]
 97%|█████████▋| 361759/371472 [6:37:37<47:22,  3.42it/s]
 97%|█████████▋| 361760/371472 [6:37:37<51:24,  3.15it/s]
                                                         
{'loss': 2.6794, 'learning_rate': 1.2354284729255098e-07, 'epoch': 15.58}
+
 97%|█████████▋| 361760/371472 [6:37:37<51:24,  3.15it/s]
 97%|█████████▋| 361761/371472 [6:37:38<53:23,  3.03it/s]
 97%|█████████▋| 361762/371472 [6:37:38<50:13,  3.22it/s]
 97%|█████████▋| 361763/371472 [6:37:38<47:48,  3.39it/s]
 97%|█████████▋| 361764/371472 [6:37:39<48:05,  3.36it/s]
 97%|█████████▋| 361765/371472 [6:37:39<46:30,  3.48it/s]
 97%|█████████▋| 361766/371472 [6:37:39<45:31,  3.55it/s]
 97%|█████████▋| 361767/371472 [6:37:39<45:04,  3.59it/s]
 97%|█████████▋| 361768/371472 [6:37:40<48:03,  3.37it/s]
 97%|█████████▋| 361769/371472 [6:37:40<49:28,  3.27it/s]
 97%|█████████▋| 361770/371472 [6:37:40<47:30,  3.40it/s]
 97%|█████████▋| 361771/371472 [6:37:41<51:02,  3.17it/s]
 97%|█████████▋| 361772/371472 [6:37:41<50:54,  3.18it/s]
 97%|█████████▋| 361773/371472 [6:37:41<49:00,  3.30it/s]
 97%|█████████▋| 361774/371472 [6:37:42<47:36,  3.39it/s]
 97%|█████████▋| 361775/371472 [6:37:42<45:50,  3.53it/s]
 97%|█████████▋| 361776/371472 [6:37:42<47:19,  3.41it/s]
 97%|█████████▋| 361777/371472 [6:37:42<45:59,  3.51it/s]
 97%|█████████▋| 361778/371472 [6:37:43<45:16,  3.57it/s]
 97%|█████████▋| 361779/371472 [6:37:43<48:56,  3.30it/s]
 97%|█████████▋| 361780/371472 [6:37:43<49:40,  3.25it/s]
                                                         
{'loss': 2.5369, 'learning_rate': 1.2349436531707213e-07, 'epoch': 15.58}
+
 97%|█████████▋| 361780/371472 [6:37:43<49:40,  3.25it/s]
 97%|█████████▋| 361781/371472 [6:37:44<47:00,  3.44it/s]
 97%|█████████▋| 361782/371472 [6:37:44<45:29,  3.55it/s]
 97%|█████████▋| 361783/371472 [6:37:44<47:48,  3.38it/s]
 97%|█████████▋| 361784/371472 [6:37:44<46:05,  3.50it/s]
 97%|█████████▋| 361785/371472 [6:37:45<47:24,  3.41it/s]
 97%|█████████▋| 361786/371472 [6:37:45<46:11,  3.49it/s]
 97%|█████████▋| 361787/371472 [6:37:45<48:31,  3.33it/s]
 97%|█████████▋| 361788/371472 [6:37:46<47:04,  3.43it/s]
 97%|█████████▋| 361789/371472 [6:37:46<50:49,  3.17it/s]
 97%|█████████▋| 361790/371472 [6:37:46<50:25,  3.20it/s]
 97%|█████████▋| 361791/371472 [6:37:47<50:35,  3.19it/s]
 97%|█████████▋| 361792/371472 [6:37:47<50:41,  3.18it/s]
 97%|█████████▋| 361793/371472 [6:37:47<53:43,  3.00it/s]
 97%|█████████▋| 361794/371472 [6:37:48<50:23,  3.20it/s]
 97%|█████████▋| 361795/371472 [6:37:48<52:01,  3.10it/s]
 97%|█████████▋| 361796/371472 [6:37:48<49:24,  3.26it/s]
 97%|█████████▋| 361797/371472 [6:37:48<47:01,  3.43it/s]
 97%|█████████▋| 361798/371472 [6:37:49<46:34,  3.46it/s]
 97%|█████████▋| 361799/371472 [6:37:49<46:48,  3.44it/s]
 97%|█████████▋| 361800/371472 [6:37:49<47:41,  3.38it/s]
                                                         
{'loss': 2.4671, 'learning_rate': 1.2344588334159317e-07, 'epoch': 15.58}
+
 97%|█████████▋| 361800/371472 [6:37:49<47:41,  3.38it/s]
 97%|█████████▋| 361801/371472 [6:37:50<47:43,  3.38it/s]
 97%|█████████▋| 361802/371472 [6:37:50<47:10,  3.42it/s]
 97%|█████████▋| 361803/371472 [6:37:50<45:25,  3.55it/s]
 97%|█████████▋| 361804/371472 [6:37:51<50:30,  3.19it/s]
 97%|█████████▋| 361805/371472 [6:37:51<50:42,  3.18it/s]
 97%|█████████▋| 361806/371472 [6:37:51<49:07,  3.28it/s]
 97%|█████████▋| 361807/371472 [6:37:51<47:53,  3.36it/s]
 97%|█████████▋| 361808/371472 [6:37:52<47:31,  3.39it/s]
 97%|█████████▋| 361809/371472 [6:37:52<45:37,  3.53it/s]
 97%|█████████▋| 361810/371472 [6:37:52<46:51,  3.44it/s]
 97%|█████████▋| 361811/371472 [6:37:53<47:01,  3.42it/s]
 97%|█████████▋| 361812/371472 [6:37:53<48:12,  3.34it/s]
 97%|█████████▋| 361813/371472 [6:37:53<47:57,  3.36it/s]
 97%|█████████▋| 361814/371472 [6:37:54<47:48,  3.37it/s]
 97%|█████████▋| 361815/371472 [6:37:54<49:38,  3.24it/s]
 97%|█████████▋| 361816/371472 [6:37:54<48:21,  3.33it/s]
 97%|█████████▋| 361817/371472 [6:37:54<46:48,  3.44it/s]
 97%|█████████▋| 361818/371472 [6:37:55<45:42,  3.52it/s]
 97%|█████████▋| 361819/371472 [6:37:55<47:15,  3.40it/s]
 97%|█████████▋| 361820/371472 [6:37:55<45:16,  3.55it/s]
                                                         
{'loss': 2.4805, 'learning_rate': 1.2339740136611435e-07, 'epoch': 15.58}
+
 97%|█████████▋| 361820/371472 [6:37:55<45:16,  3.55it/s]
 97%|█████████▋| 361821/371472 [6:37:56<45:02,  3.57it/s]
 97%|█████████▋| 361822/371472 [6:37:56<44:27,  3.62it/s]
 97%|█████████▋| 361823/371472 [6:37:56<44:00,  3.65it/s]
 97%|█████████▋| 361824/371472 [6:37:56<47:54,  3.36it/s]
 97%|█████████▋| 361825/371472 [6:37:57<59:29,  2.70it/s]
 97%|█████████▋| 361826/371472 [6:37:57<58:28,  2.75it/s]
 97%|█████████▋| 361827/371472 [6:37:58<55:00,  2.92it/s]
 97%|█████████▋| 361828/371472 [6:37:58<53:28,  3.01it/s]
 97%|█████████▋| 361829/371472 [6:37:58<51:38,  3.11it/s]
 97%|█████████▋| 361830/371472 [6:37:58<48:17,  3.33it/s]
 97%|█████████▋| 361831/371472 [6:37:59<46:57,  3.42it/s]
 97%|█████████▋| 361832/371472 [6:37:59<49:09,  3.27it/s]
 97%|█████████▋| 361833/371472 [6:37:59<48:58,  3.28it/s]
 97%|█████████▋| 361834/371472 [6:38:00<48:51,  3.29it/s]
 97%|█████████▋| 361835/371472 [6:38:00<46:39,  3.44it/s]
 97%|█████████▋| 361836/371472 [6:38:00<45:33,  3.53it/s]
 97%|█████████▋| 361837/371472 [6:38:00<44:09,  3.64it/s]
 97%|█████████▋| 361838/371472 [6:38:01<43:27,  3.70it/s]
 97%|█████████▋| 361839/371472 [6:38:01<43:47,  3.67it/s]
 97%|█████████▋| 361840/371472 [6:38:01<47:14,  3.40it/s]
                                                         
{'loss': 2.5526, 'learning_rate': 1.233489193906354e-07, 'epoch': 15.59}
+
 97%|█████████▋| 361840/371472 [6:38:01<47:14,  3.40it/s]
 97%|█████████▋| 361841/371472 [6:38:02<44:51,  3.58it/s]
 97%|█████████▋| 361842/371472 [6:38:02<43:55,  3.65it/s]
 97%|█████████▋| 361843/371472 [6:38:02<45:40,  3.51it/s]
 97%|█████████▋| 361844/371472 [6:38:02<44:34,  3.60it/s]
 97%|█████████▋| 361845/371472 [6:38:03<46:51,  3.42it/s]
 97%|█████████▋| 361846/371472 [6:38:03<47:11,  3.40it/s]
 97%|█████████▋| 361847/371472 [6:38:03<48:37,  3.30it/s]
 97%|█████████▋| 361848/371472 [6:38:04<48:34,  3.30it/s]
 97%|█████████▋| 361849/371472 [6:38:04<46:12,  3.47it/s]
 97%|█████████▋| 361850/371472 [6:38:04<49:31,  3.24it/s]
 97%|█████████▋| 361851/371472 [6:38:05<49:03,  3.27it/s]
 97%|█████████▋| 361852/371472 [6:38:05<46:47,  3.43it/s]
 97%|█████████▋| 361853/371472 [6:38:05<46:54,  3.42it/s]
 97%|█████████▋| 361854/371472 [6:38:05<47:10,  3.40it/s]
 97%|█████████▋| 361855/371472 [6:38:06<45:42,  3.51it/s]
 97%|█████████▋| 361856/371472 [6:38:06<44:34,  3.60it/s]
 97%|█████████▋| 361857/371472 [6:38:06<45:37,  3.51it/s]
 97%|█████████▋| 361858/371472 [6:38:07<45:50,  3.50it/s]
 97%|█████████▋| 361859/371472 [6:38:07<44:26,  3.60it/s]
 97%|█████████▋| 361860/371472 [6:38:07<48:18,  3.32it/s]
                                                         
{'loss': 2.5908, 'learning_rate': 1.2330043741515655e-07, 'epoch': 15.59}
+
 97%|█████████▋| 361860/371472 [6:38:07<48:18,  3.32it/s]
 97%|█████████▋| 361861/371472 [6:38:07<45:21,  3.53it/s]
 97%|█████████▋| 361862/371472 [6:38:08<44:50,  3.57it/s]
 97%|█████████▋| 361863/371472 [6:38:08<43:36,  3.67it/s]
 97%|█████████▋| 361864/371472 [6:38:08<42:54,  3.73it/s]
 97%|█████████▋| 361865/371472 [6:38:08<43:24,  3.69it/s]
 97%|█████████▋| 361866/371472 [6:38:09<42:51,  3.74it/s]
 97%|█████████▋| 361867/371472 [6:38:09<44:43,  3.58it/s]
 97%|█████████▋| 361868/371472 [6:38:09<44:15,  3.62it/s]
 97%|█████████▋| 361869/371472 [6:38:10<42:45,  3.74it/s]
 97%|█████████▋| 361870/371472 [6:38:10<42:30,  3.76it/s]
 97%|█████████▋| 361871/371472 [6:38:10<44:04,  3.63it/s]
 97%|█████████▋| 361872/371472 [6:38:10<47:43,  3.35it/s]
 97%|█████████▋| 361873/371472 [6:38:11<47:13,  3.39it/s]
 97%|█████████▋| 361874/371472 [6:38:11<46:43,  3.42it/s]
 97%|█████████▋| 361875/371472 [6:38:11<48:37,  3.29it/s]
 97%|█████████▋| 361876/371472 [6:38:12<47:08,  3.39it/s]
 97%|█████████▋| 361877/371472 [6:38:12<48:40,  3.29it/s]
 97%|█████████▋| 361878/371472 [6:38:12<49:11,  3.25it/s]
 97%|█████████▋| 361879/371472 [6:38:13<47:33,  3.36it/s]
 97%|█████████▋| 361880/371472 [6:38:13<51:45,  3.09it/s]
                                                         
{'loss': 2.5348, 'learning_rate': 1.2325195543967762e-07, 'epoch': 15.59}
+
 97%|█████████▋| 361880/371472 [6:38:13<51:45,  3.09it/s]
 97%|█████████▋| 361881/371472 [6:38:13<49:04,  3.26it/s]
 97%|█████████▋| 361882/371472 [6:38:13<46:53,  3.41it/s]
 97%|█████████▋| 361883/371472 [6:38:14<45:41,  3.50it/s]
 97%|█████████▋| 361884/371472 [6:38:14<46:17,  3.45it/s]
 97%|█████████▋| 361885/371472 [6:38:14<47:56,  3.33it/s]
 97%|█████████▋| 361886/371472 [6:38:15<51:46,  3.09it/s]
 97%|█████████▋| 361887/371472 [6:38:15<48:57,  3.26it/s]
 97%|█████████▋| 361888/371472 [6:38:15<50:22,  3.17it/s]
 97%|█████████▋| 361889/371472 [6:38:16<53:46,  2.97it/s]
 97%|█████████▋| 361890/371472 [6:38:16<51:25,  3.11it/s]
 97%|█████████▋| 361891/371472 [6:38:16<52:31,  3.04it/s]
 97%|█████████▋| 361892/371472 [6:38:17<54:21,  2.94it/s]
 97%|█████████▋| 361893/371472 [6:38:17<52:53,  3.02it/s]
 97%|█████████▋| 361894/371472 [6:38:17<49:33,  3.22it/s]
 97%|█████████▋| 361895/371472 [6:38:18<47:55,  3.33it/s]
 97%|█████████▋| 361896/371472 [6:38:18<46:30,  3.43it/s]
 97%|█████████▋| 361897/371472 [6:38:18<45:35,  3.50it/s]
 97%|█████████▋| 361898/371472 [6:38:18<44:05,  3.62it/s]
 97%|█████████▋| 361899/371472 [6:38:19<43:33,  3.66it/s]
 97%|█████████▋| 361900/371472 [6:38:19<43:21,  3.68it/s]
                                                         
{'loss': 2.6776, 'learning_rate': 1.2320347346419877e-07, 'epoch': 15.59}
+
 97%|█████████▋| 361900/371472 [6:38:19<43:21,  3.68it/s]
 97%|█████████▋| 361901/371472 [6:38:19<44:32,  3.58it/s]
 97%|█████████▋| 361902/371472 [6:38:19<43:29,  3.67it/s]
 97%|█████████▋| 361903/371472 [6:38:20<45:18,  3.52it/s]
 97%|█████████▋| 361904/371472 [6:38:20<44:39,  3.57it/s]
 97%|█████████▋| 361905/371472 [6:38:20<47:26,  3.36it/s]
 97%|█████████▋| 361906/371472 [6:38:21<46:03,  3.46it/s]
 97%|█████████▋| 361907/371472 [6:38:21<44:55,  3.55it/s]
 97%|█████████▋| 361908/371472 [6:38:21<45:12,  3.53it/s]
 97%|█████████▋| 361909/371472 [6:38:21<45:36,  3.49it/s]
 97%|█████████▋| 361910/371472 [6:38:22<44:08,  3.61it/s]
 97%|█████████▋| 361911/371472 [6:38:22<43:25,  3.67it/s]
 97%|█████████▋| 361912/371472 [6:38:22<42:01,  3.79it/s]
 97%|█████████▋| 361913/371472 [6:38:23<43:14,  3.68it/s]
 97%|█████████▋| 361914/371472 [6:38:23<44:17,  3.60it/s]
 97%|█████████▋| 361915/371472 [6:38:23<42:37,  3.74it/s]
 97%|█████████▋| 361916/371472 [6:38:23<45:32,  3.50it/s]
 97%|█████████▋| 361917/371472 [6:38:24<44:29,  3.58it/s]
 97%|█████████▋| 361918/371472 [6:38:24<46:43,  3.41it/s]
 97%|█████████▋| 361919/371472 [6:38:24<48:16,  3.30it/s]
 97%|█████████▋| 361920/371472 [6:38:25<47:06,  3.38it/s]
                                                         
{'loss': 2.5266, 'learning_rate': 1.2315499148871984e-07, 'epoch': 15.59}
+
 97%|█████████▋| 361920/371472 [6:38:25<47:06,  3.38it/s]
 97%|█████████▋| 361921/371472 [6:38:25<44:58,  3.54it/s]
 97%|█████████▋| 361922/371472 [6:38:25<45:25,  3.50it/s]
 97%|█████████▋| 361923/371472 [6:38:25<47:10,  3.37it/s]
 97%|█████████▋| 361924/371472 [6:38:26<47:02,  3.38it/s]
 97%|█████████▋| 361925/371472 [6:38:26<46:46,  3.40it/s]
 97%|█████████▋| 361926/371472 [6:38:26<45:37,  3.49it/s]
 97%|█████████▋| 361927/371472 [6:38:27<46:42,  3.41it/s]
 97%|█████████▋| 361928/371472 [6:38:27<45:52,  3.47it/s]
 97%|█████████▋| 361929/371472 [6:38:27<44:42,  3.56it/s]
 97%|█████████▋| 361930/371472 [6:38:27<43:46,  3.63it/s]
 97%|█████████▋| 361931/371472 [6:38:28<45:05,  3.53it/s]
 97%|█████████▋| 361932/371472 [6:38:28<46:08,  3.45it/s]
 97%|█████████▋| 361933/371472 [6:38:28<44:54,  3.54it/s]
 97%|█████████▋| 361934/371472 [6:38:29<43:31,  3.65it/s]
 97%|█████████▋| 361935/371472 [6:38:29<44:10,  3.60it/s]
 97%|█████████▋| 361936/371472 [6:38:29<44:10,  3.60it/s]
 97%|█████████▋| 361937/371472 [6:38:29<42:56,  3.70it/s]
 97%|█████████▋| 361938/371472 [6:38:30<47:03,  3.38it/s]
 97%|█████████▋| 361939/371472 [6:38:30<45:31,  3.49it/s]
 97%|█████████▋| 361940/371472 [6:38:30<46:23,  3.42it/s]
                                                         
{'loss': 2.5518, 'learning_rate': 1.23106509513241e-07, 'epoch': 15.59}
+
 97%|█████████▋| 361940/371472 [6:38:30<46:23,  3.42it/s]
 97%|█████████▋| 361941/371472 [6:38:31<46:04,  3.45it/s]
 97%|█████████▋| 361942/371472 [6:38:31<50:12,  3.16it/s]
 97%|█████████▋| 361943/371472 [6:38:31<50:48,  3.13it/s]
 97%|█████████▋| 361944/371472 [6:38:32<48:14,  3.29it/s]
 97%|█████████▋| 361945/371472 [6:38:32<52:26,  3.03it/s]
 97%|█████████▋| 361946/371472 [6:38:32<56:06,  2.83it/s]
 97%|█████████▋| 361947/371472 [6:38:33<54:56,  2.89it/s]
 97%|█████████▋| 361948/371472 [6:38:33<50:24,  3.15it/s]
 97%|█████████▋| 361949/371472 [6:38:33<48:48,  3.25it/s]
 97%|█████████▋| 361950/371472 [6:38:33<46:46,  3.39it/s]
 97%|█████████▋| 361951/371472 [6:38:34<46:26,  3.42it/s]
 97%|█████████▋| 361952/371472 [6:38:34<45:23,  3.50it/s]
 97%|█████████▋| 361953/371472 [6:38:34<47:36,  3.33it/s]
 97%|█████████▋| 361954/371472 [6:38:35<48:08,  3.30it/s]
 97%|█████████▋| 361955/371472 [6:38:35<49:55,  3.18it/s]
 97%|█████████▋| 361956/371472 [6:38:35<47:24,  3.35it/s]
 97%|█████████▋| 361957/371472 [6:38:36<46:37,  3.40it/s]
 97%|█████████▋| 361958/371472 [6:38:36<44:11,  3.59it/s]
 97%|█████████▋| 361959/371472 [6:38:36<44:22,  3.57it/s]
 97%|█████████▋| 361960/371472 [6:38:36<46:34,  3.40it/s]
                                                         
{'loss': 2.6189, 'learning_rate': 1.2305802753776204e-07, 'epoch': 15.59}
+
 97%|█████████▋| 361960/371472 [6:38:36<46:34,  3.40it/s]
 97%|█████████▋| 361961/371472 [6:38:37<48:35,  3.26it/s]
 97%|█████████▋| 361962/371472 [6:38:37<46:26,  3.41it/s]
 97%|█████████▋| 361963/371472 [6:38:37<45:37,  3.47it/s]
 97%|█████████▋| 361964/371472 [6:38:38<47:02,  3.37it/s]
 97%|█████████▋| 361965/371472 [6:38:38<46:39,  3.40it/s]
 97%|█████████▋| 361966/371472 [6:38:38<48:14,  3.28it/s]
 97%|█████████▋| 361967/371472 [6:38:39<57:02,  2.78it/s]
 97%|█████████▋| 361968/371472 [6:38:39<51:48,  3.06it/s]
 97%|█████████▋| 361969/371472 [6:38:39<49:17,  3.21it/s]
 97%|█████████▋| 361970/371472 [6:38:40<46:55,  3.38it/s]
 97%|█████████▋| 361971/371472 [6:38:40<51:06,  3.10it/s]
 97%|█████████▋| 361972/371472 [6:38:40<47:34,  3.33it/s]
 97%|█████████▋| 361973/371472 [6:38:40<45:51,  3.45it/s]
 97%|█████████▋| 361974/371472 [6:38:41<43:11,  3.67it/s]
 97%|█████████▋| 361975/371472 [6:38:41<41:58,  3.77it/s]
 97%|█████████▋| 361976/371472 [6:38:41<43:04,  3.67it/s]
 97%|█████████▋| 361977/371472 [6:38:42<46:23,  3.41it/s]
 97%|█████████▋| 361978/371472 [6:38:42<46:46,  3.38it/s]
 97%|█████████▋| 361979/371472 [6:38:42<47:56,  3.30it/s]
 97%|█████████▋| 361980/371472 [6:38:42<46:18,  3.42it/s]
                                                         
{'loss': 2.5711, 'learning_rate': 1.230095455622832e-07, 'epoch': 15.59}
+
 97%|█████████▋| 361980/371472 [6:38:42<46:18,  3.42it/s]
 97%|█████████▋| 361981/371472 [6:38:43<44:17,  3.57it/s]
 97%|█████████▋| 361982/371472 [6:38:43<43:19,  3.65it/s]
 97%|█████████���| 361983/371472 [6:38:43<44:06,  3.59it/s]
 97%|█████████▋| 361984/371472 [6:38:43<44:19,  3.57it/s]
 97%|█████████▋| 361985/371472 [6:38:44<43:12,  3.66it/s]
 97%|█████████▋| 361986/371472 [6:38:44<43:22,  3.65it/s]
 97%|█████████▋| 361987/371472 [6:38:44<44:31,  3.55it/s]
 97%|█████████▋| 361988/371472 [6:38:45<43:39,  3.62it/s]
 97%|█████████▋| 361989/371472 [6:38:45<51:58,  3.04it/s]
 97%|█████████▋| 361990/371472 [6:38:45<47:41,  3.31it/s]
 97%|█████████▋| 361991/371472 [6:38:46<47:02,  3.36it/s]
 97%|█████████▋| 361992/371472 [6:38:46<45:12,  3.49it/s]
 97%|█████████▋| 361993/371472 [6:38:46<45:46,  3.45it/s]
 97%|█████████▋| 361994/371472 [6:38:46<46:37,  3.39it/s]
 97%|█████████▋| 361995/371472 [6:38:47<44:39,  3.54it/s]
 97%|█████████▋| 361996/371472 [6:38:47<46:07,  3.42it/s]
 97%|█████████▋| 361997/371472 [6:38:47<50:27,  3.13it/s]
 97%|█████████▋| 361998/371472 [6:38:48<47:34,  3.32it/s]
 97%|█████████▋| 361999/371472 [6:38:48<44:30,  3.55it/s]
 97%|█████████▋| 362000/371472 [6:38:48<44:07,  3.58it/s]
                                                         
{'loss': 2.5771, 'learning_rate': 1.2296106358680426e-07, 'epoch': 15.59}
+
 97%|█████████▋| 362000/371472 [6:38:48<44:07,  3.58it/s]
 97%|█████████▋| 362001/371472 [6:38:48<43:06,  3.66it/s]
 97%|█████████▋| 362002/371472 [6:38:49<42:04,  3.75it/s]
 97%|█████████▋| 362003/371472 [6:38:49<41:36,  3.79it/s]
 97%|█████████▋| 362004/371472 [6:38:49<41:44,  3.78it/s]
 97%|█████████▋| 362005/371472 [6:38:49<41:58,  3.76it/s]
 97%|█████████▋| 362006/371472 [6:38:50<41:41,  3.78it/s]
 97%|█████████▋| 362007/371472 [6:38:50<40:55,  3.85it/s]
 97%|█████████▋| 362008/371472 [6:38:50<40:54,  3.86it/s]
 97%|█████████▋| 362009/371472 [6:38:50<40:44,  3.87it/s]
 97%|█████████▋| 362010/371472 [6:38:51<41:00,  3.84it/s]
 97%|█████████▋| 362011/371472 [6:38:51<42:07,  3.74it/s]
 97%|█████████▋| 362012/371472 [6:38:51<42:35,  3.70it/s]
 97%|█████████▋| 362013/371472 [6:38:52<44:18,  3.56it/s]
 97%|█████████▋| 362014/371472 [6:38:52<43:33,  3.62it/s]
 97%|█████████▋| 362015/371472 [6:38:52<41:57,  3.76it/s]
 97%|█████████▋| 362016/371472 [6:38:52<45:47,  3.44it/s]
 97%|█████████▋| 362017/371472 [6:38:53<44:12,  3.57it/s]
 97%|█████████▋| 362018/371472 [6:38:53<43:57,  3.58it/s]
 97%|█████████▋| 362019/371472 [6:38:53<44:52,  3.51it/s]
 97%|█████████▋| 362020/371472 [6:38:54<44:48,  3.52it/s]
                                                         
{'loss': 2.7437, 'learning_rate': 1.229125816113254e-07, 'epoch': 15.59}
+
 97%|█████████▋| 362020/371472 [6:38:54<44:48,  3.52it/s]
 97%|█████████▋| 362021/371472 [6:38:54<42:48,  3.68it/s]
 97%|█████████▋| 362022/371472 [6:38:54<43:49,  3.59it/s]
 97%|█████████▋| 362023/371472 [6:38:54<44:10,  3.56it/s]
 97%|█████████▋| 362024/371472 [6:38:55<46:34,  3.38it/s]
 97%|█████████▋| 362025/371472 [6:38:55<43:58,  3.58it/s]
 97%|█████████▋| 362026/371472 [6:38:55<43:45,  3.60it/s]
 97%|█████████▋| 362027/371472 [6:38:56<42:38,  3.69it/s]
 97%|█████████▋| 362028/371472 [6:38:56<42:06,  3.74it/s]
 97%|█████████▋| 362029/371472 [6:38:56<41:58,  3.75it/s]
 97%|█████████▋| 362030/371472 [6:38:56<48:30,  3.24it/s]
 97%|█████████▋| 362031/371472 [6:38:57<46:26,  3.39it/s]
 97%|█████████▋| 362032/371472 [6:38:57<48:17,  3.26it/s]
 97%|█████████▋| 362033/371472 [6:38:57<46:23,  3.39it/s]
 97%|█████████▋| 362034/371472 [6:38:58<45:14,  3.48it/s]
 97%|█████████▋| 362035/371472 [6:38:58<43:29,  3.62it/s]
 97%|█████████▋| 362036/371472 [6:38:58<42:37,  3.69it/s]
 97%|█████████▋| 362037/371472 [6:38:58<41:50,  3.76it/s]
 97%|█████████▋| 362038/371472 [6:38:59<40:41,  3.86it/s]
 97%|█████████▋| 362039/371472 [6:38:59<41:36,  3.78it/s]
 97%|█████████▋| 362040/371472 [6:38:59<41:43,  3.77it/s]
                                                         
{'loss': 2.5378, 'learning_rate': 1.2286409963584648e-07, 'epoch': 15.59}
+
 97%|█████████▋| 362040/371472 [6:38:59<41:43,  3.77it/s]
 97%|█████████▋| 362041/371472 [6:38:59<42:12,  3.72it/s]
 97%|█████████▋| 362042/371472 [6:39:00<41:14,  3.81it/s]
 97%|█████████▋| 362043/371472 [6:39:00<41:31,  3.78it/s]
 97%|█████████▋| 362044/371472 [6:39:00<43:32,  3.61it/s]
 97%|█████████▋| 362045/371472 [6:39:00<43:10,  3.64it/s]
 97%|█████████▋| 362046/371472 [6:39:01<42:09,  3.73it/s]
 97%|█████████▋| 362047/371472 [6:39:01<41:52,  3.75it/s]
 97%|█████████▋| 362048/371472 [6:39:01<42:25,  3.70it/s]
 97%|█████████▋| 362049/371472 [6:39:02<43:07,  3.64it/s]
 97%|█████████▋| 362050/371472 [6:39:02<43:24,  3.62it/s]
 97%|█████████▋| 362051/371472 [6:39:02<44:37,  3.52it/s]
 97%|█████████▋| 362052/371472 [6:39:02<44:47,  3.50it/s]
 97%|█████████▋| 362053/371472 [6:39:03<45:51,  3.42it/s]
 97%|█████████▋| 362054/371472 [6:39:03<44:25,  3.53it/s]
 97%|█████████▋| 362055/371472 [6:39:03<44:38,  3.52it/s]
 97%|█████████▋| 362056/371472 [6:39:04<45:33,  3.44it/s]
 97%|█████████▋| 362057/371472 [6:39:04<43:45,  3.59it/s]
 97%|█████████▋| 362058/371472 [6:39:04<44:52,  3.50it/s]
 97%|█████████▋| 362059/371472 [6:39:04<42:49,  3.66it/s]
 97%|█████████▋| 362060/371472 [6:39:05<40:47,  3.85it/s]
                                                         
{'loss': 2.6151, 'learning_rate': 1.2281561766036763e-07, 'epoch': 15.59}
+
 97%|█████████▋| 362060/371472 [6:39:05<40:47,  3.85it/s]
 97%|█████████▋| 362061/371472 [6:39:05<43:22,  3.62it/s]
 97%|█████████▋| 362062/371472 [6:39:05<43:03,  3.64it/s]
 97%|█████████▋| 362063/371472 [6:39:06<45:55,  3.41it/s]
 97%|█████████▋| 362064/371472 [6:39:06<45:42,  3.43it/s]
 97%|█████████▋| 362065/371472 [6:39:06<45:42,  3.43it/s]
 97%|█████████▋| 362066/371472 [6:39:06<44:49,  3.50it/s]
 97%|█████████▋| 362067/371472 [6:39:07<44:28,  3.52it/s]
 97%|█████████▋| 362068/371472 [6:39:07<44:41,  3.51it/s]
 97%|█████████▋| 362069/371472 [6:39:07<44:42,  3.51it/s]
 97%|█████████▋| 362070/371472 [6:39:08<46:21,  3.38it/s]
 97%|█████████▋| 362071/371472 [6:39:08<44:18,  3.54it/s]
 97%|█████████▋| 362072/371472 [6:39:08<42:05,  3.72it/s]
 97%|█████████▋| 362073/371472 [6:39:08<44:21,  3.53it/s]
 97%|█████████▋| 362074/371472 [6:39:09<43:05,  3.63it/s]
 97%|█████████▋| 362075/371472 [6:39:09<42:19,  3.70it/s]
 97%|█████████▋| 362076/371472 [6:39:09<41:50,  3.74it/s]
 97%|█████████▋| 362077/371472 [6:39:09<41:13,  3.80it/s]
 97%|█████████▋| 362078/371472 [6:39:10<41:13,  3.80it/s]
 97%|█████████▋| 362079/371472 [6:39:10<41:15,  3.79it/s]
 97%|█████████▋| 362080/371472 [6:39:10<45:54,  3.41it/s]
                                                         
{'loss': 2.5708, 'learning_rate': 1.2276713568488868e-07, 'epoch': 15.6}
+
 97%|█████████▋| 362080/371472 [6:39:10<45:54,  3.41it/s]
 97%|█████████▋| 362081/371472 [6:39:11<46:58,  3.33it/s]
 97%|█████████▋| 362082/371472 [6:39:11<43:45,  3.58it/s]
 97%|█████████▋| 362083/371472 [6:39:11<42:58,  3.64it/s]
 97%|█████████▋| 362084/371472 [6:39:11<42:56,  3.64it/s]
 97%|█████████▋| 362085/371472 [6:39:12<41:33,  3.76it/s]
 97%|█████████▋| 362086/371472 [6:39:12<40:55,  3.82it/s]
 97%|█████████▋| 362087/371472 [6:39:12<39:26,  3.97it/s]
 97%|█████████▋| 362088/371472 [6:39:12<39:33,  3.95it/s]
 97%|█████████▋| 362089/371472 [6:39:13<39:34,  3.95it/s]
 97%|█████████▋| 362090/371472 [6:39:13<42:03,  3.72it/s]
 97%|█████████▋| 362091/371472 [6:39:13<43:09,  3.62it/s]
 97%|█████████▋| 362092/371472 [6:39:13<41:30,  3.77it/s]
 97%|█████████▋| 362093/371472 [6:39:14<42:59,  3.64it/s]
 97%|█████████▋| 362094/371472 [6:39:14<44:55,  3.48it/s]
 97%|█████████▋| 362095/371472 [6:39:14<47:02,  3.32it/s]
 97%|█████████▋| 362096/371472 [6:39:15<48:25,  3.23it/s]
 97%|█████████▋| 362097/371472 [6:39:15<47:41,  3.28it/s]
 97%|█████████▋| 362098/371472 [6:39:15<46:31,  3.36it/s]
 97%|█████████▋| 362099/371472 [6:39:16<44:50,  3.48it/s]
 97%|█████████▋| 362100/371472 [6:39:16<47:53,  3.26it/s]
                                                         
{'loss': 2.552, 'learning_rate': 1.2271865370940985e-07, 'epoch': 15.6}
+
 97%|█████████▋| 362100/371472 [6:39:16<47:53,  3.26it/s]
 97%|█████████▋| 362101/371472 [6:39:16<47:10,  3.31it/s]
 97%|█████████▋| 362102/371472 [6:39:16<44:24,  3.52it/s]
 97%|█████████▋| 362103/371472 [6:39:17<43:03,  3.63it/s]
 97%|█████████▋| 362104/371472 [6:39:17<42:05,  3.71it/s]
 97%|█████████▋| 362105/371472 [6:39:17<44:31,  3.51it/s]
 97%|█████████▋| 362106/371472 [6:39:18<45:01,  3.47it/s]
 97%|█████████▋| 362107/371472 [6:39:18<42:29,  3.67it/s]
 97%|█████████▋| 362108/371472 [6:39:18<45:13,  3.45it/s]
 97%|█████████▋| 362109/371472 [6:39:19<49:41,  3.14it/s]
 97%|█████████▋| 362110/371472 [6:39:19<49:09,  3.17it/s]
 97%|█████████▋| 362111/371472 [6:39:19<47:51,  3.26it/s]
 97%|█████████▋| 362112/371472 [6:39:19<46:10,  3.38it/s]
 97%|█████████▋| 362113/371472 [6:39:20<45:03,  3.46it/s]
 97%|█████████▋| 362114/371472 [6:39:20<47:23,  3.29it/s]
 97%|█████████▋| 362115/371472 [6:39:20<45:10,  3.45it/s]
 97%|█████████▋| 362116/371472 [6:39:21<44:29,  3.51it/s]
 97%|█████████▋| 362117/371472 [6:39:21<44:56,  3.47it/s]
 97%|█████████▋| 362118/371472 [6:39:21<43:00,  3.63it/s]
 97%|█████████▋| 362119/371472 [6:39:21<42:19,  3.68it/s]
 97%|█████████▋| 362120/371472 [6:39:22<42:48,  3.64it/s]
                                                         
{'loss': 2.6538, 'learning_rate': 1.226701717339309e-07, 'epoch': 15.6}
+
 97%|█████████▋| 362120/371472 [6:39:22<42:48,  3.64it/s]
 97%|█████████▋| 362121/371472 [6:39:22<43:52,  3.55it/s]
 97%|█████████▋| 362122/371472 [6:39:22<45:00,  3.46it/s]
 97%|█████████▋| 362123/371472 [6:39:23<50:05,  3.11it/s]
 97%|█████████▋| 362124/371472 [6:39:23<46:00,  3.39it/s]
 97%|█████████▋| 362125/371472 [6:39:23<47:23,  3.29it/s]
 97%|█████████▋| 362126/371472 [6:39:24<49:57,  3.12it/s]
 97%|█████████▋| 362127/371472 [6:39:24<46:07,  3.38it/s]
 97%|█████████▋| 362128/371472 [6:39:24<43:49,  3.55it/s]
 97%|█████████▋| 362129/371472 [6:39:24<44:32,  3.50it/s]
 97%|█████████▋| 362130/371472 [6:39:25<48:30,  3.21it/s]
 97%|█████████▋| 362131/371472 [6:39:25<47:27,  3.28it/s]
 97%|█████████▋| 362132/371472 [6:39:25<44:31,  3.50it/s]
 97%|█████████▋| 362133/371472 [6:39:25<42:49,  3.64it/s]
 97%|█████████▋| 362134/371472 [6:39:26<42:49,  3.63it/s]
 97%|█████████▋| 362135/371472 [6:39:26<41:20,  3.76it/s]
 97%|█████████▋| 362136/371472 [6:39:26<42:24,  3.67it/s]
 97%|█████████▋| 362137/371472 [6:39:27<45:27,  3.42it/s]
 97%|█████████▋| 362138/371472 [6:39:27<44:55,  3.46it/s]
 97%|█████████▋| 362139/371472 [6:39:27<43:22,  3.59it/s]
 97%|█████████▋| 362140/371472 [6:39:27<44:03,  3.53it/s]
                                                         
{'loss': 2.5349, 'learning_rate': 1.2262168975845205e-07, 'epoch': 15.6}
+
 97%|█████████▋| 362140/371472 [6:39:27<44:03,  3.53it/s]
 97%|█████████▋| 362141/371472 [6:39:28<43:28,  3.58it/s]
 97%|█████████▋| 362142/371472 [6:39:28<43:00,  3.62it/s]
 97%|█████████▋| 362143/371472 [6:39:28<42:31,  3.66it/s]
 97%|█████████▋| 362144/371472 [6:39:29<44:49,  3.47it/s]
 97%|█████████▋| 362145/371472 [6:39:29<47:13,  3.29it/s]
 97%|█████████▋| 362146/371472 [6:39:29<45:59,  3.38it/s]
 97%|█████████▋| 362147/371472 [6:39:29<45:20,  3.43it/s]
 97%|█████████▋| 362148/371472 [6:39:30<44:36,  3.48it/s]
 97%|█████████▋| 362149/371472 [6:39:30<42:52,  3.62it/s]
 97%|█████████▋| 362150/371472 [6:39:30<42:05,  3.69it/s]
 97%|█████████▋| 362151/371472 [6:39:31<42:26,  3.66it/s]
 97%|█████████▋| 362152/371472 [6:39:31<41:31,  3.74it/s]
 97%|█████████▋| 362153/371472 [6:39:31<41:17,  3.76it/s]
 97%|█████████▋| 362154/371472 [6:39:31<41:20,  3.76it/s]
 97%|█████████▋| 362155/371472 [6:39:32<41:30,  3.74it/s]
 97%|█████████▋| 362156/371472 [6:39:32<44:11,  3.51it/s]
 97%|█████████▋| 362157/371472 [6:39:32<43:21,  3.58it/s]
 97%|█████████▋| 362158/371472 [6:39:32<42:31,  3.65it/s]
 97%|█████████▋| 362159/371472 [6:39:33<43:01,  3.61it/s]
 97%|█████████▋| 362160/371472 [6:39:33<41:41,  3.72it/s]
                                                         
{'loss': 2.5672, 'learning_rate': 1.2257320778297312e-07, 'epoch': 15.6}
+
 97%|█████████▋| 362160/371472 [6:39:33<41:41,  3.72it/s]
 97%|█████████▋| 362161/371472 [6:39:33<41:46,  3.71it/s]
 97%|█████████▋| 362162/371472 [6:39:34<40:28,  3.83it/s]
 97%|█████████▋| 362163/371472 [6:39:34<40:29,  3.83it/s]
 97%|█████████▋| 362164/371472 [6:39:34<40:01,  3.88it/s]
 97%|█████████▋| 362165/371472 [6:39:34<39:05,  3.97it/s]
 97%|█████████▋| 362166/371472 [6:39:35<41:38,  3.72it/s]
 97%|█████████▋| 362167/371472 [6:39:35<41:56,  3.70it/s]
 97%|█████████▋| 362168/371472 [6:39:35<41:49,  3.71it/s]
 97%|█████████▋| 362169/371472 [6:39:35<42:02,  3.69it/s]
 97%|█████████▋| 362170/371472 [6:39:36<42:37,  3.64it/s]
 97%|█████████▋| 362171/371472 [6:39:36<48:44,  3.18it/s]
 97%|█████████▋| 362172/371472 [6:39:36<46:03,  3.37it/s]
 97%|█████████▋| 362173/371472 [6:39:37<46:10,  3.36it/s]
 97%|█████████▋| 362174/371472 [6:39:37<48:41,  3.18it/s]
 97%|█████████▋| 362175/371472 [6:39:37<45:23,  3.41it/s]
 97%|█████████▋| 362176/371472 [6:39:37<44:17,  3.50it/s]
 97%|█████████▋| 362177/371472 [6:39:38<42:53,  3.61it/s]
 97%|█████████▋| 362178/371472 [6:39:38<45:50,  3.38it/s]
 97%|█████████▋| 362179/371472 [6:39:38<44:29,  3.48it/s]
 97%|█████████▋| 362180/371472 [6:39:39<43:22,  3.57it/s]
                                                         
{'loss': 2.6594, 'learning_rate': 1.2252472580749427e-07, 'epoch': 15.6}
+
 97%|█████████▋| 362180/371472 [6:39:39<43:22,  3.57it/s]
 97%|█████████▋| 362181/371472 [6:39:39<48:15,  3.21it/s]
 97%|█████████▋| 362182/371472 [6:39:39<46:28,  3.33it/s]
 97%|█████████▋| 362183/371472 [6:39:40<46:11,  3.35it/s]
 97%|█████████▋| 362184/371472 [6:39:40<45:38,  3.39it/s]
 97%|█████████▋| 362185/371472 [6:39:40<43:53,  3.53it/s]
 98%|█████████▊| 362186/371472 [6:39:40<45:06,  3.43it/s]
 98%|█████████▊| 362187/371472 [6:39:41<44:25,  3.48it/s]
 98%|█████████▊| 362188/371472 [6:39:41<46:03,  3.36it/s]
 98%|█████████▊| 362189/371472 [6:39:41<44:22,  3.49it/s]
 98%|█████████▊| 362190/371472 [6:39:42<42:51,  3.61it/s]
 98%|█████████▊| 362191/371472 [6:39:42<41:50,  3.70it/s]
 98%|█████████▊| 362192/371472 [6:39:42<41:29,  3.73it/s]
 98%|█████████▊| 362193/371472 [6:39:42<42:19,  3.65it/s]
 98%|█████████▊| 362194/371472 [6:39:43<44:22,  3.48it/s]
 98%|█████████▊| 362195/371472 [6:39:43<43:41,  3.54it/s]
 98%|█████████▊| 362196/371472 [6:39:43<42:18,  3.65it/s]
 98%|█████████▊| 362197/371472 [6:39:43<43:03,  3.59it/s]
 98%|█████████▊| 362198/371472 [6:39:44<44:54,  3.44it/s]
 98%|█████████▊| 362199/371472 [6:39:44<48:26,  3.19it/s]
 98%|█████████▊| 362200/371472 [6:39:44<46:47,  3.30it/s]
                                                         
{'loss': 2.5935, 'learning_rate': 1.2247624383201534e-07, 'epoch': 15.6}
+
 98%|█████████▊| 362200/371472 [6:39:44<46:47,  3.30it/s]
 98%|█████████▊| 362201/371472 [6:39:45<44:45,  3.45it/s]
 98%|█████████▊| 362202/371472 [6:39:45<44:42,  3.46it/s]
 98%|█████████▊| 362203/371472 [6:39:45<42:08,  3.67it/s]
 98%|█████████▊| 362204/371472 [6:39:45<41:24,  3.73it/s]
 98%|█████████▊| 362205/371472 [6:39:46<43:40,  3.54it/s]
 98%|█████████▊| 362206/371472 [6:39:46<42:27,  3.64it/s]
 98%|█████████▊| 362207/371472 [6:39:46<44:11,  3.49it/s]
 98%|█████████▊| 362208/371472 [6:39:47<43:41,  3.53it/s]
 98%|█████████▊| 362209/371472 [6:39:47<44:32,  3.47it/s]
 98%|█████████▊| 362210/371472 [6:39:47<44:04,  3.50it/s]
 98%|█████████▊| 362211/371472 [6:39:48<44:15,  3.49it/s]
 98%|█████████▊| 362212/371472 [6:39:48<43:41,  3.53it/s]
 98%|█████████▊| 362213/371472 [6:39:48<44:14,  3.49it/s]
 98%|█████████▊| 362214/371472 [6:39:48<43:19,  3.56it/s]
 98%|█████████▊| 362215/371472 [6:39:49<42:25,  3.64it/s]
 98%|█████████▊| 362216/371472 [6:39:49<42:18,  3.65it/s]
 98%|█████████▊| 362217/371472 [6:39:49<41:54,  3.68it/s]
 98%|█████████▊| 362218/371472 [6:39:49<42:50,  3.60it/s]
 98%|█████████▊| 362219/371472 [6:39:50<43:55,  3.51it/s]
 98%|█████████▊| 362220/371472 [6:39:50<43:31,  3.54it/s]
                                                         
{'loss': 2.4667, 'learning_rate': 1.224277618565364e-07, 'epoch': 15.6}
+
 98%|█████████▊| 362220/371472 [6:39:50<43:31,  3.54it/s]
 98%|█████████▊| 362221/371472 [6:39:50<44:15,  3.48it/s]
 98%|█████████▊| 362222/371472 [6:39:51<41:50,  3.68it/s]
 98%|█████████▊| 362223/371472 [6:39:51<42:59,  3.58it/s]
 98%|█████████▊| 362224/371472 [6:39:51<43:36,  3.53it/s]
 98%|█████████▊| 362225/371472 [6:39:51<42:41,  3.61it/s]
 98%|█████████▊| 362226/371472 [6:39:52<43:05,  3.58it/s]
 98%|█████████▊| 362227/371472 [6:39:52<42:40,  3.61it/s]
 98%|█████████▊| 362228/371472 [6:39:52<43:03,  3.58it/s]
 98%|█████████▊| 362229/371472 [6:39:53<43:29,  3.54it/s]
 98%|█████████▊| 362230/371472 [6:39:53<45:14,  3.40it/s]
 98%|█████████▊| 362231/371472 [6:39:53<43:42,  3.52it/s]
 98%|█████████▊| 362232/371472 [6:39:53<42:18,  3.64it/s]
 98%|█████████▊| 362233/371472 [6:39:54<44:20,  3.47it/s]
 98%|█████████▊| 362234/371472 [6:39:54<43:20,  3.55it/s]
 98%|█████████▊| 362235/371472 [6:39:54<41:40,  3.69it/s]
 98%|█████████▊| 362236/371472 [6:39:54<40:55,  3.76it/s]
 98%|█████████▊| 362237/371472 [6:39:55<39:16,  3.92it/s]
 98%|█████████▊| 362238/371472 [6:39:55<39:40,  3.88it/s]
 98%|█████████▊| 362239/371472 [6:39:55<40:57,  3.76it/s]
 98%|█████████▊| 362240/371472 [6:39:55<40:18,  3.82it/s]
                                                         
{'loss': 2.5434, 'learning_rate': 1.2237927988105757e-07, 'epoch': 15.6}
+
 98%|█████████▊| 362240/371472 [6:39:55<40:18,  3.82it/s]
 98%|█████████▊| 362241/371472 [6:39:56<44:15,  3.48it/s]
 98%|█████████▊| 362242/371472 [6:39:56<48:10,  3.19it/s]
 98%|█████████▊| 362243/371472 [6:39:56<46:12,  3.33it/s]
 98%|█████████▊| 362244/371472 [6:39:57<45:10,  3.40it/s]
 98%|█████████▊| 362245/371472 [6:39:57<48:41,  3.16it/s]
 98%|█████████▊| 362246/371472 [6:39:57<49:01,  3.14it/s]
 98%|█████████▊| 362247/371472 [6:39:58<48:07,  3.19it/s]
 98%|█████████▊| 362248/371472 [6:39:58<48:08,  3.19it/s]
 98%|█████████▊| 362249/371472 [6:39:58<45:08,  3.40it/s]
 98%|█████████▊| 362250/371472 [6:39:59<46:19,  3.32it/s]
 98%|█████████▊| 362251/371472 [6:39:59<43:32,  3.53it/s]
 98%|█████████▊| 362252/371472 [6:39:59<41:29,  3.70it/s]
 98%|█████████▊| 362253/371472 [6:39:59<43:53,  3.50it/s]
 98%|█████████▊| 362254/371472 [6:40:00<45:32,  3.37it/s]
 98%|█████████▊| 362255/371472 [6:40:00<45:31,  3.37it/s]
 98%|█████████▊| 362256/371472 [6:40:00<43:04,  3.57it/s]
 98%|█████████▊| 362257/371472 [6:40:01<48:12,  3.19it/s]
 98%|█████████▊| 362258/371472 [6:40:01<50:42,  3.03it/s]
 98%|█████████▊| 362259/371472 [6:40:01<49:16,  3.12it/s]
 98%|█████████▊| 362260/371472 [6:40:02<46:42,  3.29it/s]
                                                         
{'loss': 2.5538, 'learning_rate': 1.223307979055786e-07, 'epoch': 15.6}
+
 98%|█████████▊| 362260/371472 [6:40:02<46:42,  3.29it/s]
 98%|█████████▊| 362261/371472 [6:40:02<44:57,  3.41it/s]
 98%|█████████▊| 362262/371472 [6:40:02<43:14,  3.55it/s]
 98%|█████████▊| 362263/371472 [6:40:02<41:35,  3.69it/s]
 98%|█████████▊| 362264/371472 [6:40:03<43:50,  3.50it/s]
 98%|█████████▊| 362265/371472 [6:40:03<44:16,  3.47it/s]
 98%|█████████▊| 362266/371472 [6:40:03<45:07,  3.40it/s]
 98%|█████████▊| 362267/371472 [6:40:04<45:17,  3.39it/s]
 98%|█████████▊| 362268/371472 [6:40:04<44:27,  3.45it/s]
 98%|█████████▊| 362269/371472 [6:40:04<42:56,  3.57it/s]
 98%|█████████▊| 362270/371472 [6:40:04<42:49,  3.58it/s]
 98%|█████████▊| 362271/371472 [6:40:05<42:07,  3.64it/s]
 98%|█████████▊| 362272/371472 [6:40:05<40:38,  3.77it/s]
 98%|█████████▊| 362273/371472 [6:40:05<41:50,  3.66it/s]
 98%|█████████▊| 362274/371472 [6:40:05<42:06,  3.64it/s]
 98%|█████████▊| 362275/371472 [6:40:06<41:13,  3.72it/s]
 98%|█████████▊| 362276/371472 [6:40:06<43:01,  3.56it/s]
 98%|█████████▊| 362277/371472 [6:40:06<41:34,  3.69it/s]
 98%|█████████▊| 362278/371472 [6:40:07<40:21,  3.80it/s]
 98%|█████████▊| 362279/371472 [6:40:07<41:11,  3.72it/s]
 98%|█████████▊| 362280/371472 [6:40:07<41:44,  3.67it/s]
                                                         
{'loss': 2.6149, 'learning_rate': 1.2228231593009976e-07, 'epoch': 15.6}
+
 98%|█████████▊| 362280/371472 [6:40:07<41:44,  3.67it/s]
 98%|█████████▊| 362281/371472 [6:40:07<40:14,  3.81it/s]
 98%|█████████▊| 362282/371472 [6:40:08<40:49,  3.75it/s]
 98%|█████████▊| 362283/371472 [6:40:08<46:20,  3.30it/s]
 98%|█████████▊| 362284/371472 [6:40:08<44:12,  3.46it/s]
 98%|█████████▊| 362285/371472 [6:40:09<46:51,  3.27it/s]
 98%|█████████▊| 362286/371472 [6:40:09<44:41,  3.43it/s]
 98%|█████████▊| 362287/371472 [6:40:09<45:49,  3.34it/s]
 98%|█████████▊| 362288/371472 [6:40:10<47:29,  3.22it/s]
 98%|█████████▊| 362289/371472 [6:40:10<45:35,  3.36it/s]
 98%|█████████▊| 362290/371472 [6:40:10<45:26,  3.37it/s]
 98%|█████████▊| 362291/371472 [6:40:10<45:33,  3.36it/s]
 98%|█████████▊| 362292/371472 [6:40:11<44:32,  3.43it/s]
 98%|█████████▊| 362293/371472 [6:40:11<46:35,  3.28it/s]
 98%|█████████▊| 362294/371472 [6:40:11<46:12,  3.31it/s]
 98%|█████████▊| 362295/371472 [6:40:12<43:28,  3.52it/s]
 98%|█████████▊| 362296/371472 [6:40:12<44:16,  3.45it/s]
 98%|█████████▊| 362297/371472 [6:40:12<42:54,  3.56it/s]
 98%|█████████▊| 362298/371472 [6:40:12<45:17,  3.38it/s]
 98%|█████████▊| 362299/371472 [6:40:13<45:14,  3.38it/s]
 98%|█████████▊| 362300/371472 [6:40:13<44:21,  3.45it/s]
                                                         
{'loss': 2.5803, 'learning_rate': 1.2223383395462083e-07, 'epoch': 15.6}
+
 98%|█████████▊| 362300/371472 [6:40:13<44:21,  3.45it/s]
 98%|█████████▊| 362301/371472 [6:40:13<42:33,  3.59it/s]
 98%|█████████▊| 362302/371472 [6:40:14<41:11,  3.71it/s]
 98%|█████████▊| 362303/371472 [6:40:14<41:19,  3.70it/s]
 98%|█████████▊| 362304/371472 [6:40:14<41:57,  3.64it/s]
 98%|█████████▊| 362305/371472 [6:40:14<45:34,  3.35it/s]
 98%|█████████▊| 362306/371472 [6:40:15<43:02,  3.55it/s]
 98%|█████████▊| 362307/371472 [6:40:15<46:31,  3.28it/s]
 98%|█████████▊| 362308/371472 [6:40:15<43:31,  3.51it/s]
 98%|█████████▊| 362309/371472 [6:40:16<42:38,  3.58it/s]
 98%|█████████▊| 362310/371472 [6:40:16<42:21,  3.61it/s]
 98%|█████████▊| 362311/371472 [6:40:16<43:01,  3.55it/s]
 98%|█████████▊| 362312/371472 [6:40:16<41:10,  3.71it/s]
 98%|█████████▊| 362313/371472 [6:40:17<41:16,  3.70it/s]
 98%|█████████▊| 362314/371472 [6:40:17<39:29,  3.86it/s]
 98%|█████████▊| 362315/371472 [6:40:17<39:24,  3.87it/s]
 98%|█████████▊| 362316/371472 [6:40:17<38:38,  3.95it/s]
 98%|█████████▊| 362317/371472 [6:40:18<40:01,  3.81it/s]
 98%|█████████▊| 362318/371472 [6:40:18<40:49,  3.74it/s]
 98%|█████████▊| 362319/371472 [6:40:18<44:08,  3.46it/s]
 98%|█████████▊| 362320/371472 [6:40:19<47:53,  3.19it/s]
                                                         
{'loss': 2.6754, 'learning_rate': 1.2218535197914198e-07, 'epoch': 15.61}
+
 98%|█████████▊| 362320/371472 [6:40:19<47:53,  3.19it/s]
 98%|█████████▊| 362321/371472 [6:40:19<48:03,  3.17it/s]
 98%|█████████▊| 362322/371472 [6:40:19<46:21,  3.29it/s]
 98%|█████████▊| 362323/371472 [6:40:20<46:45,  3.26it/s]
 98%|█████████▊| 362324/371472 [6:40:20<45:07,  3.38it/s]
 98%|█████████▊| 362325/371472 [6:40:20<47:54,  3.18it/s]
 98%|█████████▊| 362326/371472 [6:40:20<45:43,  3.33it/s]
 98%|█████████▊| 362327/371472 [6:40:21<44:19,  3.44it/s]
 98%|█████████▊| 362328/371472 [6:40:21<44:57,  3.39it/s]
 98%|█████████▊| 362329/371472 [6:40:21<42:17,  3.60it/s]
 98%|█████████▊| 362330/371472 [6:40:21<41:32,  3.67it/s]
 98%|█████��███▊| 362331/371472 [6:40:22<42:29,  3.59it/s]
 98%|█████████▊| 362332/371472 [6:40:22<45:49,  3.32it/s]
 98%|█████████▊| 362333/371472 [6:40:22<44:01,  3.46it/s]
 98%|█████████▊| 362334/371472 [6:40:23<44:04,  3.45it/s]
 98%|█████████▊| 362335/371472 [6:40:23<42:05,  3.62it/s]
 98%|█████████▊| 362336/371472 [6:40:23<41:14,  3.69it/s]
 98%|█████████▊| 362337/371472 [6:40:23<41:26,  3.67it/s]
 98%|█████████▊| 362338/371472 [6:40:24<40:07,  3.79it/s]
 98%|█████████▊| 362339/371472 [6:40:24<43:07,  3.53it/s]
 98%|█████████▊| 362340/371472 [6:40:24<43:39,  3.49it/s]
                                                         
{'loss': 2.5449, 'learning_rate': 1.2213687000366303e-07, 'epoch': 15.61}
+
 98%|█████████▊| 362340/371472 [6:40:24<43:39,  3.49it/s]
 98%|█████████▊| 362341/371472 [6:40:25<42:07,  3.61it/s]
 98%|█████████▊| 362342/371472 [6:40:25<42:01,  3.62it/s]
 98%|█████████▊| 362343/371472 [6:40:25<42:32,  3.58it/s]
 98%|█████████▊| 362344/371472 [6:40:25<40:26,  3.76it/s]
 98%|█████████▊| 362345/371472 [6:40:26<39:31,  3.85it/s]
 98%|█████████▊| 362346/371472 [6:40:26<40:08,  3.79it/s]
 98%|█████████▊| 362347/371472 [6:40:26<42:14,  3.60it/s]
 98%|█████████▊| 362348/371472 [6:40:26<41:05,  3.70it/s]
 98%|█████████▊| 362349/371472 [6:40:27<42:15,  3.60it/s]
 98%|█████████▊| 362350/371472 [6:40:27<40:43,  3.73it/s]
 98%|█████████▊| 362351/371472 [6:40:27<44:59,  3.38it/s]
 98%|█████████▊| 362352/371472 [6:40:28<42:36,  3.57it/s]
 98%|█████████▊| 362353/371472 [6:40:28<41:01,  3.70it/s]
 98%|█████████▊| 362354/371472 [6:40:28<41:54,  3.63it/s]
 98%|█████████▊| 362355/371472 [6:40:28<40:30,  3.75it/s]
 98%|█████████▊| 362356/371472 [6:40:29<40:54,  3.71it/s]
 98%|█████████▊| 362357/371472 [6:40:29<42:43,  3.56it/s]
 98%|█████████▊| 362358/371472 [6:40:29<41:20,  3.67it/s]
 98%|█████████▊| 362359/371472 [6:40:30<41:55,  3.62it/s]
 98%|█████████▊| 362360/371472 [6:40:30<46:07,  3.29it/s]
                                                         
{'loss': 2.7549, 'learning_rate': 1.220883880281842e-07, 'epoch': 15.61}
+
 98%|█████████▊| 362360/371472 [6:40:30<46:07,  3.29it/s]
 98%|█████████▊| 362361/371472 [6:40:30<44:59,  3.37it/s]
 98%|█████████▊| 362362/371472 [6:40:30<46:25,  3.27it/s]
 98%|█████████▊| 362363/371472 [6:40:31<45:56,  3.31it/s]
 98%|█████████▊| 362364/371472 [6:40:31<47:08,  3.22it/s]
 98%|█████████▊| 362365/371472 [6:40:31<49:37,  3.06it/s]
 98%|█████████▊| 362366/371472 [6:40:32<46:12,  3.28it/s]
 98%|█████████▊| 362367/371472 [6:40:32<43:04,  3.52it/s]
 98%|█████████▊| 362368/371472 [6:40:32<41:33,  3.65it/s]
 98%|█████████▊| 362369/371472 [6:40:32<40:52,  3.71it/s]
 98%|█████████▊| 362370/371472 [6:40:33<40:58,  3.70it/s]
 98%|█████████▊| 362371/371472 [6:40:33<40:00,  3.79it/s]
 98%|█████████▊| 362372/371472 [6:40:33<42:45,  3.55it/s]
 98%|█████████▊| 362373/371472 [6:40:34<41:29,  3.65it/s]
 98%|█████████▊| 362374/371472 [6:40:34<44:27,  3.41it/s]
 98%|█████████▊| 362375/371472 [6:40:34<45:15,  3.35it/s]
 98%|█████████▊| 362376/371472 [6:40:34<42:40,  3.55it/s]
 98%|█████████▊| 362377/371472 [6:40:35<46:55,  3.23it/s]
 98%|█████████▊| 362378/371472 [6:40:35<46:47,  3.24it/s]
 98%|█████████▊| 362379/371472 [6:40:35<47:20,  3.20it/s]
 98%|█████████▊| 362380/371472 [6:40:36<47:17,  3.20it/s]
                                                         
{'loss': 2.437, 'learning_rate': 1.2203990605270525e-07, 'epoch': 15.61}
+
 98%|█████████▊| 362380/371472 [6:40:36<47:17,  3.20it/s]
 98%|█████████▊| 362381/371472 [6:40:36<46:40,  3.25it/s]
 98%|█████████▊| 362382/371472 [6:40:36<46:25,  3.26it/s]
 98%|█████████▊| 362383/371472 [6:40:37<45:50,  3.30it/s]
 98%|█████████▊| 362384/371472 [6:40:37<45:04,  3.36it/s]
 98%|█████████▊| 362385/371472 [6:40:37<43:33,  3.48it/s]
 98%|█████████▊| 362386/371472 [6:40:38<43:09,  3.51it/s]
 98%|█████████▊| 362387/371472 [6:40:38<42:36,  3.55it/s]
 98%|█████████▊| 362388/371472 [6:40:38<43:24,  3.49it/s]
 98%|█████████▊| 362389/371472 [6:40:38<42:13,  3.59it/s]
 98%|█████████▊| 362390/371472 [6:40:39<41:15,  3.67it/s]
 98%|█████████▊| 362391/371472 [6:40:39<41:19,  3.66it/s]
 98%|█████████▊| 362392/371472 [6:40:39<40:10,  3.77it/s]
 98%|█████████▊| 362393/371472 [6:40:39<40:32,  3.73it/s]
 98%|█████████▊| 362394/371472 [6:40:40<43:21,  3.49it/s]
 98%|█████████▊| 362395/371472 [6:40:40<42:20,  3.57it/s]
 98%|█████████▊| 362396/371472 [6:40:40<42:42,  3.54it/s]
 98%|█████████▊| 362397/371472 [6:40:41<41:50,  3.61it/s]
 98%|█████████▊| 362398/371472 [6:40:41<41:24,  3.65it/s]
 98%|█████████▊| 362399/371472 [6:40:41<40:58,  3.69it/s]
 98%|█████████▊| 362400/371472 [6:40:41<40:46,  3.71it/s]
                                                         
{'loss': 2.5268, 'learning_rate': 1.219914240772264e-07, 'epoch': 15.61}
+
 98%|█████████▊| 362400/371472 [6:40:41<40:46,  3.71it/s]
 98%|█████████▊| 362401/371472 [6:40:42<41:20,  3.66it/s]
 98%|█████████▊| 362402/371472 [6:40:42<41:27,  3.65it/s]
 98%|█████████▊| 362403/371472 [6:40:42<43:14,  3.50it/s]
 98%|█████████▊| 362404/371472 [6:40:42<42:33,  3.55it/s]
 98%|█████████▊| 362405/371472 [6:40:43<43:02,  3.51it/s]
 98%|█████████▊| 362406/371472 [6:40:43<43:26,  3.48it/s]
 98%|█████████▊| 362407/371472 [6:40:43<43:44,  3.45it/s]
 98%|█████████▊| 362408/371472 [6:40:44<44:19,  3.41it/s]
 98%|█████████▊| 362409/371472 [6:40:44<42:26,  3.56it/s]
 98%|█████████▊| 362410/371472 [6:40:44<46:24,  3.25it/s]
 98%|█████████▊| 362411/371472 [6:40:45<43:54,  3.44it/s]
 98%|█████████▊| 362412/371472 [6:40:45<42:19,  3.57it/s]
 98%|█████████▊| 362413/371472 [6:40:45<42:05,  3.59it/s]
 98%|█████████▊| 362414/371472 [6:40:45<42:38,  3.54it/s]
 98%|█████████▊| 362415/371472 [6:40:46<42:22,  3.56it/s]
 98%|█████████▊| 362416/371472 [6:40:46<43:31,  3.47it/s]
 98%|█████████▊| 362417/371472 [6:40:46<42:47,  3.53it/s]
 98%|█████████▊| 362418/371472 [6:40:46<41:06,  3.67it/s]
 98%|█████████▊| 362419/371472 [6:40:47<41:02,  3.68it/s]
 98%|█████████▊| 362420/371472 [6:40:47<41:27,  3.64it/s]
                                                         
{'loss': 2.6461, 'learning_rate': 1.2194294210174747e-07, 'epoch': 15.61}
+
 98%|█████████▊| 362420/371472 [6:40:47<41:27,  3.64it/s]
 98%|█████████▊| 362421/371472 [6:40:47<41:53,  3.60it/s]
 98%|█████████▊| 362422/371472 [6:40:48<40:46,  3.70it/s]
 98%|█████████▊| 362423/371472 [6:40:48<40:57,  3.68it/s]
 98%|█████████▊| 362424/371472 [6:40:48<44:48,  3.37it/s]
 98%|█████████▊| 362425/371472 [6:40:48<46:00,  3.28it/s]
 98%|█████████▊| 362426/371472 [6:40:49<44:50,  3.36it/s]
 98%|█████████▊| 362427/371472 [6:40:49<43:33,  3.46it/s]
 98%|█████████▊| 362428/371472 [6:40:49<44:13,  3.41it/s]
 98%|█████████▊| 362429/371472 [6:40:50<45:31,  3.31it/s]
 98%|█████████▊| 362430/371472 [6:40:50<44:28,  3.39it/s]
 98%|█████████▊| 362431/371472 [6:40:50<42:33,  3.54it/s]
 98%|█████████▊| 362432/371472 [6:40:50<41:47,  3.61it/s]
 98%|█████████▊| 362433/371472 [6:40:51<41:24,  3.64it/s]
 98%|█████████▊| 362434/371472 [6:40:51<40:52,  3.69it/s]
 98%|█████████▊| 362435/371472 [6:40:51<39:32,  3.81it/s]
 98%|█████████▊| 362436/371472 [6:40:52<41:23,  3.64it/s]
 98%|█████████▊| 362437/371472 [6:40:52<40:39,  3.70it/s]
 98%|█████████▊| 362438/371472 [6:40:52<42:35,  3.54it/s]
 98%|█████████▊| 362439/371472 [6:40:52<40:43,  3.70it/s]
 98%|█████████▊| 362440/371472 [6:40:53<41:07,  3.66it/s]
                                                         
{'loss': 2.5633, 'learning_rate': 1.2189446012626862e-07, 'epoch': 15.61}
+
 98%|█████████▊| 362440/371472 [6:40:53<41:07,  3.66it/s]
 98%|█████████▊| 362441/371472 [6:40:53<43:25,  3.47it/s]
 98%|█████████▊| 362442/371472 [6:40:53<41:24,  3.63it/s]
 98%|█████████▊| 362443/371472 [6:40:53<40:57,  3.67it/s]
 98%|█████████▊| 362444/371472 [6:40:54<44:47,  3.36it/s]
 98%|█████████▊| 362445/371472 [6:40:54<43:03,  3.49it/s]
 98%|█████████▊| 362446/371472 [6:40:54<42:16,  3.56it/s]
 98%|█████████▊| 362447/371472 [6:40:55<42:30,  3.54it/s]
 98%|█████████▊| 362448/371472 [6:40:55<41:33,  3.62it/s]
 98%|█████████▊| 362449/371472 [6:40:55<44:03,  3.41it/s]
 98%|█████████▊| 362450/371472 [6:40:56<42:47,  3.51it/s]
 98%|█████████▊| 362451/371472 [6:40:56<41:10,  3.65it/s]
 98%|█████████▊| 362452/371472 [6:40:56<41:00,  3.67it/s]
 98%|█████████▊| 362453/371472 [6:40:56<41:45,  3.60it/s]
 98%|█████████▊| 362454/371472 [6:40:57<41:18,  3.64it/s]
 98%|█████████▊| 362455/371472 [6:40:57<42:22,  3.55it/s]
 98%|█████████▊| 362456/371472 [6:40:57<41:24,  3.63it/s]
 98%|█████████▊| 362457/371472 [6:40:57<42:41,  3.52it/s]
 98%|█████████▊| 362458/371472 [6:40:58<42:06,  3.57it/s]
 98%|█████████▊| 362459/371472 [6:40:58<43:54,  3.42it/s]
 98%|█████████▊| 362460/371472 [6:40:58<42:45,  3.51it/s]
                                                         
{'loss': 2.7042, 'learning_rate': 1.218459781507897e-07, 'epoch': 15.61}
+
 98%|█████████▊| 362460/371472 [6:40:58<42:45,  3.51it/s]
 98%|█████████▊| 362461/371472 [6:40:59<45:54,  3.27it/s]
 98%|█████████▊| 362462/371472 [6:40:59<44:04,  3.41it/s]
 98%|█████████▊| 362463/371472 [6:40:59<42:42,  3.52it/s]
 98%|█████████▊| 362464/371472 [6:40:59<40:46,  3.68it/s]
 98%|█████████▊| 362465/371472 [6:41:00<41:15,  3.64it/s]
 98%|█████████▊| 362466/371472 [6:41:00<41:48,  3.59it/s]
 98%|█████████▊| 362467/371472 [6:41:00<41:27,  3.62it/s]
 98%|█████████▊| 362468/371472 [6:41:01<40:41,  3.69it/s]
 98%|█████████▊| 362469/371472 [6:41:01<44:35,  3.37it/s]
 98%|█████████▊| 362470/371472 [6:41:01<44:05,  3.40it/s]
 98%|█████████▊| 362471/371472 [6:41:01<42:37,  3.52it/s]
 98%|█████████▊| 362472/371472 [6:41:02<42:19,  3.54it/s]
 98%|█████████▊| 362473/371472 [6:41:02<40:53,  3.67it/s]
 98%|█████████▊| 362474/371472 [6:41:02<39:30,  3.80it/s]
 98%|█████████▊| 362475/371472 [6:41:02<38:56,  3.85it/s]
 98%|█████████▊| 362476/371472 [6:41:03<43:40,  3.43it/s]
 98%|█████████▊| 362477/371472 [6:41:03<42:17,  3.54it/s]
 98%|█████████▊| 362478/371472 [6:41:03<40:16,  3.72it/s]
 98%|█████████▊| 362479/371472 [6:41:04<41:56,  3.57it/s]
 98%|█████████▊| 362480/371472 [6:41:04<40:44,  3.68it/s]
                                                         
{'loss': 2.774, 'learning_rate': 1.2179749617531085e-07, 'epoch': 15.61}
+
 98%|█████████▊| 362480/371472 [6:41:04<40:44,  3.68it/s]
 98%|█████████▊| 362481/371472 [6:41:04<40:07,  3.73it/s]
 98%|█████████▊| 362482/371472 [6:41:04<38:55,  3.85it/s]
 98%|█████████▊| 362483/371472 [6:41:05<38:45,  3.87it/s]
 98%|█████████▊| 362484/371472 [6:41:05<38:13,  3.92it/s]
 98%|█████████▊| 362485/371472 [6:41:05<38:22,  3.90it/s]
 98%|█████████▊| 362486/371472 [6:41:05<39:20,  3.81it/s]
 98%|█████████▊| 362487/371472 [6:41:06<40:16,  3.72it/s]
 98%|█████████▊| 362488/371472 [6:41:06<43:19,  3.46it/s]
 98%|█████████▊| 362489/371472 [6:41:06<42:27,  3.53it/s]
 98%|█████████▊| 362490/371472 [6:41:07<42:07,  3.55it/s]
 98%|█████████▊| 362491/371472 [6:41:07<44:14,  3.38it/s]
 98%|█████████▊| 362492/371472 [6:41:07<42:38,  3.51it/s]
 98%|█████████▊| 362493/371472 [6:41:07<41:49,  3.58it/s]
 98%|█████████▊| 362494/371472 [6:41:08<41:36,  3.60it/s]
 98%|█████████▊| 362495/371472 [6:41:08<41:50,  3.58it/s]
 98%|█████████▊| 362496/371472 [6:41:08<42:28,  3.52it/s]
 98%|█████████▊| 362497/371472 [6:41:09<44:58,  3.33it/s]
 98%|█████████▊| 362498/371472 [6:41:09<42:21,  3.53it/s]
 98%|█████████▊| 362499/371472 [6:41:09<41:58,  3.56it/s]
 98%|█████████▊| 362500/371472 [6:41:09<40:09,  3.72it/s]
                                                         
{'loss': 2.3987, 'learning_rate': 1.217490141998319e-07, 'epoch': 15.61}
+
 98%|█████████▊| 362500/371472 [6:41:09<40:09,  3.72it/s]
 98%|█████████▊| 362501/371472 [6:41:10<40:56,  3.65it/s]
 98%|█████████▊| 362502/371472 [6:41:10<41:40,  3.59it/s]
 98%|█████████▊| 362503/371472 [6:41:10<40:23,  3.70it/s]
 98%|█████████▊| 362504/371472 [6:41:10<38:59,  3.83it/s]
 98%|█████████▊| 362505/371472 [6:41:11<38:51,  3.85it/s]
 98%|█████████▊| 362506/371472 [6:41:11<42:58,  3.48it/s]
 98%|█████████▊| 362507/371472 [6:41:11<42:10,  3.54it/s]
 98%|█████████▊| 362508/371472 [6:41:12<41:50,  3.57it/s]
 98%|█████████▊| 362509/371472 [6:41:12<42:35,  3.51it/s]
 98%|█████████▊| 362510/371472 [6:41:12<41:42,  3.58it/s]
 98%|█████████▊| 362511/371472 [6:41:12<42:46,  3.49it/s]
 98%|█████████▊| 362512/371472 [6:41:13<41:18,  3.62it/s]
 98%|█████████▊| 362513/371472 [6:41:13<40:46,  3.66it/s]
 98%|█████████▊| 362514/371472 [6:41:13<40:13,  3.71it/s]
 98%|█████████▊| 362515/371472 [6:41:14<40:29,  3.69it/s]
 98%|█████████▊| 362516/371472 [6:41:14<39:52,  3.74it/s]
 98%|█████████▊| 362517/371472 [6:41:14<42:19,  3.53it/s]
 98%|█████████▊| 362518/371472 [6:41:14<44:31,  3.35it/s]
 98%|█████████▊| 362519/371472 [6:41:15<43:54,  3.40it/s]
 98%|█████████▊| 362520/371472 [6:41:15<43:28,  3.43it/s]
                                                         
{'loss': 2.7462, 'learning_rate': 1.2170053222435307e-07, 'epoch': 15.61}
+
 98%|█████████▊| 362520/371472 [6:41:15<43:28,  3.43it/s]
 98%|█████████▊| 362521/371472 [6:41:15<43:23,  3.44it/s]
 98%|█████████▊| 362522/371472 [6:41:16<44:55,  3.32it/s]
 98%|█████████▊| 362523/371472 [6:41:16<44:15,  3.37it/s]
 98%|█████████▊| 362524/371472 [6:41:16<46:01,  3.24it/s]
 98%|█████████▊| 362525/371472 [6:41:17<48:12,  3.09it/s]
 98%|█████████▊| 362526/371472 [6:41:17<46:13,  3.23it/s]
 98%|█████████▊| 362527/371472 [6:41:17<43:55,  3.39it/s]
 98%|█████████▊| 362528/371472 [6:41:17<42:06,  3.54it/s]
 98%|█████████▊| 362529/371472 [6:41:18<41:02,  3.63it/s]
 98%|█████████▊| 362530/371472 [6:41:18<40:44,  3.66it/s]
 98%|█████████▊| 362531/371472 [6:41:18<42:03,  3.54it/s]
 98%|█████████▊| 362532/371472 [6:41:18<39:54,  3.73it/s]
 98%|█████████▊| 362533/371472 [6:41:19<39:47,  3.74it/s]
 98%|█████████▊| 362534/371472 [6:41:19<39:31,  3.77it/s]
 98%|█████████▊| 362535/371472 [6:41:19<40:22,  3.69it/s]
 98%|█████████▊| 362536/371472 [6:41:20<39:22,  3.78it/s]
 98%|█████████▊| 362537/371472 [6:41:20<41:32,  3.58it/s]
 98%|█████████▊| 362538/371472 [6:41:20<41:16,  3.61it/s]
 98%|█████████▊| 362539/371472 [6:41:20<41:02,  3.63it/s]
 98%|█████████▊| 362540/371472 [6:41:21<42:24,  3.51it/s]
                                                         
{'loss': 2.4385, 'learning_rate': 1.2165205024887411e-07, 'epoch': 15.62}
+
 98%|█████████▊| 362540/371472 [6:41:21<42:24,  3.51it/s]
 98%|█████████▊| 362541/371472 [6:41:21<41:41,  3.57it/s]
 98%|█████████▊| 362542/371472 [6:41:21<45:22,  3.28it/s]
 98%|█████████▊| 362543/371472 [6:41:22<43:44,  3.40it/s]
 98%|█████████▊| 362544/371472 [6:41:22<41:59,  3.54it/s]
 98%|█████████▊| 362545/371472 [6:41:22<41:33,  3.58it/s]
 98%|█████████▊| 362546/371472 [6:41:22<41:39,  3.57it/s]
 98%|█████████▊| 362547/371472 [6:41:23<45:37,  3.26it/s]
 98%|█████████▊| 362548/371472 [6:41:23<45:36,  3.26it/s]
 98%|█████████▊| 362549/371472 [6:41:23<42:59,  3.46it/s]
 98%|█████████▊| 362550/371472 [6:41:24<43:09,  3.45it/s]
 98%|█████████▊| 362551/371472 [6:41:24<43:41,  3.40it/s]
 98%|█████████▊| 362552/371472 [6:41:24<44:34,  3.34it/s]
 98%|█████████▊| 362553/371472 [6:41:24<43:06,  3.45it/s]
 98%|█████████▊| 362554/371472 [6:41:25<42:53,  3.47it/s]
 98%|█████████▊| 362555/371472 [6:41:25<42:14,  3.52it/s]
 98%|█████████▊| 362556/371472 [6:41:25<42:16,  3.51it/s]
 98%|█████████▊| 362557/371472 [6:41:26<41:32,  3.58it/s]
 98%|█████████▊| 362558/371472 [6:41:26<42:47,  3.47it/s]
 98%|█████████▊| 362559/371472 [6:41:26<42:09,  3.52it/s]
 98%|█████████▊| 362560/371472 [6:41:26<41:58,  3.54it/s]
                                                         
{'loss': 2.497, 'learning_rate': 1.2160356827339527e-07, 'epoch': 15.62}
+
 98%|█████████▊| 362560/371472 [6:41:26<41:58,  3.54it/s]
 98%|██████��██▊| 362561/371472 [6:41:27<40:58,  3.62it/s]
 98%|█████████▊| 362562/371472 [6:41:27<39:32,  3.76it/s]
 98%|█████████▊| 362563/371472 [6:41:27<42:03,  3.53it/s]
 98%|█████████▊| 362564/371472 [6:41:28<42:36,  3.48it/s]
 98%|█████████▊| 362565/371472 [6:41:28<41:46,  3.55it/s]
 98%|█████████▊| 362566/371472 [6:41:28<42:26,  3.50it/s]
 98%|█████████▊| 362567/371472 [6:41:28<41:34,  3.57it/s]
 98%|█████████▊| 362568/371472 [6:41:29<45:51,  3.24it/s]
 98%|█████████▊| 362569/371472 [6:41:29<48:50,  3.04it/s]
 98%|█████████▊| 362570/371472 [6:41:29<47:25,  3.13it/s]
 98%|█████████▊| 362571/371472 [6:41:30<45:30,  3.26it/s]
 98%|█████████▊| 362572/371472 [6:41:30<45:00,  3.30it/s]
 98%|█████████▊| 362573/371472 [6:41:30<45:05,  3.29it/s]
 98%|█████████▊| 362574/371472 [6:41:31<43:19,  3.42it/s]
 98%|█████████▊| 362575/371472 [6:41:31<41:20,  3.59it/s]
 98%|█████████▊| 362576/371472 [6:41:31<42:13,  3.51it/s]
 98%|█████████▊| 362577/371472 [6:41:32<46:15,  3.20it/s]
 98%|█████████▊| 362578/371472 [6:41:32<44:44,  3.31it/s]
 98%|█████████▊| 362579/371472 [6:41:32<44:24,  3.34it/s]
 98%|█████████▊| 362580/371472 [6:41:32<42:35,  3.48it/s]
                                                         
{'loss': 2.5052, 'learning_rate': 1.2155508629791634e-07, 'epoch': 15.62}
+
 98%|█████████▊| 362580/371472 [6:41:32<42:35,  3.48it/s]
 98%|█████████▊| 362581/371472 [6:41:33<42:04,  3.52it/s]
 98%|█████████▊| 362582/371472 [6:41:33<41:02,  3.61it/s]
 98%|█████████▊| 362583/371472 [6:41:33<40:47,  3.63it/s]
 98%|█████████▊| 362584/371472 [6:41:33<39:44,  3.73it/s]
 98%|█████████▊| 362585/371472 [6:41:34<39:02,  3.79it/s]
 98%|█████████▊| 362586/371472 [6:41:34<39:02,  3.79it/s]
 98%|█████████▊| 362587/371472 [6:41:34<39:05,  3.79it/s]
 98%|█████████▊| 362588/371472 [6:41:34<40:11,  3.68it/s]
 98%|█████████▊| 362589/371472 [6:41:35<42:28,  3.49it/s]
 98%|█████████▊| 362590/371472 [6:41:35<41:42,  3.55it/s]
 98%|█████████▊| 362591/371472 [6:41:35<42:41,  3.47it/s]
 98%|█████████▊| 362592/371472 [6:41:36<40:47,  3.63it/s]
 98%|█████████▊| 362593/371472 [6:41:36<40:43,  3.63it/s]
 98%|█████████▊| 362594/371472 [6:41:36<41:33,  3.56it/s]
 98%|█████████▊| 362595/371472 [6:41:37<42:12,  3.50it/s]
 98%|█████████▊| 362596/371472 [6:41:37<41:07,  3.60it/s]
 98%|█████████▊| 362597/371472 [6:41:37<41:21,  3.58it/s]
 98%|█████████▊| 362598/371472 [6:41:37<40:45,  3.63it/s]
 98%|█████████▊| 362599/371472 [6:41:38<39:23,  3.75it/s]
 98%|█████████▊| 362600/371472 [6:41:38<38:54,  3.80it/s]
                                                         
{'loss': 2.5635, 'learning_rate': 1.215066043224375e-07, 'epoch': 15.62}
+
 98%|█████████▊| 362600/371472 [6:41:38<38:54,  3.80it/s]
 98%|█████████▊| 362601/371472 [6:41:38<38:41,  3.82it/s]
 98%|█████████▊| 362602/371472 [6:41:38<38:38,  3.83it/s]
 98%|█████████▊| 362603/371472 [6:41:39<38:48,  3.81it/s]
 98%|█████████▊| 362604/371472 [6:41:39<39:12,  3.77it/s]
 98%|█████████▊| 362605/371472 [6:41:39<39:20,  3.76it/s]
 98%|█████████▊| 362606/371472 [6:41:39<39:58,  3.70it/s]
 98%|█████████▊| 362607/371472 [6:41:40<38:48,  3.81it/s]
 98%|█████████▊| 362608/371472 [6:41:40<39:55,  3.70it/s]
 98%|█████████▊| 362609/371472 [6:41:40<39:57,  3.70it/s]
 98%|█████████▊| 362610/371472 [6:41:40<39:30,  3.74it/s]
 98%|█████████▊| 362611/371472 [6:41:41<39:11,  3.77it/s]
 98%|█████████▊| 362612/371472 [6:41:41<38:53,  3.80it/s]
 98%|█████████▊| 362613/371472 [6:41:41<38:44,  3.81it/s]
 98%|█████████▊| 362614/371472 [6:41:41<37:25,  3.94it/s]
 98%|█████████▊| 362615/371472 [6:41:42<38:32,  3.83it/s]
 98%|█████████▊| 362616/371472 [6:41:42<39:47,  3.71it/s]
 98%|█████████▊| 362617/371472 [6:41:42<38:50,  3.80it/s]
 98%|█████████▊| 362618/371472 [6:41:43<40:14,  3.67it/s]
 98%|█████████▊| 362619/371472 [6:41:43<43:22,  3.40it/s]
 98%|█████████▊| 362620/371472 [6:41:43<43:05,  3.42it/s]
                                                         
{'loss': 2.6826, 'learning_rate': 1.2145812234695856e-07, 'epoch': 15.62}
+
 98%|█████████▊| 362620/371472 [6:41:43<43:05,  3.42it/s]
 98%|█████████▊| 362621/371472 [6:41:44<42:51,  3.44it/s]
 98%|█████████▊| 362622/371472 [6:41:44<41:39,  3.54it/s]
 98%|█████████▊| 362623/371472 [6:41:44<39:38,  3.72it/s]
 98%|█████████▊| 362624/371472 [6:41:44<39:33,  3.73it/s]
 98%|█████████▊| 362625/371472 [6:41:45<42:02,  3.51it/s]
 98%|█████████▊| 362626/371472 [6:41:45<41:15,  3.57it/s]
 98%|█████████▊| 362627/371472 [6:41:45<41:50,  3.52it/s]
 98%|█████████▊| 362628/371472 [6:41:45<42:03,  3.50it/s]
 98%|█████████▊| 362629/371472 [6:41:46<43:17,  3.40it/s]
 98%|█████████▊| 362630/371472 [6:41:46<44:31,  3.31it/s]
 98%|█████████▊| 362631/371472 [6:41:46<43:25,  3.39it/s]
 98%|█████████▊| 362632/371472 [6:41:47<41:19,  3.57it/s]
 98%|█████████▊| 362633/371472 [6:41:47<41:59,  3.51it/s]
 98%|█████████▊| 362634/371472 [6:41:47<40:34,  3.63it/s]
 98%|█████████▊| 362635/371472 [6:41:47<39:47,  3.70it/s]
 98%|█████████▊| 362636/371472 [6:41:48<39:12,  3.76it/s]
 98%|█████████▊| 362637/371472 [6:41:48<39:02,  3.77it/s]
 98%|█████████▊| 362638/371472 [6:41:48<40:33,  3.63it/s]
 98%|█████████▊| 362639/371472 [6:41:49<41:53,  3.51it/s]
 98%|█████████▊| 362640/371472 [6:41:49<44:49,  3.28it/s]
                                                         
{'loss': 2.5188, 'learning_rate': 1.214096403714797e-07, 'epoch': 15.62}
+
 98%|█████████▊| 362640/371472 [6:41:49<44:49,  3.28it/s]
 98%|█████████▊| 362641/371472 [6:41:49<43:37,  3.37it/s]
 98%|█████████▊| 362642/371472 [6:41:49<41:36,  3.54it/s]
 98%|█████████▊| 362643/371472 [6:41:50<43:30,  3.38it/s]
 98%|█████████▊| 362644/371472 [6:41:50<42:00,  3.50it/s]
 98%|█████████▊| 362645/371472 [6:41:50<43:10,  3.41it/s]
 98%|█████████▊| 362646/371472 [6:41:51<41:02,  3.58it/s]
 98%|█████████▊| 362647/371472 [6:41:51<42:27,  3.46it/s]
 98%|█████████▊| 362648/371472 [6:41:51<44:46,  3.28it/s]
 98%|█████████▊| 362649/371472 [6:41:52<43:17,  3.40it/s]
 98%|█████████▊| 362650/371472 [6:41:52<44:26,  3.31it/s]
 98%|█████████▊| 362651/371472 [6:41:52<41:47,  3.52it/s]
 98%|█████████▊| 362652/371472 [6:41:52<41:50,  3.51it/s]
 98%|█████████▊| 362653/371472 [6:41:53<43:40,  3.37it/s]
 98%|█████████▊| 362654/371472 [6:41:53<43:34,  3.37it/s]
 98%|█████████▊| 362655/371472 [6:41:53<41:43,  3.52it/s]
 98%|█████████▊| 362656/371472 [6:41:53<40:33,  3.62it/s]
 98%|█████████▊| 362657/371472 [6:41:54<39:48,  3.69it/s]
 98%|█████████▊| 362658/371472 [6:41:54<39:28,  3.72it/s]
 98%|█████████▊| 362659/371472 [6:41:54<38:53,  3.78it/s]
 98%|█████████▊| 362660/371472 [6:41:55<38:25,  3.82it/s]
                                                         
{'loss': 2.6908, 'learning_rate': 1.2136115839600076e-07, 'epoch': 15.62}
+
 98%|█████████▊| 362660/371472 [6:41:55<38:25,  3.82it/s]
 98%|█████████▊| 362661/371472 [6:41:55<39:56,  3.68it/s]
 98%|█████████▊| 362662/371472 [6:41:55<42:29,  3.46it/s]
 98%|█████████▊| 362663/371472 [6:41:55<45:03,  3.26it/s]
 98%|█████████▊| 362664/371472 [6:41:56<46:49,  3.14it/s]
 98%|█████████▊| 362665/371472 [6:41:56<46:02,  3.19it/s]
 98%|█████████▊| 362666/371472 [6:41:56<44:16,  3.31it/s]
 98%|█████████▊| 362667/371472 [6:41:57<42:26,  3.46it/s]
 98%|█████████▊| 362668/371472 [6:41:57<42:04,  3.49it/s]
 98%|█████████▊| 362669/371472 [6:41:57<41:24,  3.54it/s]
 98%|█████████▊| 362670/371472 [6:41:57<40:06,  3.66it/s]
 98%|█████████▊| 362671/371472 [6:41:58<39:18,  3.73it/s]
 98%|█████████▊| 362672/371472 [6:41:58<39:45,  3.69it/s]
 98%|█████████▊| 362673/371472 [6:41:58<38:43,  3.79it/s]
 98%|█████████▊| 362674/371472 [6:41:59<40:07,  3.65it/s]
 98%|█████████▊| 362675/371472 [6:41:59<41:25,  3.54it/s]
 98%|█████████▊| 362676/371472 [6:41:59<40:45,  3.60it/s]
 98%|█████████▊| 362677/371472 [6:41:59<43:55,  3.34it/s]
 98%|█████████▊| 362678/371472 [6:42:00<42:47,  3.42it/s]
 98%|█████████▊| 362679/371472 [6:42:00<40:38,  3.61it/s]
 98%|█████████▊| 362680/371472 [6:42:00<42:29,  3.45it/s]
                                                         
{'loss': 2.5963, 'learning_rate': 1.2131267642052193e-07, 'epoch': 15.62}
+
 98%|█████████▊| 362680/371472 [6:42:00<42:29,  3.45it/s]
 98%|█████████▊| 362681/371472 [6:42:01<45:13,  3.24it/s]
 98%|█████████▊| 362682/371472 [6:42:01<42:56,  3.41it/s]
 98%|█████████▊| 362683/371472 [6:42:01<43:07,  3.40it/s]
 98%|█████████▊| 362684/371472 [6:42:01<41:56,  3.49it/s]
 98%|█████████▊| 362685/371472 [6:42:02<41:15,  3.55it/s]
 98%|█████████▊| 362686/371472 [6:42:02<41:09,  3.56it/s]
 98%|█████████▊| 362687/371472 [6:42:02<42:14,  3.47it/s]
 98%|█████████▊| 362688/371472 [6:42:03<46:59,  3.12it/s]
 98%|█████████▊| 362689/371472 [6:42:03<43:44,  3.35it/s]
 98%|█████████▊| 362690/371472 [6:42:03<44:08,  3.32it/s]
 98%|█████████▊| 362691/371472 [6:42:04<41:49,  3.50it/s]
 98%|█████████▊| 362692/371472 [6:42:04<41:42,  3.51it/s]
 98%|█████████▊| 362693/371472 [6:42:04<41:11,  3.55it/s]
 98%|█████████▊| 362694/371472 [6:42:04<39:19,  3.72it/s]
 98%|█████████▊| 362695/371472 [6:42:05<38:30,  3.80it/s]
 98%|█████████▊| 362696/371472 [6:42:05<38:10,  3.83it/s]
 98%|█████████▊| 362697/371472 [6:42:05<39:01,  3.75it/s]
 98%|█████████▊| 362698/371472 [6:42:05<38:39,  3.78it/s]
 98%|█████████▊| 362699/371472 [6:42:06<38:39,  3.78it/s]
 98%|█████████▊| 362700/371472 [6:42:06<39:13,  3.73it/s]
                                                         
{'loss': 2.5804, 'learning_rate': 1.2126419444504298e-07, 'epoch': 15.62}
+
 98%|█████████▊| 362700/371472 [6:42:06<39:13,  3.73it/s]
 98%|█████████▊| 362701/371472 [6:42:06<44:13,  3.31it/s]
 98%|█████████▊| 362702/371472 [6:42:07<43:42,  3.34it/s]
 98%|█████████▊| 362703/371472 [6:42:07<42:50,  3.41it/s]
 98%|█████████▊| 362704/371472 [6:42:07<42:34,  3.43it/s]
 98%|█████████▊| 362705/371472 [6:42:07<41:16,  3.54it/s]
 98%|█████████▊| 362706/371472 [6:42:08<41:06,  3.55it/s]
 98%|█████████▊| 362707/371472 [6:42:08<40:08,  3.64it/s]
 98%|█████████▊| 362708/371472 [6:42:08<39:12,  3.73it/s]
 98%|█████████▊| 362709/371472 [6:42:08<38:57,  3.75it/s]
 98%|█████████▊| 362710/371472 [6:42:09<39:14,  3.72it/s]
 98%|█████████▊| 362711/371472 [6:42:09<38:56,  3.75it/s]
 98%|█████████▊| 362712/371472 [6:42:09<39:33,  3.69it/s]
 98%|█████████▊| 362713/371472 [6:42:10<40:41,  3.59it/s]
 98%|█████████▊| 362714/371472 [6:42:10<40:18,  3.62it/s]
 98%|█████████▊| 362715/371472 [6:42:10<38:33,  3.79it/s]
 98%|█████████▊| 362716/371472 [6:42:10<38:49,  3.76it/s]
 98%|█████████▊| 362717/371472 [6:42:11<38:20,  3.81it/s]
 98%|█████████▊| 362718/371472 [6:42:11<37:59,  3.84it/s]
 98%|█████████▊| 362719/371472 [6:42:11<37:14,  3.92it/s]
 98%|█████████▊| 362720/371472 [6:42:11<41:18,  3.53it/s]
                                                         
{'loss': 2.6059, 'learning_rate': 1.2121571246956413e-07, 'epoch': 15.62}
+
 98%|█████████▊| 362720/371472 [6:42:11<41:18,  3.53it/s]
 98%|█████████▊| 362721/371472 [6:42:12<41:26,  3.52it/s]
 98%|█████████▊| 362722/371472 [6:42:12<42:05,  3.46it/s]
 98%|█████████▊| 362723/371472 [6:42:12<41:38,  3.50it/s]
 98%|█████████▊| 362724/371472 [6:42:13<40:34,  3.59it/s]
 98%|█████████▊| 362725/371472 [6:42:13<41:51,  3.48it/s]
 98%|█████████▊| 362726/371472 [6:42:13<41:04,  3.55it/s]
 98%|█████████▊| 362727/371472 [6:42:13<40:39,  3.58it/s]
 98%|█████████▊| 362728/371472 [6:42:14<40:13,  3.62it/s]
 98%|█████████▊| 362729/371472 [6:42:14<40:27,  3.60it/s]
 98%|█████████▊| 362730/371472 [6:42:14<41:22,  3.52it/s]
 98%|█████████▊| 362731/371472 [6:42:15<41:23,  3.52it/s]
 98%|█████████▊| 362732/371472 [6:42:15<39:56,  3.65it/s]
 98%|█████████▊| 362733/371472 [6:42:15<41:20,  3.52it/s]
 98%|█████████▊| 362734/371472 [6:42:15<41:29,  3.51it/s]
 98%|█████████▊| 362735/371472 [6:42:16<40:22,  3.61it/s]
 98%|█████████▊| 362736/371472 [6:42:16<42:28,  3.43it/s]
 98%|█████████▊| 362737/371472 [6:42:16<42:10,  3.45it/s]
 98%|█████████▊| 362738/371472 [6:42:17<42:32,  3.42it/s]
 98%|█████████▊| 362739/371472 [6:42:17<42:06,  3.46it/s]
 98%|█████████▊| 362740/371472 [6:42:17<41:19,  3.52it/s]
                                                         
{'loss': 2.4746, 'learning_rate': 1.211672304940852e-07, 'epoch': 15.62}
+
 98%|█████████▊| 362740/371472 [6:42:17<41:19,  3.52it/s]
 98%|█████████▊| 362741/371472 [6:42:17<40:34,  3.59it/s]
 98%|█████████▊| 362742/371472 [6:42:18<41:17,  3.52it/s]
 98%|█████████▊| 362743/371472 [6:42:18<40:35,  3.58it/s]
 98%|█████████▊| 362744/371472 [6:42:18<40:35,  3.58it/s]
 98%|█████████▊| 362745/371472 [6:42:19<40:38,  3.58it/s]
 98%|█████████▊| 362746/371472 [6:42:19<40:44,  3.57it/s]
 98%|█████████▊| 362747/371472 [6:42:19<42:34,  3.41it/s]
 98%|█████████▊| 362748/371472 [6:42:19<41:14,  3.53it/s]
 98%|█████████▊| 362749/371472 [6:42:20<41:18,  3.52it/s]
 98%|█████████▊| 362750/371472 [6:42:20<41:22,  3.51it/s]
 98%|█████████▊| 362751/371472 [6:42:20<43:12,  3.36it/s]
 98%|█████████▊| 362752/371472 [6:42:21<42:38,  3.41it/s]
 98%|█████████▊| 362753/371472 [6:42:21<40:43,  3.57it/s]
 98%|█████████▊| 362754/371472 [6:42:21<40:04,  3.63it/s]
 98%|█████████▊| 362755/371472 [6:42:21<43:41,  3.32it/s]
 98%|█████████▊| 362756/371472 [6:42:22<43:02,  3.37it/s]
 98%|█████████▊| 362757/371472 [6:42:22<42:07,  3.45it/s]
 98%|█████████▊| 362758/371472 [6:42:22<42:50,  3.39it/s]
 98%|█████████▊| 362759/371472 [6:42:23<40:57,  3.55it/s]
 98%|█████████▊| 362760/371472 [6:42:23<41:49,  3.47it/s]
                                                         
{'loss': 2.4865, 'learning_rate': 1.2111874851860625e-07, 'epoch': 15.62}
+
 98%|█████████▊| 362760/371472 [6:42:23<41:49,  3.47it/s]
 98%|█████████▊| 362761/371472 [6:42:23<47:16,  3.07it/s]
 98%|█████████▊| 362762/371472 [6:42:24<47:04,  3.08it/s]
 98%|█████████▊| 362763/371472 [6:42:24<47:29,  3.06it/s]
 98%|█████████▊| 362764/371472 [6:42:24<44:50,  3.24it/s]
 98%|█████████▊| 362765/371472 [6:42:25<44:47,  3.24it/s]
 98%|█████████▊| 362766/371472 [6:42:25<42:32,  3.41it/s]
 98%|█████████▊| 362767/371472 [6:42:25<42:47,  3.39it/s]
 98%|█████████▊| 362768/371472 [6:42:25<42:07,  3.44it/s]
 98%|█████████▊| 362769/371472 [6:42:26<40:28,  3.58it/s]
 98%|█████████▊| 362770/371472 [6:42:26<38:54,  3.73it/s]
 98%|█████████▊| 362771/371472 [6:42:26<38:21,  3.78it/s]
 98%|█████████▊| 362772/371472 [6:42:26<38:20,  3.78it/s]
 98%|█████████▊| 362773/371472 [6:42:27<40:29,  3.58it/s]
 98%|█████████▊| 362774/371472 [6:42:27<41:59,  3.45it/s]
 98%|█████████▊| 362775/371472 [6:42:27<40:24,  3.59it/s]
 98%|█████████▊| 362776/371472 [6:42:28<40:09,  3.61it/s]
 98%|█████████▊| 362777/371472 [6:42:28<41:18,  3.51it/s]
 98%|█████████▊| 362778/371472 [6:42:28<40:09,  3.61it/s]
 98%|█████████▊| 362779/371472 [6:42:28<39:17,  3.69it/s]
 98%|█████████▊| 362780/371472 [6:42:29<39:08,  3.70it/s]
                                                         
{'loss': 2.634, 'learning_rate': 1.210702665431274e-07, 'epoch': 15.63}
+
 98%|█████████▊| 362780/371472 [6:42:29<39:08,  3.70it/s]
 98%|█████████▊| 362781/371472 [6:42:29<42:00,  3.45it/s]
 98%|█████████▊| 362782/371472 [6:42:29<39:57,  3.63it/s]
 98%|█████████▊| 362783/371472 [6:42:29<39:23,  3.68it/s]
 98%|█████████▊| 362784/371472 [6:42:30<40:09,  3.61it/s]
 98%|█████████▊| 362785/371472 [6:42:30<40:12,  3.60it/s]
 98%|█████████▊| 362786/371472 [6:42:30<40:39,  3.56it/s]
 98%|█████████▊| 362787/371472 [6:42:31<39:55,  3.63it/s]
 98%|█████████▊| 362788/371472 [6:42:31<38:55,  3.72it/s]
 98%|█████████▊| 362789/371472 [6:42:31<38:26,  3.76it/s]
 98%|█████████▊| 362790/371472 [6:42:31<39:55,  3.62it/s]
 98%|█████████▊| 362791/371472 [6:42:32<38:11,  3.79it/s]
 98%|█████████▊| 362792/371472 [6:42:32<40:24,  3.58it/s]
 98%|█████████▊| 362793/371472 [6:42:32<40:26,  3.58it/s]
 98%|█████████▊| 362794/371472 [6:42:32<39:55,  3.62it/s]
 98%|█████████▊| 362795/371472 [6:42:33<40:39,  3.56it/s]
 98%|█████████▊| 362796/371472 [6:42:33<39:53,  3.63it/s]
 98%|█████████▊| 362797/371472 [6:42:33<39:45,  3.64it/s]
 98%|█████████▊| 362798/371472 [6:42:34<41:19,  3.50it/s]
 98%|█████████▊| 362799/371472 [6:42:34<39:44,  3.64it/s]
 98%|█████████▊| 362800/371472 [6:42:34<43:00,  3.36it/s]
                                                         
{'loss': 2.6028, 'learning_rate': 1.2102178456764847e-07, 'epoch': 15.63}
+
 98%|█████████▊| 362800/371472 [6:42:34<43:00,  3.36it/s]
 98%|█████████▊| 362801/371472 [6:42:35<43:06,  3.35it/s]
 98%|█████████▊| 362802/371472 [6:42:35<41:41,  3.47it/s]
 98%|█████████▊| 362803/371472 [6:42:35<41:46,  3.46it/s]
 98%|█████████▊| 362804/371472 [6:42:35<41:17,  3.50it/s]
 98%|█████████▊| 362805/371472 [6:42:36<40:32,  3.56it/s]
 98%|█████████▊| 362806/371472 [6:42:36<40:47,  3.54it/s]
 98%|█████████▊| 362807/371472 [6:42:36<43:28,  3.32it/s]
 98%|█████████▊| 362808/371472 [6:42:37<42:14,  3.42it/s]
 98%|█████████▊| 362809/371472 [6:42:37<41:27,  3.48it/s]
 98%|█████████▊| 362810/371472 [6:42:37<39:48,  3.63it/s]
 98%|█████████▊| 362811/371472 [6:42:37<40:46,  3.54it/s]
 98%|█████████▊| 362812/371472 [6:42:38<43:31,  3.32it/s]
 98%|█████████▊| 362813/371472 [6:42:38<43:19,  3.33it/s]
 98%|█████████▊| 362814/371472 [6:42:38<43:43,  3.30it/s]
 98%|█████████▊| 362815/371472 [6:42:39<42:23,  3.40it/s]
 98%|█████████▊| 362816/371472 [6:42:39<41:30,  3.48it/s]
 98%|█████████▊| 362817/371472 [6:42:39<41:26,  3.48it/s]
 98%|█████████▊| 362818/371472 [6:42:40<45:45,  3.15it/s]
 98%|█████████▊| 362819/371472 [6:42:40<43:47,  3.29it/s]
 98%|█████████▊| 362820/371472 [6:42:40<42:18,  3.41it/s]
                                                         
{'loss': 2.55, 'learning_rate': 1.2097330259216962e-07, 'epoch': 15.63}
+
 98%|█████████▊| 362820/371472 [6:42:40<42:18,  3.41it/s]
 98%|█████████▊| 362821/371472 [6:42:40<40:18,  3.58it/s]
 98%|█████████▊| 362822/371472 [6:42:41<46:18,  3.11it/s]
 98%|█████████▊| 362823/371472 [6:42:41<43:31,  3.31it/s]
 98%|█████████▊| 362824/371472 [6:42:41<41:46,  3.45it/s]
 98%|█████████▊| 362825/371472 [6:42:42<42:24,  3.40it/s]
 98%|█████████▊| 362826/371472 [6:42:42<57:38,  2.50it/s]
 98%|█████████▊| 362827/371472 [6:42:43<52:40,  2.74it/s]
 98%|█████████▊| 362828/371472 [6:42:43<48:04,  3.00it/s]
 98%|█████████▊| 362829/371472 [6:42:43<44:58,  3.20it/s]
 98%|█████████▊| 362830/371472 [6:42:43<43:35,  3.30it/s]
 98%|█████████▊| 362831/371472 [6:42:44<42:00,  3.43it/s]
 98%|█████████▊| 362832/371472 [6:42:44<40:03,  3.59it/s]
 98%|█████████▊| 362833/371472 [6:42:44<42:56,  3.35it/s]
 98%|█████████▊| 362834/371472 [6:42:44<42:07,  3.42it/s]
 98%|█████████▊| 362835/371472 [6:42:45<40:08,  3.59it/s]
 98%|█████████▊| 362836/371472 [6:42:45<41:02,  3.51it/s]
 98%|█████████▊| 362837/371472 [6:42:45<42:31,  3.38it/s]
 98%|█████████▊| 362838/371472 [6:42:46<41:00,  3.51it/s]
 98%|█████████▊| 362839/371472 [6:42:46<39:11,  3.67it/s]
 98%|█████████▊| 362840/371472 [6:42:46<38:38,  3.72it/s]
                                                         
{'loss': 2.6876, 'learning_rate': 1.209248206166907e-07, 'epoch': 15.63}
+
 98%|█████████▊| 362840/371472 [6:42:46<38:38,  3.72it/s]
 98%|█████████▊| 362841/371472 [6:42:46<40:34,  3.54it/s]
 98%|█████████▊| 362842/371472 [6:42:47<40:44,  3.53it/s]
 98%|█████████▊| 362843/371472 [6:42:47<41:27,  3.47it/s]
 98%|█████████▊| 362844/371472 [6:42:47<40:43,  3.53it/s]
 98%|█████████▊| 362845/371472 [6:42:48<40:30,  3.55it/s]
 98%|█████████▊| 362846/371472 [6:42:48<41:34,  3.46it/s]
 98%|█████████▊| 362847/371472 [6:42:48<40:17,  3.57it/s]
 98%|█████████▊| 362848/371472 [6:42:48<41:50,  3.44it/s]
 98%|█████████▊| 362849/371472 [6:42:49<41:08,  3.49it/s]
 98%|█████████▊| 362850/371472 [6:42:49<39:07,  3.67it/s]
 98%|█████████▊| 362851/371472 [6:42:49<39:55,  3.60it/s]
 98%|█████████▊| 362852/371472 [6:42:50<40:56,  3.51it/s]
 98%|█████████▊| 362853/371472 [6:42:50<40:33,  3.54it/s]
 98%|█████████▊| 362854/371472 [6:42:50<43:54,  3.27it/s]
 98%|█████████▊| 362855/371472 [6:42:50<42:25,  3.39it/s]
 98%|█████████▊| 362856/371472 [6:42:51<42:19,  3.39it/s]
 98%|█████████▊| 362857/371472 [6:42:51<40:23,  3.55it/s]
 98%|█████████▊| 362858/371472 [6:42:51<38:40,  3.71it/s]
 98%|█████████▊| 362859/371472 [6:42:51<38:10,  3.76it/s]
 98%|█████████▊| 362860/371472 [6:42:52<36:50,  3.90it/s]
                                                         
{'loss': 2.6096, 'learning_rate': 1.2087633864121184e-07, 'epoch': 15.63}
+
 98%|█████████▊| 362860/371472 [6:42:52<36:50,  3.90it/s]
 98%|█████████▊| 362861/371472 [6:42:52<38:35,  3.72it/s]
 98%|█████████▊| 362862/371472 [6:42:52<37:42,  3.81it/s]
 98%|█████████▊| 362863/371472 [6:42:53<40:51,  3.51it/s]
 98%|█████████▊| 362864/371472 [6:42:53<39:31,  3.63it/s]
 98%|█████████▊| 362865/371472 [6:42:53<39:47,  3.60it/s]
 98%|█████████▊| 362866/371472 [6:42:53<39:29,  3.63it/s]
 98%|█████████▊| 362867/371472 [6:42:54<42:11,  3.40it/s]
 98%|█████████▊| 362868/371472 [6:42:54<41:39,  3.44it/s]
 98%|█████████▊| 362869/371472 [6:42:54<42:54,  3.34it/s]
 98%|█████████▊| 362870/371472 [6:42:55<42:47,  3.35it/s]
 98%|█████████▊| 362871/371472 [6:42:55<41:57,  3.42it/s]
 98%|█████████▊| 362872/371472 [6:42:55<43:14,  3.31it/s]
 98%|█████████▊| 362873/371472 [6:42:55<41:22,  3.46it/s]
 98%|█████████▊| 362874/371472 [6:42:56<40:28,  3.54it/s]
 98%|█████████▊| 362875/371472 [6:42:56<39:00,  3.67it/s]
 98%|█████████▊| 362876/371472 [6:42:56<40:36,  3.53it/s]
 98%|█████████▊| 362877/371472 [6:42:57<41:06,  3.48it/s]
 98%|█████████▊| 362878/371472 [6:42:57<42:49,  3.34it/s]
 98%|█████████▊| 362879/371472 [6:42:57<41:22,  3.46it/s]
 98%|█████████▊| 362880/371472 [6:42:57<39:21,  3.64it/s]
                                                         
{'loss': 2.4786, 'learning_rate': 1.2082785666573289e-07, 'epoch': 15.63}
+
 98%|█████████▊| 362880/371472 [6:42:57<39:21,  3.64it/s]
 98%|█████████▊| 362881/371472 [6:42:58<38:44,  3.70it/s]
 98%|█████████▊| 362882/371472 [6:42:58<41:22,  3.46it/s]
 98%|█████████▊| 362883/371472 [6:42:58<42:15,  3.39it/s]
 98%|█████████▊| 362884/371472 [6:42:59<40:37,  3.52it/s]
 98%|█████████▊| 362885/371472 [6:42:59<38:53,  3.68it/s]
 98%|█████████▊| 362886/371472 [6:42:59<38:18,  3.74it/s]
 98%|█████████▊| 362887/371472 [6:43:00<45:39,  3.13it/s]
 98%|█████████▊| 362888/371472 [6:43:00<44:00,  3.25it/s]
 98%|█████████▊| 362889/371472 [6:43:00<48:33,  2.95it/s]
 98%|█████████▊| 362890/371472 [6:43:01<47:26,  3.01it/s]
 98%|█████████▊| 362891/371472 [6:43:01<44:22,  3.22it/s]
 98%|█████████▊| 362892/371472 [6:43:01<44:44,  3.20it/s]
 98%|█████████▊| 362893/371472 [6:43:01<41:51,  3.42it/s]
 98%|█████████▊| 362894/371472 [6:43:02<40:34,  3.52it/s]
 98%|█████████▊| 362895/371472 [6:43:02<41:31,  3.44it/s]
 98%|█████████▊| 362896/371472 [6:43:02<41:47,  3.42it/s]
 98%|█████████▊| 362897/371472 [6:43:03<41:16,  3.46it/s]
 98%|█████████▊| 362898/371472 [6:43:03<43:10,  3.31it/s]
 98%|█████████▊| 362899/371472 [6:43:03<41:10,  3.47it/s]
 98%|█████████▊| 362900/371472 [6:43:03<42:13,  3.38it/s]
                                                         
{'loss': 2.71, 'learning_rate': 1.2077937469025406e-07, 'epoch': 15.63}
+
 98%|█████████▊| 362900/371472 [6:43:03<42:13,  3.38it/s]
 98%|█████████▊| 362901/371472 [6:43:04<40:46,  3.50it/s]
 98%|█████████▊| 362902/371472 [6:43:04<40:53,  3.49it/s]
 98%|█████████▊| 362903/371472 [6:43:04<40:31,  3.52it/s]
 98%|█████████▊| 362904/371472 [6:43:05<39:45,  3.59it/s]
 98%|█████████▊| 362905/371472 [6:43:05<39:12,  3.64it/s]
 98%|█████████▊| 362906/371472 [6:43:05<39:33,  3.61it/s]
 98%|█████████▊| 362907/371472 [6:43:05<39:09,  3.64it/s]
 98%|█████████▊| 362908/371472 [6:43:06<39:13,  3.64it/s]
 98%|��████████▊| 362909/371472 [6:43:06<39:21,  3.63it/s]
 98%|█████████▊| 362910/371472 [6:43:06<40:46,  3.50it/s]
 98%|█████████▊| 362911/371472 [6:43:06<40:38,  3.51it/s]
 98%|█████████▊| 362912/371472 [6:43:07<38:51,  3.67it/s]
 98%|█████████▊| 362913/371472 [6:43:07<38:47,  3.68it/s]
 98%|█████████▊| 362914/371472 [6:43:07<37:59,  3.75it/s]
 98%|█████████▊| 362915/371472 [6:43:07<37:37,  3.79it/s]
 98%|█████████▊| 362916/371472 [6:43:08<38:39,  3.69it/s]
 98%|█████████▊| 362917/371472 [6:43:08<40:21,  3.53it/s]
 98%|█████████▊| 362918/371472 [6:43:08<38:34,  3.70it/s]
 98%|█████████▊| 362919/371472 [6:43:09<38:26,  3.71it/s]
 98%|█████████▊| 362920/371472 [6:43:09<39:43,  3.59it/s]
                                                         
{'loss': 2.3971, 'learning_rate': 1.207308927147751e-07, 'epoch': 15.63}
+
 98%|█████████▊| 362920/371472 [6:43:09<39:43,  3.59it/s]
 98%|█████████▊| 362921/371472 [6:43:09<40:45,  3.50it/s]
 98%|█████████▊| 362922/371472 [6:43:09<39:17,  3.63it/s]
 98%|█████████▊| 362923/371472 [6:43:10<41:50,  3.40it/s]
 98%|█████████▊| 362924/371472 [6:43:10<40:58,  3.48it/s]
 98%|█████████▊| 362925/371472 [6:43:10<40:04,  3.56it/s]
 98%|█████████▊| 362926/371472 [6:43:11<40:01,  3.56it/s]
 98%|█████████▊| 362927/371472 [6:43:11<38:31,  3.70it/s]
 98%|█████████▊| 362928/371472 [6:43:11<39:28,  3.61it/s]
 98%|█████████▊| 362929/371472 [6:43:11<38:01,  3.74it/s]
 98%|█████████▊| 362930/371472 [6:43:12<40:36,  3.51it/s]
 98%|█████████▊| 362931/371472 [6:43:12<41:01,  3.47it/s]
 98%|█████████▊| 362932/371472 [6:43:12<38:53,  3.66it/s]
 98%|█████████▊| 362933/371472 [6:43:13<39:45,  3.58it/s]
 98%|█████████▊| 362934/371472 [6:43:13<38:18,  3.71it/s]
 98%|█████████▊| 362935/371472 [6:43:13<39:33,  3.60it/s]
 98%|█████████▊| 362936/371472 [6:43:13<40:04,  3.55it/s]
 98%|█████████▊| 362937/371472 [6:43:14<41:16,  3.45it/s]
 98%|█████████▊| 362938/371472 [6:43:14<40:54,  3.48it/s]
 98%|█████████▊| 362939/371472 [6:43:14<41:12,  3.45it/s]
 98%|█████████▊| 362940/371472 [6:43:15<43:13,  3.29it/s]
                                                         
{'loss': 2.6482, 'learning_rate': 1.2068241073929626e-07, 'epoch': 15.63}
+
 98%|█████████▊| 362940/371472 [6:43:15<43:13,  3.29it/s]
 98%|█████████▊| 362941/371472 [6:43:15<40:48,  3.48it/s]
 98%|█████████▊| 362942/371472 [6:43:15<39:06,  3.64it/s]
 98%|█████████▊| 362943/371472 [6:43:15<38:56,  3.65it/s]
 98%|█████████▊| 362944/371472 [6:43:16<37:59,  3.74it/s]
 98%|█████████▊| 362945/371472 [6:43:16<37:34,  3.78it/s]
 98%|█████████▊| 362946/371472 [6:43:16<37:50,  3.76it/s]
 98%|█████████▊| 362947/371472 [6:43:16<39:20,  3.61it/s]
 98%|█████████▊| 362948/371472 [6:43:17<43:03,  3.30it/s]
 98%|█████████▊| 362949/371472 [6:43:17<40:57,  3.47it/s]
 98%|█████████▊| 362950/371472 [6:43:17<38:52,  3.65it/s]
 98%|█████████▊| 362951/371472 [6:43:18<37:36,  3.78it/s]
 98%|█████████▊| 362952/371472 [6:43:18<36:55,  3.85it/s]
 98%|█████████▊| 362953/371472 [6:43:18<38:25,  3.70it/s]
 98%|█████████▊| 362954/371472 [6:43:18<40:35,  3.50it/s]
 98%|█████████▊| 362955/371472 [6:43:19<40:23,  3.51it/s]
 98%|█████████▊| 362956/371472 [6:43:19<42:48,  3.32it/s]
 98%|█████████▊| 362957/371472 [6:43:19<41:45,  3.40it/s]
 98%|█████████▊| 362958/371472 [6:43:20<39:46,  3.57it/s]
 98%|█████████▊| 362959/371472 [6:43:20<39:50,  3.56it/s]
 98%|█████████▊| 362960/371472 [6:43:20<39:19,  3.61it/s]
                                                         
{'loss': 2.6494, 'learning_rate': 1.2063392876381733e-07, 'epoch': 15.63}
+
 98%|█████████▊| 362960/371472 [6:43:20<39:19,  3.61it/s]
 98%|█████████▊| 362961/371472 [6:43:20<37:34,  3.77it/s]
 98%|█████████▊| 362962/371472 [6:43:21<38:22,  3.70it/s]
 98%|█████████▊| 362963/371472 [6:43:21<38:47,  3.66it/s]
 98%|█████████▊| 362964/371472 [6:43:21<39:48,  3.56it/s]
 98%|█████████▊| 362965/371472 [6:43:21<38:47,  3.65it/s]
 98%|█████████▊| 362966/371472 [6:43:22<39:36,  3.58it/s]
 98%|█████████▊| 362967/371472 [6:43:22<43:53,  3.23it/s]
 98%|█████████▊| 362968/371472 [6:43:22<42:10,  3.36it/s]
 98%|█████████▊| 362969/371472 [6:43:23<40:34,  3.49it/s]
 98%|█████████▊| 362970/371472 [6:43:23<39:14,  3.61it/s]
 98%|█████████▊| 362971/371472 [6:43:23<38:47,  3.65it/s]
 98%|█████████▊| 362972/371472 [6:43:23<38:28,  3.68it/s]
 98%|█████████▊| 362973/371472 [6:43:24<38:14,  3.70it/s]
 98%|█████████▊| 362974/371472 [6:43:24<40:10,  3.53it/s]
 98%|█████████▊| 362975/371472 [6:43:24<41:10,  3.44it/s]
 98%|█████████▊| 362976/371472 [6:43:25<40:17,  3.51it/s]
 98%|█████████▊| 362977/371472 [6:43:25<39:07,  3.62it/s]
 98%|█████████▊| 362978/371472 [6:43:25<39:36,  3.57it/s]
 98%|█████████▊| 362979/371472 [6:43:25<38:48,  3.65it/s]
 98%|█████████▊| 362980/371472 [6:43:26<40:42,  3.48it/s]
                                                         
{'loss': 2.6278, 'learning_rate': 1.2058544678833848e-07, 'epoch': 15.63}
+
 98%|█████████▊| 362980/371472 [6:43:26<40:42,  3.48it/s]
 98%|█████████▊| 362981/371472 [6:43:26<43:11,  3.28it/s]
 98%|█████████▊| 362982/371472 [6:43:26<45:49,  3.09it/s]
 98%|█████████▊| 362983/371472 [6:43:27<43:11,  3.28it/s]
 98%|█████████▊| 362984/371472 [6:43:27<40:51,  3.46it/s]
 98%|█████████▊| 362985/371472 [6:43:27<43:32,  3.25it/s]
 98%|█████████▊| 362986/371472 [6:43:28<43:32,  3.25it/s]
 98%|█████████▊| 362987/371472 [6:43:28<40:48,  3.46it/s]
 98%|█████████▊| 362988/371472 [6:43:28<42:28,  3.33it/s]
 98%|█████████▊| 362989/371472 [6:43:29<43:28,  3.25it/s]
 98%|█████████▊| 362990/371472 [6:43:29<44:47,  3.16it/s]
 98%|█████████▊| 362991/371472 [6:43:29<44:18,  3.19it/s]
 98%|█████████▊| 362992/371472 [6:43:30<45:12,  3.13it/s]
 98%|█████████▊| 362993/371472 [6:43:30<43:29,  3.25it/s]
 98%|█████████▊| 362994/371472 [6:43:30<41:13,  3.43it/s]
 98%|█████████▊| 362995/371472 [6:43:30<40:14,  3.51it/s]
 98%|█████████▊| 362996/371472 [6:43:31<40:32,  3.48it/s]
 98%|█████████▊| 362997/371472 [6:43:31<40:03,  3.53it/s]
 98%|█████████▊| 362998/371472 [6:43:31<42:18,  3.34it/s]
 98%|█████████▊| 362999/371472 [6:43:31<40:08,  3.52it/s]
 98%|█████████▊| 363000/371472 [6:43:32<39:01,  3.62it/s]
                                                         
{'loss': 2.5964, 'learning_rate': 1.2053696481285955e-07, 'epoch': 15.64}
+
 98%|█████████▊| 363000/371472 [6:43:32<39:01,  3.62it/s]
 98%|█████████▊| 363001/371472 [6:43:32<40:30,  3.49it/s]
 98%|█████████▊| 363002/371472 [6:43:32<40:57,  3.45it/s]
 98%|█████████▊| 363003/371472 [6:43:33<40:26,  3.49it/s]
 98%|█████████▊| 363004/371472 [6:43:33<40:29,  3.49it/s]
 98%|█████████▊| 363005/371472 [6:43:33<39:30,  3.57it/s]
 98%|█████████▊| 363006/371472 [6:43:33<38:28,  3.67it/s]
 98%|█████████▊| 363007/371472 [6:43:34<39:01,  3.62it/s]
 98%|█████████▊| 363008/371472 [6:43:34<37:40,  3.75it/s]
 98%|█████████▊| 363009/371472 [6:43:34<39:49,  3.54it/s]
 98%|█████████▊| 363010/371472 [6:43:35<37:49,  3.73it/s]
 98%|█████████▊| 363011/371472 [6:43:35<38:33,  3.66it/s]
 98%|█████████▊| 363012/371472 [6:43:35<38:50,  3.63it/s]
 98%|█████████▊| 363013/371472 [6:43:35<38:17,  3.68it/s]
 98%|█████████▊| 363014/371472 [6:43:36<39:20,  3.58it/s]
 98%|█████████▊| 363015/371472 [6:43:36<37:49,  3.73it/s]
 98%|█████████▊| 363016/371472 [6:43:36<38:02,  3.71it/s]
 98%|█████████▊| 363017/371472 [6:43:36<39:08,  3.60it/s]
 98%|█████████▊| 363018/371472 [6:43:37<39:54,  3.53it/s]
 98%|█████████▊| 363019/371472 [6:43:37<39:15,  3.59it/s]
 98%|█████████▊| 363020/371472 [6:43:37<39:26,  3.57it/s]
                                                         
{'loss': 2.5642, 'learning_rate': 1.204884828373807e-07, 'epoch': 15.64}
+
 98%|█████████▊| 363020/371472 [6:43:37<39:26,  3.57it/s]
 98%|█████████▊| 363021/371472 [6:43:38<39:37,  3.56it/s]
 98%|█████████▊| 363022/371472 [6:43:38<39:03,  3.61it/s]
 98%|█████████▊| 363023/371472 [6:43:38<39:00,  3.61it/s]
 98%|��████████▊| 363024/371472 [6:43:38<39:29,  3.57it/s]
 98%|█████████▊| 363025/371472 [6:43:39<38:35,  3.65it/s]
 98%|█████████▊| 363026/371472 [6:43:39<39:07,  3.60it/s]
 98%|█████████▊| 363027/371472 [6:43:39<38:57,  3.61it/s]
 98%|█████████▊| 363028/371472 [6:43:39<38:41,  3.64it/s]
 98%|█████████▊| 363029/371472 [6:43:40<37:24,  3.76it/s]
 98%|█████████▊| 363030/371472 [6:43:40<39:04,  3.60it/s]
 98%|█████████▊| 363031/371472 [6:43:40<38:56,  3.61it/s]
 98%|█████████▊| 363032/371472 [6:43:41<38:26,  3.66it/s]
 98%|█████████▊| 363033/371472 [6:43:41<43:38,  3.22it/s]
 98%|█████████▊| 363034/371472 [6:43:41<42:59,  3.27it/s]
 98%|█████████▊| 363035/371472 [6:43:42<42:13,  3.33it/s]
 98%|█████████▊| 363036/371472 [6:43:42<39:29,  3.56it/s]
 98%|█████████▊| 363037/371472 [6:43:42<37:56,  3.70it/s]
 98%|█████████▊| 363038/371472 [6:43:42<40:40,  3.46it/s]
 98%|█████████▊| 363039/371472 [6:43:43<40:15,  3.49it/s]
 98%|█████████▊| 363040/371472 [6:43:43<40:19,  3.49it/s]
                                                         
{'loss': 2.4696, 'learning_rate': 1.2044000086190175e-07, 'epoch': 15.64}
+
 98%|█████████▊| 363040/371472 [6:43:43<40:19,  3.49it/s]
 98%|█████████▊| 363041/371472 [6:43:43<41:44,  3.37it/s]
 98%|█████████▊| 363042/371472 [6:43:44<40:27,  3.47it/s]
 98%|█████████▊| 363043/371472 [6:43:44<41:08,  3.41it/s]
 98%|█████████▊| 363044/371472 [6:43:44<39:34,  3.55it/s]
 98%|█████████▊| 363045/371472 [6:43:44<38:45,  3.62it/s]
 98%|█████████▊| 363046/371472 [6:43:45<38:10,  3.68it/s]
 98%|█████████▊| 363047/371472 [6:43:45<37:50,  3.71it/s]
 98%|█████████▊| 363048/371472 [6:43:45<37:55,  3.70it/s]
 98%|█████████▊| 363049/371472 [6:43:45<37:54,  3.70it/s]
 98%|█████████▊| 363050/371472 [6:43:46<37:04,  3.79it/s]
 98%|█████████▊| 363051/371472 [6:43:46<38:26,  3.65it/s]
 98%|█████████▊| 363052/371472 [6:43:46<44:09,  3.18it/s]
 98%|█████████▊| 363053/371472 [6:43:47<42:25,  3.31it/s]
 98%|█████████▊| 363054/371472 [6:43:47<41:32,  3.38it/s]
 98%|█████████▊| 363055/371472 [6:43:47<41:51,  3.35it/s]
 98%|█████████▊| 363056/371472 [6:43:48<40:21,  3.48it/s]
 98%|█████████▊| 363057/371472 [6:43:48<39:04,  3.59it/s]
 98%|█████████▊| 363058/371472 [6:43:48<39:58,  3.51it/s]
 98%|█████████▊| 363059/371472 [6:43:48<38:44,  3.62it/s]
 98%|█████████▊| 363060/371472 [6:43:49<38:09,  3.67it/s]
                                                         
{'loss': 2.7335, 'learning_rate': 1.2039151888642293e-07, 'epoch': 15.64}
+
 98%|█████████▊| 363060/371472 [6:43:49<38:09,  3.67it/s]
 98%|█████████▊| 363061/371472 [6:43:49<38:48,  3.61it/s]
 98%|█████████▊| 363062/371472 [6:43:49<38:49,  3.61it/s]
 98%|█████████▊| 363063/371472 [6:43:49<40:42,  3.44it/s]
 98%|█████████▊| 363064/371472 [6:43:50<40:31,  3.46it/s]
 98%|█████████▊| 363065/371472 [6:43:50<39:58,  3.50it/s]
 98%|█████████▊| 363066/371472 [6:43:50<37:48,  3.71it/s]
 98%|█████████▊| 363067/371472 [6:43:51<38:07,  3.67it/s]
 98%|█████████▊| 363068/371472 [6:43:51<40:27,  3.46it/s]
 98%|█████████▊| 363069/371472 [6:43:51<41:00,  3.42it/s]
 98%|█████████▊| 363070/371472 [6:43:51<38:56,  3.60it/s]
 98%|█████████▊| 363071/371472 [6:43:52<37:42,  3.71it/s]
 98%|█████████▊| 363072/371472 [6:43:52<38:57,  3.59it/s]
 98%|█████████▊| 363073/371472 [6:43:52<43:28,  3.22it/s]
 98%|█████████▊| 363074/371472 [6:43:53<41:20,  3.39it/s]
 98%|█████████▊| 363075/371472 [6:43:53<39:42,  3.52it/s]
 98%|█████████▊| 363076/371472 [6:43:53<41:48,  3.35it/s]
 98%|█████████▊| 363077/371472 [6:43:53<40:17,  3.47it/s]
 98%|█████████▊| 363078/371472 [6:43:54<41:17,  3.39it/s]
 98%|█████████▊| 363079/371472 [6:43:54<39:19,  3.56it/s]
 98%|█████████▊| 363080/371472 [6:43:54<38:59,  3.59it/s]
                                                         
{'loss': 2.4981, 'learning_rate': 1.2034303691094397e-07, 'epoch': 15.64}
+
 98%|█████████▊| 363080/371472 [6:43:54<38:59,  3.59it/s]
 98%|█████████▊| 363081/371472 [6:43:55<39:25,  3.55it/s]
 98%|█████████▊| 363082/371472 [6:43:55<39:03,  3.58it/s]
 98%|█████████▊| 363083/371472 [6:43:55<38:56,  3.59it/s]
 98%|█████████▊| 363084/371472 [6:43:55<38:21,  3.64it/s]
 98%|█████████▊| 363085/371472 [6:43:56<36:54,  3.79it/s]
 98%|█████████▊| 363086/371472 [6:43:56<36:19,  3.85it/s]
 98%|█████████▊| 363087/371472 [6:43:56<38:19,  3.65it/s]
 98%|█████████▊| 363088/371472 [6:43:57<41:09,  3.40it/s]
 98%|█████████▊| 363089/371472 [6:43:57<40:46,  3.43it/s]
 98%|█████████▊| 363090/371472 [6:43:57<39:57,  3.50it/s]
 98%|█████████▊| 363091/371472 [6:43:57<40:54,  3.41it/s]
 98%|█████████▊| 363092/371472 [6:43:58<40:51,  3.42it/s]
 98%|█████████▊| 363093/371472 [6:43:58<41:12,  3.39it/s]
 98%|█████████▊| 363094/371472 [6:43:58<39:58,  3.49it/s]
 98%|█████████▊| 363095/371472 [6:43:59<39:10,  3.56it/s]
 98%|█████████▊| 363096/371472 [6:43:59<39:31,  3.53it/s]
 98%|█████████▊| 363097/371472 [6:43:59<40:58,  3.41it/s]
 98%|█████████▊| 363098/371472 [6:43:59<38:21,  3.64it/s]
 98%|█████████▊| 363099/371472 [6:44:00<37:20,  3.74it/s]
 98%|█████████▊| 363100/371472 [6:44:00<39:05,  3.57it/s]
                                                         
{'loss': 2.5884, 'learning_rate': 1.2029455493546512e-07, 'epoch': 15.64}
+
 98%|█████████▊| 363100/371472 [6:44:00<39:05,  3.57it/s]
 98%|█████████▊| 363101/371472 [6:44:00<39:32,  3.53it/s]
 98%|█████████▊| 363102/371472 [6:44:01<41:39,  3.35it/s]
 98%|█████████▊| 363103/371472 [6:44:01<40:09,  3.47it/s]
 98%|█████████▊| 363104/371472 [6:44:01<38:37,  3.61it/s]
 98%|█████████▊| 363105/371472 [6:44:01<39:07,  3.56it/s]
 98%|█████████▊| 363106/371472 [6:44:02<37:42,  3.70it/s]
 98%|█████████▊| 363107/371472 [6:44:02<36:29,  3.82it/s]
 98%|█████████▊| 363108/371472 [6:44:02<36:24,  3.83it/s]
 98%|█████████▊| 363109/371472 [6:44:02<36:35,  3.81it/s]
 98%|█████████▊| 363110/371472 [6:44:03<36:06,  3.86it/s]
 98%|█████████▊| 363111/371472 [6:44:03<38:08,  3.65it/s]
 98%|█████████▊| 363112/371472 [6:44:03<36:42,  3.79it/s]
 98%|█████████▊| 363113/371472 [6:44:03<38:52,  3.58it/s]
 98%|█████████▊| 363114/371472 [6:44:04<39:00,  3.57it/s]
 98%|█████████▊| 363115/371472 [6:44:04<38:44,  3.60it/s]
 98%|█████████▊| 363116/371472 [6:44:04<38:43,  3.60it/s]
 98%|█████████▊| 363117/371472 [6:44:05<37:26,  3.72it/s]
 98%|█████████▊| 363118/371472 [6:44:05<36:52,  3.77it/s]
 98%|█████████▊| 363119/371472 [6:44:05<36:35,  3.81it/s]
 98%|█████████▊| 363120/371472 [6:44:05<36:07,  3.85it/s]
                                                         
{'loss': 2.742, 'learning_rate': 1.202460729599862e-07, 'epoch': 15.64}
+
 98%|█████████▊| 363120/371472 [6:44:05<36:07,  3.85it/s]
 98%|█████████▊| 363121/371472 [6:44:06<36:46,  3.78it/s]
 98%|█████████▊| 363122/371472 [6:44:06<43:10,  3.22it/s]
 98%|█████████▊| 363123/371472 [6:44:06<44:21,  3.14it/s]
 98%|█████████▊| 363124/371472 [6:44:07<42:12,  3.30it/s]
 98%|█████████▊| 363125/371472 [6:44:07<39:46,  3.50it/s]
 98%|█████████▊| 363126/371472 [6:44:07<40:27,  3.44it/s]
 98%|█████████▊| 363127/371472 [6:44:07<38:56,  3.57it/s]
 98%|█████████▊| 363128/371472 [6:44:08<39:03,  3.56it/s]
 98%|█████████▊| 363129/371472 [6:44:08<37:31,  3.71it/s]
 98%|█████████▊| 363130/371472 [6:44:08<36:34,  3.80it/s]
 98%|█████████▊| 363131/371472 [6:44:08<36:25,  3.82it/s]
 98%|█████████▊| 363132/371472 [6:44:09<37:31,  3.70it/s]
 98%|█████████▊| 363133/371472 [6:44:09<38:39,  3.59it/s]
 98%|█████████▊| 363134/371472 [6:44:09<43:18,  3.21it/s]
 98%|█████████▊| 363135/371472 [6:44:10<42:58,  3.23it/s]
 98%|█████████▊| 363136/371472 [6:44:10<42:55,  3.24it/s]
 98%|█████████▊| 363137/371472 [6:44:10<40:28,  3.43it/s]
 98%|█████████▊| 363138/371472 [6:44:11<39:46,  3.49it/s]
 98%|█████████▊| 363139/371472 [6:44:11<39:09,  3.55it/s]
 98%|█████████▊| 363140/371472 [6:44:11<38:03,  3.65it/s]
                                                         
{'loss': 2.5444, 'learning_rate': 1.2019759098450734e-07, 'epoch': 15.64}
+
 98%|█████████▊| 363140/371472 [6:44:11<38:03,  3.65it/s]
 98%|█████████▊| 363141/371472 [6:44:11<37:51,  3.67it/s]
 98%|█████████▊| 363142/371472 [6:44:12<37:49,  3.67it/s]
 98%|█████████▊| 363143/371472 [6:44:12<39:03,  3.55it/s]
 98%|█████████▊| 363144/371472 [6:44:12<43:01,  3.23it/s]
 98%|█████████▊| 363145/371472 [6:44:13<41:24,  3.35it/s]
 98%|█████████▊| 363146/371472 [6:44:13<39:49,  3.48it/s]
 98%|█████████▊| 363147/371472 [6:44:13<38:52,  3.57it/s]
 98%|█████████▊| 363148/371472 [6:44:13<39:09,  3.54it/s]
 98%|█████████▊| 363149/371472 [6:44:14<40:42,  3.41it/s]
 98%|█████████▊| 363150/371472 [6:44:14<40:14,  3.45it/s]
 98%|█████████▊| 363151/371472 [6:44:14<39:36,  3.50it/s]
 98%|█████████▊| 363152/371472 [6:44:15<38:26,  3.61it/s]
 98%|█████████▊| 363153/371472 [6:44:15<38:15,  3.62it/s]
 98%|█████████▊| 363154/371472 [6:44:15<37:46,  3.67it/s]
 98%|█████████▊| 363155/371472 [6:44:15<37:30,  3.70it/s]
 98%|█████████▊| 363156/371472 [6:44:16<37:06,  3.73it/s]
 98%|█████████▊| 363157/371472 [6:44:16<36:44,  3.77it/s]
 98%|█████████▊| 363158/371472 [6:44:16<39:31,  3.51it/s]
 98%|█████████▊| 363159/371472 [6:44:16<38:45,  3.57it/s]
 98%|█████████▊| 363160/371472 [6:44:17<38:02,  3.64it/s]
                                                         
{'loss': 2.6159, 'learning_rate': 1.201491090090284e-07, 'epoch': 15.64}
+
 98%|█████████▊| 363160/371472 [6:44:17<38:02,  3.64it/s]
 98%|█████████▊| 363161/371472 [6:44:17<36:38,  3.78it/s]
 98%|█████████▊| 363162/371472 [6:44:17<40:30,  3.42it/s]
 98%|█████████▊| 363163/371472 [6:44:18<38:46,  3.57it/s]
 98%|█████████▊| 363164/371472 [6:44:18<38:08,  3.63it/s]
 98%|█████████▊| 363165/371472 [6:44:18<38:50,  3.56it/s]
 98%|█████████▊| 363166/371472 [6:44:18<38:12,  3.62it/s]
 98%|█████████▊| 363167/371472 [6:44:19<43:12,  3.20it/s]
 98%|█████████▊| 363168/371472 [6:44:19<40:29,  3.42it/s]
 98%|█████████▊| 363169/371472 [6:44:19<40:58,  3.38it/s]
 98%|█████████▊| 363170/371472 [6:44:20<40:41,  3.40it/s]
 98%|█████████▊| 363171/371472 [6:44:20<43:49,  3.16it/s]
 98%|█████████▊| 363172/371472 [6:44:20<41:03,  3.37it/s]
 98%|█████████▊| 363173/371472 [6:44:21<39:08,  3.53it/s]
 98%|█████████▊| 363174/371472 [6:44:21<40:54,  3.38it/s]
 98%|█████████▊| 363175/371472 [6:44:21<38:43,  3.57it/s]
 98%|█████████▊| 363176/371472 [6:44:21<38:07,  3.63it/s]
 98%|█████████▊| 363177/371472 [6:44:22<38:14,  3.62it/s]
 98%|█████████▊| 363178/371472 [6:44:22<37:18,  3.70it/s]
 98%|█████████▊| 363179/371472 [6:44:22<36:55,  3.74it/s]
 98%|█████████▊| 363180/371472 [6:44:22<37:24,  3.69it/s]
                                                         
{'loss': 2.6473, 'learning_rate': 1.2010062703354957e-07, 'epoch': 15.64}
+
 98%|█████████▊| 363180/371472 [6:44:22<37:24,  3.69it/s]
 98%|█████████▊| 363181/371472 [6:44:23<37:26,  3.69it/s]
 98%|█████████▊| 363182/371472 [6:44:23<37:32,  3.68it/s]
 98%|█████████▊| 363183/371472 [6:44:23<39:36,  3.49it/s]
 98%|█████████▊| 363184/371472 [6:44:24<39:07,  3.53it/s]
 98%|█████████▊| 363185/371472 [6:44:24<39:53,  3.46it/s]
 98%|█████████▊| 363186/371472 [6:44:24<38:05,  3.63it/s]
 98%|█████████▊| 363187/371472 [6:44:24<39:08,  3.53it/s]
 98%|█████████▊| 363188/371472 [6:44:25<38:30,  3.58it/s]
 98%|█████████▊| 363189/371472 [6:44:25<39:12,  3.52it/s]
 98%|█████████▊| 363190/371472 [6:44:25<39:17,  3.51it/s]
 98%|█████████▊| 363191/371472 [6:44:26<38:00,  3.63it/s]
 98%|█████████▊| 363192/371472 [6:44:26<38:53,  3.55it/s]
 98%|█████████▊| 363193/371472 [6:44:26<37:58,  3.63it/s]
 98%|█████████▊| 363194/371472 [6:44:26<37:45,  3.65it/s]
 98%|█████████▊| 363195/371472 [6:44:27<37:20,  3.69it/s]
 98%|█████████▊| 363196/371472 [6:44:27<41:17,  3.34it/s]
 98%|█████████▊| 363197/371472 [6:44:27<41:42,  3.31it/s]
 98%|█████████▊| 363198/371472 [6:44:28<40:14,  3.43it/s]
 98%|██████���██▊| 363199/371472 [6:44:28<39:31,  3.49it/s]
 98%|█████████▊| 363200/371472 [6:44:28<40:41,  3.39it/s]
                                                         
{'loss': 2.4425, 'learning_rate': 1.200521450580706e-07, 'epoch': 15.64}
+
 98%|█████████▊| 363200/371472 [6:44:28<40:41,  3.39it/s]
 98%|█████████▊| 363201/371472 [6:44:28<39:08,  3.52it/s]
 98%|█████████▊| 363202/371472 [6:44:29<49:01,  2.81it/s]
 98%|█████████▊| 363203/371472 [6:44:29<45:50,  3.01it/s]
 98%|█████████▊| 363204/371472 [6:44:30<44:47,  3.08it/s]
 98%|█████████▊| 363205/371472 [6:44:30<43:17,  3.18it/s]
 98%|█████████▊| 363206/371472 [6:44:30<41:41,  3.30it/s]
 98%|█████████▊| 363207/371472 [6:44:30<39:45,  3.46it/s]
 98%|█████████▊| 363208/371472 [6:44:31<37:20,  3.69it/s]
 98%|█████████▊| 363209/371472 [6:44:31<38:25,  3.58it/s]
 98%|█████████▊| 363210/371472 [6:44:31<38:53,  3.54it/s]
 98%|█████████▊| 363211/371472 [6:44:31<37:53,  3.63it/s]
 98%|█████████▊| 363212/371472 [6:44:32<41:43,  3.30it/s]
 98%|█████████▊| 363213/371472 [6:44:32<41:05,  3.35it/s]
 98%|█████████▊| 363214/371472 [6:44:32<39:48,  3.46it/s]
 98%|█████████▊| 363215/371472 [6:44:33<39:42,  3.47it/s]
 98%|█████████▊| 363216/371472 [6:44:33<39:52,  3.45it/s]
 98%|█████████▊| 363217/371472 [6:44:33<38:48,  3.55it/s]
 98%|█████████▊| 363218/371472 [6:44:33<36:48,  3.74it/s]
 98%|█████████▊| 363219/371472 [6:44:34<36:51,  3.73it/s]
 98%|█████████▊| 363220/371472 [6:44:34<44:01,  3.12it/s]
                                                         
{'loss': 2.5308, 'learning_rate': 1.2000366308259176e-07, 'epoch': 15.64}
+
 98%|█████████▊| 363220/371472 [6:44:34<44:01,  3.12it/s]
 98%|█████████▊| 363221/371472 [6:44:34<43:12,  3.18it/s]
 98%|█████████▊| 363222/371472 [6:44:35<42:03,  3.27it/s]
 98%|█████████▊| 363223/371472 [6:44:35<39:34,  3.47it/s]
 98%|█████████▊| 363224/371472 [6:44:35<43:28,  3.16it/s]
 98%|█████████▊| 363225/371472 [6:44:36<41:18,  3.33it/s]
 98%|█████████▊| 363226/371472 [6:44:36<39:44,  3.46it/s]
 98%|█████████▊| 363227/371472 [6:44:36<38:43,  3.55it/s]
 98%|█████████▊| 363228/371472 [6:44:36<37:57,  3.62it/s]
 98%|█████████▊| 363229/371472 [6:44:37<36:56,  3.72it/s]
 98%|█████████▊| 363230/371472 [6:44:37<37:28,  3.66it/s]
 98%|█████████▊| 363231/371472 [6:44:37<36:53,  3.72it/s]
 98%|█████████▊| 363232/371472 [6:44:37<37:01,  3.71it/s]
 98%|█████████▊| 363233/371472 [6:44:38<37:55,  3.62it/s]
 98%|█████████▊| 363234/371472 [6:44:38<37:21,  3.67it/s]
 98%|█████████▊| 363235/371472 [6:44:38<36:55,  3.72it/s]
 98%|█████████▊| 363236/371472 [6:44:39<37:46,  3.63it/s]
 98%|█████████▊| 363237/371472 [6:44:39<38:15,  3.59it/s]
 98%|█████████▊| 363238/371472 [6:44:39<38:24,  3.57it/s]
 98%|█████████▊| 363239/371472 [6:44:39<38:10,  3.59it/s]
 98%|█████████▊| 363240/371472 [6:44:40<38:13,  3.59it/s]
                                                         
{'loss': 2.4085, 'learning_rate': 1.1995518110711283e-07, 'epoch': 15.65}
+
 98%|█████████▊| 363240/371472 [6:44:40<38:13,  3.59it/s]
 98%|█████████▊| 363241/371472 [6:44:40<37:02,  3.70it/s]
 98%|█████████▊| 363242/371472 [6:44:40<41:50,  3.28it/s]
 98%|█████████▊| 363243/371472 [6:44:41<40:43,  3.37it/s]
 98%|█████████▊| 363244/371472 [6:44:41<40:20,  3.40it/s]
 98%|█████████▊| 363245/371472 [6:44:41<42:23,  3.24it/s]
 98%|█████████▊| 363246/371472 [6:44:41<40:06,  3.42it/s]
 98%|█████████▊| 363247/371472 [6:44:42<38:03,  3.60it/s]
 98%|█████████▊| 363248/371472 [6:44:42<36:44,  3.73it/s]
 98%|█████████▊| 363249/371472 [6:44:42<36:54,  3.71it/s]
 98%|█████████▊| 363250/371472 [6:44:43<37:00,  3.70it/s]
 98%|█████████▊| 363251/371472 [6:44:43<37:07,  3.69it/s]
 98%|█████████▊| 363252/371472 [6:44:43<37:39,  3.64it/s]
 98%|█████████▊| 363253/371472 [6:44:43<37:31,  3.65it/s]
 98%|█████████▊| 363254/371472 [6:44:44<36:58,  3.70it/s]
 98%|█████████▊| 363255/371472 [6:44:44<37:15,  3.68it/s]
 98%|█████████▊| 363256/371472 [6:44:44<37:04,  3.69it/s]
 98%|█████████▊| 363257/371472 [6:44:44<37:45,  3.63it/s]
 98%|█████████▊| 363258/371472 [6:44:45<37:52,  3.61it/s]
 98%|█████████▊| 363259/371472 [6:44:45<36:12,  3.78it/s]
 98%|█████████▊| 363260/371472 [6:44:45<37:13,  3.68it/s]
                                                         
{'loss': 2.4244, 'learning_rate': 1.1990669913163399e-07, 'epoch': 15.65}
+
 98%|█████████▊| 363260/371472 [6:44:45<37:13,  3.68it/s]
 98%|█████████▊| 363261/371472 [6:44:46<37:52,  3.61it/s]
 98%|█████████▊| 363262/371472 [6:44:46<36:23,  3.76it/s]
 98%|█████████▊| 363263/371472 [6:44:46<37:38,  3.64it/s]
 98%|█████████▊| 363264/371472 [6:44:46<36:54,  3.71it/s]
 98%|█████████▊| 363265/371472 [6:44:47<36:13,  3.78it/s]
 98%|█████████▊| 363266/371472 [6:44:47<35:31,  3.85it/s]
 98%|█████████▊| 363267/371472 [6:44:47<36:01,  3.80it/s]
 98%|█████████▊| 363268/371472 [6:44:47<36:50,  3.71it/s]
 98%|█████████▊| 363269/371472 [6:44:48<35:56,  3.80it/s]
 98%|█████████▊| 363270/371472 [6:44:48<36:48,  3.71it/s]
 98%|█████████▊| 363271/371472 [6:44:48<36:44,  3.72it/s]
 98%|█████████▊| 363272/371472 [6:44:48<37:27,  3.65it/s]
 98%|█████████▊| 363273/371472 [6:44:49<37:27,  3.65it/s]
 98%|█████████▊| 363274/371472 [6:44:49<36:24,  3.75it/s]
 98%|█████████▊| 363275/371472 [6:44:49<36:58,  3.70it/s]
 98%|█████████▊| 363276/371472 [6:44:50<36:15,  3.77it/s]
 98%|█████████▊| 363277/371472 [6:44:50<41:39,  3.28it/s]
 98%|█████████▊| 363278/371472 [6:44:50<40:05,  3.41it/s]
 98%|█████████▊| 363279/371472 [6:44:50<40:04,  3.41it/s]
 98%|█████████▊| 363280/371472 [6:44:51<39:54,  3.42it/s]
                                                         
{'loss': 2.6898, 'learning_rate': 1.1985821715615506e-07, 'epoch': 15.65}
+
 98%|█████████▊| 363280/371472 [6:44:51<39:54,  3.42it/s]
 98%|█████████▊| 363281/371472 [6:44:51<40:14,  3.39it/s]
 98%|█████████▊| 363282/371472 [6:44:51<38:52,  3.51it/s]
 98%|█████████▊| 363283/371472 [6:44:52<38:12,  3.57it/s]
 98%|█████████▊| 363284/371472 [6:44:52<38:18,  3.56it/s]
 98%|█████████▊| 363285/371472 [6:44:52<38:10,  3.57it/s]
 98%|█████████▊| 363286/371472 [6:44:52<38:17,  3.56it/s]
 98%|█████████▊| 363287/371472 [6:44:53<37:11,  3.67it/s]
 98%|█████████▊| 363288/371472 [6:44:53<36:57,  3.69it/s]
 98%|█████████▊| 363289/371472 [6:44:53<37:03,  3.68it/s]
 98%|█████████▊| 363290/371472 [6:44:54<37:35,  3.63it/s]
 98%|█████████▊| 363291/371472 [6:44:54<36:27,  3.74it/s]
 98%|█████████▊| 363292/371472 [6:44:54<35:58,  3.79it/s]
 98%|█████████▊| 363293/371472 [6:44:54<36:46,  3.71it/s]
 98%|█████████▊| 363294/371472 [6:44:55<37:02,  3.68it/s]
 98%|█████████▊| 363295/371472 [6:44:55<36:21,  3.75it/s]
 98%|█████████▊| 363296/371472 [6:44:55<37:54,  3.59it/s]
 98%|█████████▊| 363297/371472 [6:44:55<37:29,  3.63it/s]
 98%|█████████▊| 363298/371472 [6:44:56<38:35,  3.53it/s]
 98%|█████████▊| 363299/371472 [6:44:56<38:22,  3.55it/s]
 98%|█████████▊| 363300/371472 [6:44:56<39:49,  3.42it/s]
                                                         
{'loss': 2.6376, 'learning_rate': 1.198097351806761e-07, 'epoch': 15.65}
+
 98%|█████████▊| 363300/371472 [6:44:56<39:49,  3.42it/s]
 98%|█████████▊| 363301/371472 [6:44:57<39:17,  3.47it/s]
 98%|█████████▊| 363302/371472 [6:44:57<38:36,  3.53it/s]
 98%|█████████▊| 363303/371472 [6:44:57<38:32,  3.53it/s]
 98%|█████████▊| 363304/371472 [6:44:57<37:52,  3.59it/s]
 98%|█████████▊| 363305/371472 [6:44:58<38:42,  3.52it/s]
 98%|█████████▊| 363306/371472 [6:44:58<37:46,  3.60it/s]
 98%|█████████▊| 363307/371472 [6:44:58<39:07,  3.48it/s]
 98%|█████████▊| 363308/371472 [6:44:59<37:38,  3.61it/s]
 98%|█████████▊| 363309/371472 [6:44:59<37:03,  3.67it/s]
 98%|█████████▊| 363310/371472 [6:44:59<36:05,  3.77it/s]
 98%|█████████▊| 363311/371472 [6:44:59<37:46,  3.60it/s]
 98%|█████████▊| 363312/371472 [6:45:00<37:46,  3.60it/s]
 98%|█████████▊| 363313/371472 [6:45:00<37:29,  3.63it/s]
 98%|██████���██▊| 363314/371472 [6:45:00<36:50,  3.69it/s]
 98%|█████████▊| 363315/371472 [6:45:00<38:14,  3.56it/s]
 98%|█████████▊| 363316/371472 [6:45:01<39:38,  3.43it/s]
 98%|█████████▊| 363317/371472 [6:45:01<41:11,  3.30it/s]
 98%|█████████▊| 363318/371472 [6:45:01<40:10,  3.38it/s]
 98%|█████████▊| 363319/371472 [6:45:02<39:23,  3.45it/s]
 98%|█████████▊| 363320/371472 [6:45:02<39:09,  3.47it/s]
                                                         
{'loss': 2.5261, 'learning_rate': 1.1976125320519728e-07, 'epoch': 15.65}
+
 98%|█████████▊| 363320/371472 [6:45:02<39:09,  3.47it/s]
 98%|█████████▊| 363321/371472 [6:45:02<39:18,  3.46it/s]
 98%|█████████▊| 363322/371472 [6:45:03<41:44,  3.25it/s]
 98%|█████████▊| 363323/371472 [6:45:03<41:11,  3.30it/s]
 98%|█████████▊| 363324/371472 [6:45:03<39:35,  3.43it/s]
 98%|█████████▊| 363325/371472 [6:45:03<38:35,  3.52it/s]
 98%|█████████▊| 363326/371472 [6:45:04<38:41,  3.51it/s]
 98%|█████████▊| 363327/371472 [6:45:04<37:29,  3.62it/s]
 98%|█████████▊| 363328/371472 [6:45:04<38:26,  3.53it/s]
 98%|█████████▊| 363329/371472 [6:45:05<39:00,  3.48it/s]
 98%|█████████▊| 363330/371472 [6:45:05<38:35,  3.52it/s]
 98%|█████████▊| 363331/371472 [6:45:05<38:02,  3.57it/s]
 98%|█████████▊| 363332/371472 [6:45:05<39:33,  3.43it/s]
 98%|█████████▊| 363333/371472 [6:45:06<40:26,  3.35it/s]
 98%|█████████▊| 363334/371472 [6:45:06<41:19,  3.28it/s]
 98%|█████████▊| 363335/371472 [6:45:06<40:46,  3.33it/s]
 98%|█████████▊| 363336/371472 [6:45:07<39:27,  3.44it/s]
 98%|█████████▊| 363337/371472 [6:45:07<39:11,  3.46it/s]
 98%|█████████▊| 363338/371472 [6:45:07<36:59,  3.67it/s]
 98%|█████████▊| 363339/371472 [6:45:07<36:02,  3.76it/s]
 98%|█████████▊| 363340/371472 [6:45:08<37:49,  3.58it/s]
                                                         
{'loss': 2.5846, 'learning_rate': 1.1971277122971832e-07, 'epoch': 15.65}
+
 98%|█████████▊| 363340/371472 [6:45:08<37:49,  3.58it/s]
 98%|█████████▊| 363341/371472 [6:45:08<37:11,  3.64it/s]
 98%|█████████▊| 363342/371472 [6:45:08<38:22,  3.53it/s]
 98%|█████████▊| 363343/371472 [6:45:09<38:11,  3.55it/s]
 98%|█████████▊| 363344/371472 [6:45:09<40:10,  3.37it/s]
 98%|█████████▊| 363345/371472 [6:45:09<39:08,  3.46it/s]
 98%|█████████▊| 363346/371472 [6:45:09<37:54,  3.57it/s]
 98%|█████████▊| 363347/371472 [6:45:10<37:26,  3.62it/s]
 98%|█████████▊| 363348/371472 [6:45:10<38:21,  3.53it/s]
 98%|█████████▊| 363349/371472 [6:45:10<40:52,  3.31it/s]
 98%|█████████▊| 363350/371472 [6:45:11<39:38,  3.41it/s]
 98%|█████████▊| 363351/371472 [6:45:11<39:31,  3.42it/s]
 98%|█████████▊| 363352/371472 [6:45:11<38:14,  3.54it/s]
 98%|█████████▊| 363353/371472 [6:45:11<38:07,  3.55it/s]
 98%|█████████▊| 363354/371472 [6:45:12<37:41,  3.59it/s]
 98%|█████████▊| 363355/371472 [6:45:12<36:20,  3.72it/s]
 98%|█████████▊| 363356/371472 [6:45:12<36:32,  3.70it/s]
 98%|█████████▊| 363357/371472 [6:45:12<36:57,  3.66it/s]
 98%|█████████▊| 363358/371472 [6:45:13<37:11,  3.64it/s]
 98%|█████████▊| 363359/371472 [6:45:13<36:46,  3.68it/s]
 98%|█████████▊| 363360/371472 [6:45:13<35:54,  3.77it/s]
                                                         
{'loss': 2.5755, 'learning_rate': 1.1966428925423948e-07, 'epoch': 15.65}
+
 98%|█████████▊| 363360/371472 [6:45:13<35:54,  3.77it/s]
 98%|█████████▊| 363361/371472 [6:45:14<36:56,  3.66it/s]
 98%|█████████▊| 363362/371472 [6:45:14<37:01,  3.65it/s]
 98%|█████████▊| 363363/371472 [6:45:14<35:56,  3.76it/s]
 98%|█████████▊| 363364/371472 [6:45:14<37:22,  3.62it/s]
 98%|█████████▊| 363365/371472 [6:45:15<42:06,  3.21it/s]
 98%|█████████▊| 363366/371472 [6:45:15<45:04,  3.00it/s]
 98%|█████████▊| 363367/371472 [6:45:15<44:29,  3.04it/s]
 98%|█████████▊| 363368/371472 [6:45:16<41:55,  3.22it/s]
 98%|█████████▊| 363369/371472 [6:45:16<41:29,  3.26it/s]
 98%|█████████▊| 363370/371472 [6:45:16<40:30,  3.33it/s]
 98%|█████████▊| 363371/371472 [6:45:17<38:52,  3.47it/s]
 98%|█████████▊| 363372/371472 [6:45:17<38:30,  3.51it/s]
 98%|█████████▊| 363373/371472 [6:45:17<37:37,  3.59it/s]
 98%|█████████▊| 363374/371472 [6:45:17<37:07,  3.64it/s]
 98%|█████████▊| 363375/371472 [6:45:18<37:00,  3.65it/s]
 98%|█████████▊| 363376/371472 [6:45:18<35:19,  3.82it/s]
 98%|█████████▊| 363377/371472 [6:45:18<36:15,  3.72it/s]
 98%|█████████▊| 363378/371472 [6:45:18<36:02,  3.74it/s]
 98%|█████████▊| 363379/371472 [6:45:19<39:34,  3.41it/s]
 98%|█████████▊| 363380/371472 [6:45:19<38:13,  3.53it/s]
                                                         
{'loss': 2.4491, 'learning_rate': 1.1961580727876055e-07, 'epoch': 15.65}
+
 98%|█████████▊| 363380/371472 [6:45:19<38:13,  3.53it/s]
 98%|█████████▊| 363381/371472 [6:45:19<37:27,  3.60it/s]
 98%|█████████▊| 363382/371472 [6:45:20<36:59,  3.65it/s]
 98%|█████████▊| 363383/371472 [6:45:20<36:30,  3.69it/s]
 98%|█████████▊| 363384/371472 [6:45:20<36:53,  3.65it/s]
 98%|█████████▊| 363385/371472 [6:45:20<35:46,  3.77it/s]
 98%|█████████▊| 363386/371472 [6:45:21<37:14,  3.62it/s]
 98%|█████████▊| 363387/371472 [6:45:21<39:20,  3.42it/s]
 98%|█████████▊| 363388/371472 [6:45:21<38:41,  3.48it/s]
 98%|█████████▊| 363389/371472 [6:45:22<36:41,  3.67it/s]
 98%|█████████▊| 363390/371472 [6:45:22<36:39,  3.67it/s]
 98%|█████████▊| 363391/371472 [6:45:22<35:32,  3.79it/s]
 98%|█████████▊| 363392/371472 [6:45:22<34:45,  3.87it/s]
 98%|█████████▊| 363393/371472 [6:45:23<35:36,  3.78it/s]
 98%|█████████▊| 363394/371472 [6:45:23<37:09,  3.62it/s]
 98%|█████████▊| 363395/371472 [6:45:23<38:34,  3.49it/s]
 98%|█████████▊| 363396/371472 [6:45:24<40:27,  3.33it/s]
 98%|█████████▊| 363397/371472 [6:45:24<40:58,  3.28it/s]
 98%|█████████▊| 363398/371472 [6:45:24<42:16,  3.18it/s]
 98%|█████████▊| 363399/371472 [6:45:24<39:39,  3.39it/s]
 98%|█████████▊| 363400/371472 [6:45:25<37:54,  3.55it/s]
                                                         
{'loss': 2.5875, 'learning_rate': 1.195673253032817e-07, 'epoch': 15.65}
+
 98%|█████████▊| 363400/371472 [6:45:25<37:54,  3.55it/s]
 98%|█████████▊| 363401/371472 [6:45:25<37:54,  3.55it/s]
 98%|█████████▊| 363402/371472 [6:45:25<37:05,  3.63it/s]
 98%|█████████▊| 363403/371472 [6:45:26<37:07,  3.62it/s]
 98%|█████████▊| 363404/371472 [6:45:26<35:26,  3.79it/s]
 98%|█████████▊| 363405/371472 [6:45:26<40:55,  3.29it/s]
 98%|█████████▊| 363406/371472 [6:45:26<40:19,  3.33it/s]
 98%|█████████▊| 363407/371472 [6:45:27<41:34,  3.23it/s]
 98%|█████████▊| 363408/371472 [6:45:27<40:23,  3.33it/s]
 98%|█████████▊| 363409/371472 [6:45:27<38:24,  3.50it/s]
 98%|█████████▊| 363410/371472 [6:45:28<37:45,  3.56it/s]
 98%|█████████▊| 363411/371472 [6:45:28<42:00,  3.20it/s]
 98%|█████████▊| 363412/371472 [6:45:28<38:43,  3.47it/s]
 98%|█████████▊| 363413/371472 [6:45:28<38:27,  3.49it/s]
 98%|█████████▊| 363414/371472 [6:45:29<38:05,  3.53it/s]
 98%|█████████▊| 363415/371472 [6:45:29<37:00,  3.63it/s]
 98%|█████████▊| 363416/371472 [6:45:29<37:07,  3.62it/s]
 98%|█████████▊| 363417/371472 [6:45:30<36:18,  3.70it/s]
 98%|█████████▊| 363418/371472 [6:45:30<36:47,  3.65it/s]
 98%|█████████▊| 363419/371472 [6:45:30<37:41,  3.56it/s]
 98%|█████████▊| 363420/371472 [6:45:30<38:42,  3.47it/s]
                                                         
{'loss': 2.526, 'learning_rate': 1.1951884332780274e-07, 'epoch': 15.65}
+
 98%|█████████▊| 363420/371472 [6:45:30<38:42,  3.47it/s]
 98%|█████████▊| 363421/371472 [6:45:31<40:14,  3.33it/s]
 98%|█████████▊| 363422/371472 [6:45:31<39:07,  3.43it/s]
 98%|█████████▊| 363423/371472 [6:45:31<37:01,  3.62it/s]
 98%|█████████▊| 363424/371472 [6:45:32<39:23,  3.41it/s]
 98%|█████████▊| 363425/371472 [6:45:32<38:50,  3.45it/s]
 98%|█████████▊| 363426/371472 [6:45:32<39:45,  3.37it/s]
 98%|█████████▊| 363427/371472 [6:45:32<38:37,  3.47it/s]
 98%|█████████▊| 363428/371472 [6:45:33<37:23,  3.59it/s]
 98%|██████���██▊| 363429/371472 [6:45:33<36:22,  3.68it/s]
 98%|█████████▊| 363430/371472 [6:45:33<36:14,  3.70it/s]
 98%|█████████▊| 363431/371472 [6:45:34<38:48,  3.45it/s]
 98%|█████████▊| 363432/371472 [6:45:34<37:49,  3.54it/s]
 98%|█████████▊| 363433/371472 [6:45:34<37:41,  3.55it/s]
 98%|█████████▊| 363434/371472 [6:45:34<36:43,  3.65it/s]
 98%|█████████▊| 363435/371472 [6:45:35<36:54,  3.63it/s]
 98%|█████████▊| 363436/371472 [6:45:35<36:58,  3.62it/s]
 98%|█████████▊| 363437/371472 [6:45:35<37:28,  3.57it/s]
 98%|█████████▊| 363438/371472 [6:45:36<38:11,  3.51it/s]
 98%|█████████▊| 363439/371472 [6:45:36<37:03,  3.61it/s]
 98%|█████████▊| 363440/371472 [6:45:36<37:16,  3.59it/s]
                                                         
{'loss': 2.6976, 'learning_rate': 1.1947036135232392e-07, 'epoch': 15.65}
+
 98%|█████████▊| 363440/371472 [6:45:36<37:16,  3.59it/s]
 98%|█████████▊| 363441/371472 [6:45:36<36:53,  3.63it/s]
 98%|█████████▊| 363442/371472 [6:45:37<38:28,  3.48it/s]
 98%|█████████▊| 363443/371472 [6:45:37<39:21,  3.40it/s]
 98%|█████████▊| 363444/371472 [6:45:37<38:05,  3.51it/s]
 98%|█████████▊| 363445/371472 [6:45:37<38:40,  3.46it/s]
 98%|█████████▊| 363446/371472 [6:45:38<39:06,  3.42it/s]
 98%|█████████▊| 363447/371472 [6:45:38<38:28,  3.48it/s]
 98%|█████████▊| 363448/371472 [6:45:38<41:33,  3.22it/s]
 98%|█████████▊| 363449/371472 [6:45:39<41:49,  3.20it/s]
 98%|█████████▊| 363450/371472 [6:45:39<39:35,  3.38it/s]
 98%|█████████▊| 363451/371472 [6:45:39<39:56,  3.35it/s]
 98%|█████████▊| 363452/371472 [6:45:40<38:46,  3.45it/s]
 98%|█████████▊| 363453/371472 [6:45:40<36:44,  3.64it/s]
 98%|█████████▊| 363454/371472 [6:45:40<39:34,  3.38it/s]
 98%|█████████▊| 363455/371472 [6:45:41<41:00,  3.26it/s]
 98%|█████████▊| 363456/371472 [6:45:41<39:51,  3.35it/s]
 98%|█████████▊| 363457/371472 [6:45:41<38:43,  3.45it/s]
 98%|█████████▊| 363458/371472 [6:45:41<38:22,  3.48it/s]
 98%|█████████▊| 363459/371472 [6:45:42<37:18,  3.58it/s]
 98%|█████████▊| 363460/371472 [6:45:42<39:46,  3.36it/s]
                                                         
{'loss': 2.4426, 'learning_rate': 1.1942187937684496e-07, 'epoch': 15.65}
+
 98%|█████████▊| 363460/371472 [6:45:42<39:46,  3.36it/s]
 98%|█████████▊| 363461/371472 [6:45:42<38:54,  3.43it/s]
 98%|█████████▊| 363462/371472 [6:45:42<37:10,  3.59it/s]
 98%|█████████▊| 363463/371472 [6:45:43<36:11,  3.69it/s]
 98%|█████████▊| 363464/371472 [6:45:43<35:15,  3.79it/s]
 98%|█████████▊| 363465/371472 [6:45:43<35:22,  3.77it/s]
 98%|█████████▊| 363466/371472 [6:45:44<36:47,  3.63it/s]
 98%|█████████▊| 363467/371472 [6:45:44<40:33,  3.29it/s]
 98%|█████████▊| 363468/371472 [6:45:44<38:28,  3.47it/s]
 98%|█████████▊| 363469/371472 [6:45:44<39:33,  3.37it/s]
 98%|█████████▊| 363470/371472 [6:45:45<37:59,  3.51it/s]
 98%|█████████▊| 363471/371472 [6:45:45<37:28,  3.56it/s]
 98%|█████████▊| 363472/371472 [6:45:45<38:07,  3.50it/s]
 98%|█████████▊| 363473/371472 [6:45:46<40:33,  3.29it/s]
 98%|█████████▊| 363474/371472 [6:45:46<38:54,  3.43it/s]
 98%|█████████▊| 363475/371472 [6:45:46<40:43,  3.27it/s]
 98%|█████████▊| 363476/371472 [6:45:47<38:50,  3.43it/s]
 98%|█████████▊| 363477/371472 [6:45:47<37:20,  3.57it/s]
 98%|█████████▊| 363478/371472 [6:45:47<37:00,  3.60it/s]
 98%|█████████▊| 363479/371472 [6:45:47<39:27,  3.38it/s]
 98%|█████████▊| 363480/371472 [6:45:48<39:34,  3.37it/s]
                                                         
{'loss': 2.3106, 'learning_rate': 1.1937339740136612e-07, 'epoch': 15.66}
+
 98%|█████████▊| 363480/371472 [6:45:48<39:34,  3.37it/s]
 98%|█████████▊| 363481/371472 [6:45:48<37:45,  3.53it/s]
 98%|█████████▊| 363482/371472 [6:45:48<40:34,  3.28it/s]
 98%|█████████▊| 363483/371472 [6:45:49<38:52,  3.43it/s]
 98%|█████████▊| 363484/371472 [6:45:49<37:29,  3.55it/s]
 98%|█████████▊| 363485/371472 [6:45:49<36:27,  3.65it/s]
 98%|█████████▊| 363486/371472 [6:45:49<35:53,  3.71it/s]
 98%|█████████▊| 363487/371472 [6:45:50<35:15,  3.77it/s]
 98%|█████████▊| 363488/371472 [6:45:50<35:57,  3.70it/s]
 98%|█████████▊| 363489/371472 [6:45:50<40:11,  3.31it/s]
 98%|█████████▊| 363490/371472 [6:45:51<40:14,  3.31it/s]
 98%|█████████▊| 363491/371472 [6:45:51<39:27,  3.37it/s]
 98%|█████████▊| 363492/371472 [6:45:51<38:13,  3.48it/s]
 98%|█████████▊| 363493/371472 [6:45:51<38:34,  3.45it/s]
 98%|█████████▊| 363494/371472 [6:45:52<37:48,  3.52it/s]
 98%|█████████▊| 363495/371472 [6:45:52<36:39,  3.63it/s]
 98%|█████████▊| 363496/371472 [6:45:52<35:37,  3.73it/s]
 98%|█████████▊| 363497/371472 [6:45:52<36:38,  3.63it/s]
 98%|█████████▊| 363498/371472 [6:45:53<38:06,  3.49it/s]
 98%|█████████▊| 363499/371472 [6:45:53<37:26,  3.55it/s]
 98%|█████████▊| 363500/371472 [6:45:53<37:40,  3.53it/s]
                                                         
{'loss': 2.5534, 'learning_rate': 1.193249154258872e-07, 'epoch': 15.66}
+
 98%|█████████▊| 363500/371472 [6:45:53<37:40,  3.53it/s]
 98%|█████████▊| 363501/371472 [6:45:54<37:12,  3.57it/s]
 98%|█████████▊| 363502/371472 [6:45:54<39:30,  3.36it/s]
 98%|█████████▊| 363503/371472 [6:45:54<38:16,  3.47it/s]
 98%|█████████▊| 363504/371472 [6:45:54<37:17,  3.56it/s]
 98%|█████████▊| 363505/371472 [6:45:55<36:32,  3.63it/s]
 98%|█████████▊| 363506/371472 [6:45:55<36:04,  3.68it/s]
 98%|█████████▊| 363507/371472 [6:45:55<39:15,  3.38it/s]
 98%|█████████▊| 363508/371472 [6:45:56<39:43,  3.34it/s]
 98%|█████████▊| 363509/371472 [6:45:56<38:10,  3.48it/s]
 98%|█████████▊| 363510/371472 [6:45:56<37:35,  3.53it/s]
 98%|█████████▊| 363511/371472 [6:45:57<40:37,  3.27it/s]
 98%|█████████▊| 363512/371472 [6:45:57<38:26,  3.45it/s]
 98%|█████████▊| 363513/371472 [6:45:57<36:52,  3.60it/s]
 98%|█████████▊| 363514/371472 [6:45:57<36:16,  3.66it/s]
 98%|█████████▊| 363515/371472 [6:45:58<35:31,  3.73it/s]
 98%|█████████▊| 363516/371472 [6:45:58<35:10,  3.77it/s]
 98%|█████████▊| 363517/371472 [6:45:58<34:28,  3.85it/s]
 98%|█████████▊| 363518/371472 [6:45:58<35:04,  3.78it/s]
 98%|█████████▊| 363519/371472 [6:45:59<34:09,  3.88it/s]
 98%|█████████▊| 363520/371472 [6:45:59<34:55,  3.79it/s]
                                                         
{'loss': 2.5937, 'learning_rate': 1.1927643345040834e-07, 'epoch': 15.66}
+
 98%|█████████▊| 363520/371472 [6:45:59<34:55,  3.79it/s]
 98%|█████████▊| 363521/371472 [6:45:59<36:22,  3.64it/s]
 98%|█████████▊| 363522/371472 [6:45:59<35:00,  3.78it/s]
 98%|█████████▊| 363523/371472 [6:46:00<35:08,  3.77it/s]
 98%|█████████▊| 363524/371472 [6:46:00<38:49,  3.41it/s]
 98%|█████████▊| 363525/371472 [6:46:00<40:26,  3.27it/s]
 98%|█████████▊| 363526/371472 [6:46:01<38:52,  3.41it/s]
 98%|█████████▊| 363527/371472 [6:46:01<38:22,  3.45it/s]
 98%|█████████▊| 363528/371472 [6:46:01<38:44,  3.42it/s]
 98%|█████████▊| 363529/371472 [6:46:01<38:23,  3.45it/s]
 98%|█████████▊| 363530/371472 [6:46:02<36:12,  3.65it/s]
 98%|█████████▊| 363531/371472 [6:46:02<36:49,  3.59it/s]
 98%|█████████▊| 363532/371472 [6:46:02<37:17,  3.55it/s]
 98%|█████████▊| 363533/371472 [6:46:03<36:11,  3.66it/s]
 98%|█████████▊| 363534/371472 [6:46:03<36:46,  3.60it/s]
 98%|█████████▊| 363535/371472 [6:46:03<36:44,  3.60it/s]
 98%|█████████▊| 363536/371472 [6:46:03<36:16,  3.65it/s]
 98%|█████████▊| 363537/371472 [6:46:04<35:30,  3.72it/s]
 98%|█████████▊| 363538/371472 [6:46:04<35:57,  3.68it/s]
 98%|█████████▊| 363539/371472 [6:46:04<35:57,  3.68it/s]
 98%|█████████▊| 363540/371472 [6:46:05<37:37,  3.51it/s]
                                                         
{'loss': 2.6376, 'learning_rate': 1.192279514749294e-07, 'epoch': 15.66}
+
 98%|█████████▊| 363540/371472 [6:46:05<37:37,  3.51it/s]
 98%|█████████▊| 363541/371472 [6:46:05<37:33,  3.52it/s]
 98%|█████████▊| 363542/371472 [6:46:05<36:32,  3.62it/s]
 98%|█████████▊| 363543/371472 [6:46:05<37:53,  3.49it/s]
 98%|██████���██▊| 363544/371472 [6:46:06<36:09,  3.65it/s]
 98%|█████████▊| 363545/371472 [6:46:06<36:38,  3.61it/s]
 98%|█████████▊| 363546/371472 [6:46:06<36:34,  3.61it/s]
 98%|█████████▊| 363547/371472 [6:46:06<37:46,  3.50it/s]
 98%|█████████▊| 363548/371472 [6:46:07<36:45,  3.59it/s]
 98%|█████████▊| 363549/371472 [6:46:07<38:49,  3.40it/s]
 98%|█████████▊| 363550/371472 [6:46:07<40:28,  3.26it/s]
 98%|█████████▊| 363551/371472 [6:46:08<37:39,  3.51it/s]
 98%|█████████▊| 363552/371472 [6:46:08<36:57,  3.57it/s]
 98%|█████████▊| 363553/371472 [6:46:08<41:47,  3.16it/s]
 98%|█████████▊| 363554/371472 [6:46:09<39:46,  3.32it/s]
 98%|█████████▊| 363555/371472 [6:46:09<36:55,  3.57it/s]
 98%|█████████▊| 363556/371472 [6:46:09<37:03,  3.56it/s]
 98%|█████████▊| 363557/371472 [6:46:09<37:23,  3.53it/s]
 98%|█████████▊| 363558/371472 [6:46:10<35:37,  3.70it/s]
 98%|█████████▊| 363559/371472 [6:46:10<35:23,  3.73it/s]
 98%|█████████▊| 363560/371472 [6:46:10<36:00,  3.66it/s]
                                                         
{'loss': 2.5423, 'learning_rate': 1.1917946949945056e-07, 'epoch': 15.66}
+
 98%|█████████▊| 363560/371472 [6:46:10<36:00,  3.66it/s]
 98%|█████████▊| 363561/371472 [6:46:10<35:26,  3.72it/s]
 98%|█████████▊| 363562/371472 [6:46:11<35:14,  3.74it/s]
 98%|█████████▊| 363563/371472 [6:46:11<35:57,  3.67it/s]
 98%|█████████▊| 363564/371472 [6:46:11<35:05,  3.76it/s]
 98%|█████████▊| 363565/371472 [6:46:11<34:54,  3.78it/s]
 98%|█████████▊| 363566/371472 [6:46:12<34:08,  3.86it/s]
 98%|█████████▊| 363567/371472 [6:46:12<34:15,  3.85it/s]
 98%|█████████▊| 363568/371472 [6:46:12<35:12,  3.74it/s]
 98%|█████████▊| 363569/371472 [6:46:13<34:57,  3.77it/s]
 98%|█████████▊| 363570/371472 [6:46:13<35:02,  3.76it/s]
 98%|█████████▊| 363571/371472 [6:46:13<34:50,  3.78it/s]
 98%|█████████▊| 363572/371472 [6:46:13<35:27,  3.71it/s]
 98%|█████████▊| 363573/371472 [6:46:14<36:11,  3.64it/s]
 98%|█████████▊| 363574/371472 [6:46:14<35:52,  3.67it/s]
 98%|█████████▊| 363575/371472 [6:46:14<36:13,  3.63it/s]
 98%|█████████▊| 363576/371472 [6:46:14<36:45,  3.58it/s]
 98%|█████████▊| 363577/371472 [6:46:15<38:31,  3.41it/s]
 98%|█████████▊| 363578/371472 [6:46:15<39:00,  3.37it/s]
 98%|█████████▊| 363579/371472 [6:46:15<36:55,  3.56it/s]
 98%|█████████▊| 363580/371472 [6:46:16<36:05,  3.64it/s]
                                                         
{'loss': 2.489, 'learning_rate': 1.1913098752397162e-07, 'epoch': 15.66}
+
 98%|█████████▊| 363580/371472 [6:46:16<36:05,  3.64it/s]
 98%|█████████▊| 363581/371472 [6:46:16<35:14,  3.73it/s]
 98%|█████████▊| 363582/371472 [6:46:16<35:30,  3.70it/s]
 98%|█████████▊| 363583/371472 [6:46:16<35:24,  3.71it/s]
 98%|█████████▊| 363584/371472 [6:46:17<37:50,  3.47it/s]
 98%|█████████▊| 363585/371472 [6:46:17<37:57,  3.46it/s]
 98%|█████████▊| 363586/371472 [6:46:17<36:26,  3.61it/s]
 98%|█████████▊| 363587/371472 [6:46:18<36:14,  3.63it/s]
 98%|█████████▊| 363588/371472 [6:46:18<36:46,  3.57it/s]
 98%|█████████▊| 363589/371472 [6:46:18<35:32,  3.70it/s]
 98%|█████████▊| 363590/371472 [6:46:18<34:46,  3.78it/s]
 98%|█████████▊| 363591/371472 [6:46:19<36:25,  3.61it/s]
 98%|█████████▊| 363592/371472 [6:46:19<35:21,  3.72it/s]
 98%|█████████▊| 363593/371472 [6:46:19<35:25,  3.71it/s]
 98%|█████████▊| 363594/371472 [6:46:19<37:24,  3.51it/s]
 98%|█████████▊| 363595/371472 [6:46:20<36:22,  3.61it/s]
 98%|█████████▊| 363596/371472 [6:46:20<35:44,  3.67it/s]
 98%|█████████▊| 363597/371472 [6:46:20<34:24,  3.81it/s]
 98%|█████████▊| 363598/371472 [6:46:21<36:06,  3.63it/s]
 98%|█████████▊| 363599/371472 [6:46:21<39:53,  3.29it/s]
 98%|█████████▊| 363600/371472 [6:46:21<37:47,  3.47it/s]
                                                         
{'loss': 2.6304, 'learning_rate': 1.1908250554849277e-07, 'epoch': 15.66}
+
 98%|█████████▊| 363600/371472 [6:46:21<37:47,  3.47it/s]
 98%|█████████▊| 363601/371472 [6:46:21<36:49,  3.56it/s]
 98%|█████████▊| 363602/371472 [6:46:22<38:29,  3.41it/s]
 98%|█████████▊| 363603/371472 [6:46:22<37:37,  3.49it/s]
 98%|█████████▊| 363604/371472 [6:46:22<36:46,  3.57it/s]
 98%|█████████▊| 363605/371472 [6:46:23<35:38,  3.68it/s]
 98%|█████████▊| 363606/371472 [6:46:23<36:28,  3.59it/s]
 98%|█████████▊| 363607/371472 [6:46:23<36:03,  3.63it/s]
 98%|█████████▊| 363608/371472 [6:46:23<35:44,  3.67it/s]
 98%|█████████▊| 363609/371472 [6:46:24<38:23,  3.41it/s]
 98%|█████████▊| 363610/371472 [6:46:24<37:49,  3.46it/s]
 98%|█████████▊| 363611/371472 [6:46:24<36:37,  3.58it/s]
 98%|█████████▊| 363612/371472 [6:46:25<35:53,  3.65it/s]
 98%|█████████▊| 363613/371472 [6:46:25<35:24,  3.70it/s]
 98%|█████████▊| 363614/371472 [6:46:25<36:56,  3.54it/s]
 98%|█████████▊| 363615/371472 [6:46:25<39:17,  3.33it/s]
 98%|█████████▊| 363616/371472 [6:46:26<38:55,  3.36it/s]
 98%|█████████▊| 363617/371472 [6:46:26<37:56,  3.45it/s]
 98%|█████████▊| 363618/371472 [6:46:26<36:39,  3.57it/s]
 98%|█████████▊| 363619/371472 [6:46:27<36:09,  3.62it/s]
 98%|█████████▊| 363620/371472 [6:46:27<37:24,  3.50it/s]
                                                         
{'loss': 2.5457, 'learning_rate': 1.1903402357301383e-07, 'epoch': 15.66}
+
 98%|█████████▊| 363620/371472 [6:46:27<37:24,  3.50it/s]
 98%|█████████▊| 363621/371472 [6:46:27<36:20,  3.60it/s]
 98%|█████████▊| 363622/371472 [6:46:27<35:40,  3.67it/s]
 98%|█████████▊| 363623/371472 [6:46:28<34:33,  3.79it/s]
 98%|█████████▊| 363624/371472 [6:46:28<34:09,  3.83it/s]
 98%|█████████▊| 363625/371472 [6:46:28<35:51,  3.65it/s]
 98%|█████████▊| 363626/371472 [6:46:28<35:46,  3.66it/s]
 98%|█████████▊| 363627/371472 [6:46:29<34:53,  3.75it/s]
 98%|█████████▊| 363628/371472 [6:46:29<34:22,  3.80it/s]
 98%|█████████▊| 363629/371472 [6:46:29<33:45,  3.87it/s]
 98%|█████████▊| 363630/371472 [6:46:29<34:34,  3.78it/s]
 98%|█████████▊| 363631/371472 [6:46:30<34:03,  3.84it/s]
 98%|█████████▊| 363632/371472 [6:46:30<33:57,  3.85it/s]
 98%|█████████▊| 363633/371472 [6:46:30<34:56,  3.74it/s]
 98%|█████████▊| 363634/371472 [6:46:30<33:58,  3.84it/s]
 98%|█████████▊| 363635/371472 [6:46:31<35:35,  3.67it/s]
 98%|█████████▊| 363636/371472 [6:46:31<34:15,  3.81it/s]
 98%|█████████▊| 363637/371472 [6:46:31<34:10,  3.82it/s]
 98%|█████████▊| 363638/371472 [6:46:32<35:21,  3.69it/s]
 98%|█████████▊| 363639/371472 [6:46:32<36:23,  3.59it/s]
 98%|█████████▊| 363640/371472 [6:46:32<36:16,  3.60it/s]
                                                         
{'loss': 2.7068, 'learning_rate': 1.1898554159753499e-07, 'epoch': 15.66}
+
 98%|█████████▊| 363640/371472 [6:46:32<36:16,  3.60it/s]
 98%|█████████▊| 363641/371472 [6:46:32<35:09,  3.71it/s]
 98%|█████████▊| 363642/371472 [6:46:33<35:36,  3.66it/s]
 98%|█████████▊| 363643/371472 [6:46:33<35:55,  3.63it/s]
 98%|█████████▊| 363644/371472 [6:46:33<34:42,  3.76it/s]
 98%|█████████▊| 363645/371472 [6:46:33<34:50,  3.74it/s]
 98%|█████████▊| 363646/371472 [6:46:34<35:00,  3.73it/s]
 98%|█████████▊| 363647/371472 [6:46:34<35:45,  3.65it/s]
 98%|█████████▊| 363648/371472 [6:46:34<36:07,  3.61it/s]
 98%|█████████▊| 363649/371472 [6:46:35<36:25,  3.58it/s]
 98%|█████████▊| 363650/371472 [6:46:35<38:30,  3.39it/s]
 98%|█████████▊| 363651/371472 [6:46:35<37:44,  3.45it/s]
 98%|█████████▊| 363652/371472 [6:46:35<37:12,  3.50it/s]
 98%|█████████▊| 363653/371472 [6:46:36<35:54,  3.63it/s]
 98%|█████████▊| 363654/371472 [6:46:36<42:37,  3.06it/s]
 98%|█████████▊| 363655/371472 [6:46:36<40:36,  3.21it/s]
 98%|█████████▊| 363656/371472 [6:46:37<39:36,  3.29it/s]
 98%|█████████▊| 363657/371472 [6:46:37<37:49,  3.44it/s]
 98%|█████████▊| 363658/371472 [6:46:37<36:38,  3.55it/s]
 98%|█████████▊| 363659/371472 [6:46:38<36:23,  3.58it/s]
 98%|█████████▊| 363660/371472 [6:46:38<37:01,  3.52it/s]
                                                         
{'loss': 2.7249, 'learning_rate': 1.1893705962205605e-07, 'epoch': 15.66}
+
 98%|█████████▊| 363660/371472 [6:46:38<37:01,  3.52it/s]
 98%|█████████▊| 363661/371472 [6:46:38<38:07,  3.42it/s]
 98%|█████████▊| 363662/371472 [6:46:38<40:07,  3.24it/s]
 98%|█████████▊| 363663/371472 [6:46:39<40:04,  3.25it/s]
 98%|█████████▊| 363664/371472 [6:46:39<38:39,  3.37it/s]
 98%|█████████▊| 363665/371472 [6:46:39<37:22,  3.48it/s]
 98%|█████████▊| 363666/371472 [6:46:40<37:46,  3.44it/s]
 98%|█████████▊| 363667/371472 [6:46:40<37:21,  3.48it/s]
 98%|█████████▊| 363668/371472 [6:46:40<36:52,  3.53it/s]
 98%|█████████▊| 363669/371472 [6:46:40<36:09,  3.60it/s]
 98%|█████████▊| 363670/371472 [6:46:41<36:30,  3.56it/s]
 98%|█████████▊| 363671/371472 [6:46:41<35:00,  3.71it/s]
 98%|█████████▊| 363672/371472 [6:46:41<35:47,  3.63it/s]
 98%|█████████▊| 363673/371472 [6:46:42<34:53,  3.73it/s]
 98%|█████████▊| 363674/371472 [6:46:42<34:09,  3.80it/s]
 98%|█████████▊| 363675/371472 [6:46:42<33:44,  3.85it/s]
 98%|█████████▊| 363676/371472 [6:46:42<37:38,  3.45it/s]
 98%|█████████▊| 363677/371472 [6:46:43<36:36,  3.55it/s]
 98%|█████████▊| 363678/371472 [6:46:43<35:12,  3.69it/s]
 98%|█████████▊| 363679/371472 [6:46:43<34:37,  3.75it/s]
 98%|█████████▊| 363680/371472 [6:46:43<34:44,  3.74it/s]
                                                         
{'loss': 2.6517, 'learning_rate': 1.188885776465772e-07, 'epoch': 15.66}
+
 98%|█████████▊| 363680/371472 [6:46:43<34:44,  3.74it/s]
 98%|█████████▊| 363681/371472 [6:46:44<35:58,  3.61it/s]
 98%|█████████▊| 363682/371472 [6:46:44<36:09,  3.59it/s]
 98%|█████████▊| 363683/371472 [6:46:44<35:40,  3.64it/s]
 98%|█████████▊| 363684/371472 [6:46:45<36:24,  3.56it/s]
 98%|█████████▊| 363685/371472 [6:46:45<36:55,  3.52it/s]
 98%|█████████▊| 363686/371472 [6:46:45<35:34,  3.65it/s]
 98%|█████████▊| 363687/371472 [6:46:45<37:08,  3.49it/s]
 98%|█████████▊| 363688/371472 [6:46:46<35:07,  3.69it/s]
 98%|█████████▊| 363689/371472 [6:46:46<35:08,  3.69it/s]
 98%|█████████▊| 363690/371472 [6:46:46<35:35,  3.64it/s]
 98%|█████████▊| 363691/371472 [6:46:47<36:45,  3.53it/s]
 98%|█████████▊| 363692/371472 [6:46:47<35:58,  3.61it/s]
 98%|█████████▊| 363693/371472 [6:46:47<35:35,  3.64it/s]
 98%|█████████▊| 363694/371472 [6:46:47<35:56,  3.61it/s]
 98%|█████████▊| 363695/371472 [6:46:48<34:47,  3.73it/s]
 98%|█████████▊| 363696/371472 [6:46:48<34:51,  3.72it/s]
 98%|█████████▊| 363697/371472 [6:46:48<35:12,  3.68it/s]
 98%|█████████▊| 363698/371472 [6:46:48<34:39,  3.74it/s]
 98%|█████████▊| 363699/371472 [6:46:49<33:49,  3.83it/s]
 98%|█████████▊| 363700/371472 [6:46:49<34:44,  3.73it/s]
                                                         
{'loss': 2.5935, 'learning_rate': 1.1884009567109826e-07, 'epoch': 15.67}
+
 98%|█████████▊| 363700/371472 [6:46:49<34:44,  3.73it/s]
 98%|█████████▊| 363701/371472 [6:46:49<35:06,  3.69it/s]
 98%|█████████▊| 363702/371472 [6:46:49<34:07,  3.80it/s]
 98%|█████████▊| 363703/371472 [6:46:50<35:06,  3.69it/s]
 98%|█████████▊| 363704/371472 [6:46:50<34:23,  3.76it/s]
 98%|█████████▊| 363705/371472 [6:46:50<35:24,  3.66it/s]
 98%|█████████▊| 363706/371472 [6:46:51<34:49,  3.72it/s]
 98%|█████████▊| 363707/371472 [6:46:51<36:25,  3.55it/s]
 98%|█████████▊| 363708/371472 [6:46:51<36:17,  3.57it/s]
 98%|█████████▊| 363709/371472 [6:46:51<35:59,  3.59it/s]
 98%|█████████▊| 363710/371472 [6:46:52<36:16,  3.57it/s]
 98%|█████████▊| 363711/371472 [6:46:52<34:53,  3.71it/s]
 98%|█████████▊| 363712/371472 [6:46:52<35:46,  3.62it/s]
 98%|█████████▊| 363713/371472 [6:46:52<35:46,  3.62it/s]
 98%|█████████▊| 363714/371472 [6:46:53<36:19,  3.56it/s]
 98%|█████████▊| 363715/371472 [6:46:53<36:20,  3.56it/s]
 98%|█████████▊| 363716/371472 [6:46:53<35:37,  3.63it/s]
 98%|█████████▊| 363717/371472 [6:46:54<38:44,  3.34it/s]
 98%|█████████▊| 363718/371472 [6:46:54<40:32,  3.19it/s]
 98%|█████████▊| 363719/371472 [6:46:54<38:49,  3.33it/s]
 98%|█████████▊| 363720/371472 [6:46:55<38:47,  3.33it/s]
                                                         
{'loss': 2.5588, 'learning_rate': 1.1879161369561942e-07, 'epoch': 15.67}
+
 98%|█████████▊| 363720/371472 [6:46:55<38:47,  3.33it/s]
 98%|█████████▊| 363721/371472 [6:46:55<38:57,  3.32it/s]
 98%|█████████▊| 363722/371472 [6:46:55<39:21,  3.28it/s]
 98%|█████████▊| 363723/371472 [6:46:56<39:25,  3.28it/s]
 98%|█████████▊| 363724/371472 [6:46:56<36:55,  3.50it/s]
 98%|█████████▊| 363725/371472 [6:46:56<37:06,  3.48it/s]
 98%|█████████▊| 363726/371472 [6:46:56<36:46,  3.51it/s]
 98%|█████████▊| 363727/371472 [6:46:57<38:10,  3.38it/s]
 98%|█████████▊| 363728/371472 [6:46:57<37:04,  3.48it/s]
 98%|█████████▊| 363729/371472 [6:46:57<35:50,  3.60it/s]
 98%|█████████▊| 363730/371472 [6:46:58<39:05,  3.30it/s]
 98%|█████████▊| 363731/371472 [6:46:58<37:42,  3.42it/s]
 98%|█████████▊| 363732/371472 [6:46:58<38:09,  3.38it/s]
 98%|█████████▊| 363733/371472 [6:46:58<36:39,  3.52it/s]
 98%|█████████▊| 363734/371472 [6:46:59<36:15,  3.56it/s]
 98%|█████████▊| 363735/371472 [6:46:59<34:50,  3.70it/s]
 98%|█████████▊| 363736/371472 [6:46:59<35:15,  3.66it/s]
 98%|█████████▊| 363737/371472 [6:46:59<35:43,  3.61it/s]
 98%|█████████▊| 363738/371472 [6:47:00<35:31,  3.63it/s]
 98%|█████████▊| 363739/371472 [6:47:00<35:04,  3.67it/s]
 98%|█████████▊| 363740/371472 [6:47:00<36:45,  3.51it/s]
                                                         
{'loss': 2.4459, 'learning_rate': 1.1874313172014047e-07, 'epoch': 15.67}
+
 98%|█████████▊| 363740/371472 [6:47:00<36:45,  3.51it/s]
 98%|█████████▊| 363741/371472 [6:47:01<36:04,  3.57it/s]
 98%|█████████▊| 363742/371472 [6:47:01<35:25,  3.64it/s]
 98%|█████████▊| 363743/371472 [6:47:01<34:16,  3.76it/s]
 98%|█████████▊| 363744/371472 [6:47:01<35:17,  3.65it/s]
 98%|█████████▊| 363745/371472 [6:47:02<34:26,  3.74it/s]
 98%|█████████▊| 363746/371472 [6:47:02<35:36,  3.62it/s]
 98%|█████████▊| 363747/371472 [6:47:02<34:18,  3.75it/s]
 98%|█████████▊| 363748/371472 [6:47:02<34:57,  3.68it/s]
 98%|█████████▊| 363749/371472 [6:47:03<37:03,  3.47it/s]
 98%|█████████▊| 363750/371472 [6:47:03<34:57,  3.68it/s]
 98%|█████████▊| 363751/371472 [6:47:03<34:00,  3.78it/s]
 98%|█████████▊| 363752/371472 [6:47:04<34:37,  3.72it/s]
 98%|█████████▊| 363753/371472 [6:47:04<36:24,  3.53it/s]
 98%|█████████▊| 363754/371472 [6:47:04<34:27,  3.73it/s]
 98%|█████████▊| 363755/371472 [6:47:04<35:05,  3.67it/s]
 98%|█████████▊| 363756/371472 [6:47:05<34:36,  3.72it/s]
 98%|█████████▊| 363757/371472 [6:47:05<35:16,  3.64it/s]
 98%|█████████▊| 363758/371472 [6:47:05<35:26,  3.63it/s]
 98%|█████████▊| 363759/371472 [6:47:05<35:37,  3.61it/s]
 98%|█████████▊| 363760/371472 [6:47:06<35:55,  3.58it/s]
                                                         
{'loss': 2.5437, 'learning_rate': 1.1869464974466163e-07, 'epoch': 15.67}
+
 98%|█████████▊| 363760/371472 [6:47:06<35:55,  3.58it/s]
 98%|█████████▊| 363761/371472 [6:47:06<36:52,  3.49it/s]
 98%|█████████▊| 363762/371472 [6:47:06<37:42,  3.41it/s]
 98%|█████████▊| 363763/371472 [6:47:07<36:28,  3.52it/s]
 98%|█████████▊| 363764/371472 [6:47:07<38:34,  3.33it/s]
 98%|█████████▊| 363765/371472 [6:47:07<36:46,  3.49it/s]
 98%|█████████▊| 363766/371472 [6:47:07<36:14,  3.54it/s]
 98%|█████████▊| 363767/371472 [6:47:08<35:43,  3.59it/s]
 98%|█████████▊| 363768/371472 [6:47:08<35:19,  3.63it/s]
 98%|█████████▊| 363769/371472 [6:47:08<34:36,  3.71it/s]
 98%|█████████▊| 363770/371472 [6:47:09<35:00,  3.67it/s]
 98%|█████████▊| 363771/371472 [6:47:09<36:12,  3.54it/s]
 98%|█████████▊| 363772/371472 [6:47:09<38:30,  3.33it/s]
 98%|█████████▊| 363773/371472 [6:47:09<36:55,  3.47it/s]
 98%|█████████▊| 363774/371472 [6:47:10<37:18,  3.44it/s]
 98%|█████████▊| 363775/371472 [6:47:10<38:52,  3.30it/s]
 98%|█████████▊| 363776/371472 [6:47:10<37:18,  3.44it/s]
 98%|█████████▊| 363777/371472 [6:47:11<36:52,  3.48it/s]
 98%|█████████▊| 363778/371472 [6:47:11<35:06,  3.65it/s]
 98%|█████████▊| 363779/371472 [6:47:11<34:41,  3.70it/s]
 98%|█████████▊| 363780/371472 [6:47:11<34:59,  3.66it/s]
                                                         
{'loss': 2.5921, 'learning_rate': 1.1864616776918269e-07, 'epoch': 15.67}
+
 98%|█████████▊| 363780/371472 [6:47:11<34:59,  3.66it/s]
 98%|█████████▊| 363781/371472 [6:47:12<36:03,  3.56it/s]
 98%|█████████▊| 363782/371472 [6:47:12<36:17,  3.53it/s]
 98%|█████████▊| 363783/371472 [6:47:12<36:04,  3.55it/s]
 98%|█████████▊| 363784/371472 [6:47:13<41:43,  3.07it/s]
 98%|█████████▊| 363785/371472 [6:47:13<38:54,  3.29it/s]
 98%|█████████▊| 363786/371472 [6:47:13<36:35,  3.50it/s]
 98%|█████████▊| 363787/371472 [6:47:13<35:50,  3.57it/s]
 98%|█████████▊| 363788/371472 [6:47:14<35:47,  3.58it/s]
 98%|█████████▊| 363789/371472 [6:47:14<38:17,  3.34it/s]
 98%|█████████▊| 363790/371472 [6:47:14<38:02,  3.37it/s]
 98%|█████████▊| 363791/371472 [6:47:15<36:44,  3.48it/s]
 98%|█████████▊| 363792/371472 [6:47:15<35:38,  3.59it/s]
 98%|█████████▊| 363793/371472 [6:47:15<41:02,  3.12it/s]
 98%|█████████▊| 363794/371472 [6:47:16<37:56,  3.37it/s]
 98%|█████████▊| 363795/371472 [6:47:16<38:39,  3.31it/s]
 98%|█████████▊| 363796/371472 [6:47:16<38:45,  3.30it/s]
 98%|█████████▊| 363797/371472 [6:47:16<37:00,  3.46it/s]
 98%|█████████▊| 363798/371472 [6:47:17<36:48,  3.47it/s]
 98%|█████████▊| 363799/371472 [6:47:17<36:01,  3.55it/s]
 98%|█████████▊| 363800/371472 [6:47:17<34:35,  3.70it/s]
                                                         
{'loss': 2.6403, 'learning_rate': 1.1859768579370384e-07, 'epoch': 15.67}
+
 98%|█████████▊| 363800/371472 [6:47:17<34:35,  3.70it/s]
 98%|█████████▊| 363801/371472 [6:47:18<35:31,  3.60it/s]
 98%|█████████▊| 363802/371472 [6:47:18<36:27,  3.51it/s]
 98%|█████████▊| 363803/371472 [6:47:18<35:40,  3.58it/s]
 98%|█████████▊| 363804/371472 [6:47:19<39:42,  3.22it/s]
 98%|█████████▊| 363805/371472 [6:47:19<41:35,  3.07it/s]
 98%|█████████▊| 363806/371472 [6:47:19<39:41,  3.22it/s]
 98%|█████████▊| 363807/371472 [6:47:19<40:45,  3.13it/s]
 98%|█████████▊| 363808/371472 [6:47:20<41:25,  3.08it/s]
 98%|█████████▊| 363809/371472 [6:47:20<38:56,  3.28it/s]
 98%|█████████▊| 363810/371472 [6:47:20<36:52,  3.46it/s]
 98%|█████████▊| 363811/371472 [6:47:21<36:57,  3.46it/s]
 98%|█████████▊| 363812/371472 [6:47:21<37:32,  3.40it/s]
 98%|█████████▊| 363813/371472 [6:47:21<35:56,  3.55it/s]
 98%|█████████▊| 363814/371472 [6:47:21<35:03,  3.64it/s]
 98%|█████████▊| 363815/371472 [6:47:22<34:31,  3.70it/s]
 98%|█████████▊| 363816/371472 [6:47:22<34:07,  3.74it/s]
 98%|█████████▊| 363817/371472 [6:47:22<33:38,  3.79it/s]
 98%|█████████▊| 363818/371472 [6:47:22<34:41,  3.68it/s]
 98%|█████████▊| 363819/371472 [6:47:23<33:06,  3.85it/s]
 98%|█████████▊| 363820/371472 [6:47:23<33:38,  3.79it/s]
                                                         
{'loss': 2.8078, 'learning_rate': 1.1854920381822491e-07, 'epoch': 15.67}
+
 98%|█████████▊| 363820/371472 [6:47:23<33:38,  3.79it/s]
 98%|█████████▊| 363821/371472 [6:47:23<39:45,  3.21it/s]
 98%|█████████▊| 363822/371472 [6:47:24<38:19,  3.33it/s]
 98%|█████████▊| 363823/371472 [6:47:24<37:19,  3.41it/s]
 98%|█████████▊| 363824/371472 [6:47:24<35:51,  3.56it/s]
 98%|█████████▊| 363825/371472 [6:47:24<34:52,  3.65it/s]
 98%|█████████▊| 363826/371472 [6:47:25<34:18,  3.71it/s]
 98%|█████████▊| 363827/371472 [6:47:25<37:23,  3.41it/s]
 98%|█████████▊| 363828/371472 [6:47:25<37:17,  3.42it/s]
 98%|█████████▊| 363829/371472 [6:47:26<39:29,  3.23it/s]
 98%|█████████▊| 363830/371472 [6:47:26<38:08,  3.34it/s]
 98%|█████████▊| 363831/371472 [6:47:26<39:40,  3.21it/s]
 98%|█████████▊| 363832/371472 [6:47:27<38:44,  3.29it/s]
 98%|█████████▊| 363833/371472 [6:47:27<38:34,  3.30it/s]
 98%|█████████▊| 363834/371472 [6:47:27<37:14,  3.42it/s]
 98%|█████████▊| 363835/371472 [6:47:27<36:14,  3.51it/s]
 98%|█████████▊| 363836/371472 [6:47:28<36:14,  3.51it/s]
 98%|█████████▊| 363837/371472 [6:47:28<35:26,  3.59it/s]
 98%|█████████▊| 363838/371472 [6:47:28<35:21,  3.60it/s]
 98%|█████████▊| 363839/371472 [6:47:29<34:38,  3.67it/s]
 98%|█████████▊| 363840/371472 [6:47:29<33:58,  3.74it/s]
                                                         
{'loss': 2.3763, 'learning_rate': 1.1850072184274596e-07, 'epoch': 15.67}
+
 98%|█████████▊| 363840/371472 [6:47:29<33:58,  3.74it/s]
 98%|█████████▊| 363841/371472 [6:47:29<34:14,  3.71it/s]
 98%|█████████▊| 363842/371472 [6:47:29<33:37,  3.78it/s]
 98%|█████████▊| 363843/371472 [6:47:30<35:18,  3.60it/s]
 98%|█████████▊| 363844/371472 [6:47:30<33:49,  3.76it/s]
 98%|█████████▊| 363845/371472 [6:47:30<34:10,  3.72it/s]
 98%|█████████▊| 363846/371472 [6:47:30<35:56,  3.54it/s]
 98%|█████████▊| 363847/371472 [6:47:31<37:43,  3.37it/s]
 98%|█████████▊| 363848/371472 [6:47:31<37:25,  3.40it/s]
 98%|█████████▊| 363849/371472 [6:47:31<37:59,  3.34it/s]
 98%|█████████▊| 363850/371472 [6:47:32<38:10,  3.33it/s]
 98%|█████████▊| 363851/371472 [6:47:32<36:40,  3.46it/s]
 98%|█████████▊| 363852/371472 [6:47:32<36:42,  3.46it/s]
 98%|█████████▊| 363853/371472 [6:47:33<34:56,  3.63it/s]
 98%|█████████▊| 363854/371472 [6:47:33<35:34,  3.57it/s]
 98%|█████████▊| 363855/371472 [6:47:33<34:59,  3.63it/s]
 98%|█████████▊| 363856/371472 [6:47:33<34:41,  3.66it/s]
 98%|█████████▊| 363857/371472 [6:47:34<33:09,  3.83it/s]
 98%|█████████▊| 363858/371472 [6:47:34<32:55,  3.85it/s]
 98%|█████████▊| 363859/371472 [6:47:34<32:53,  3.86it/s]
 98%|█████████▊| 363860/371472 [6:47:34<34:25,  3.69it/s]
                                                         
{'loss': 2.4756, 'learning_rate': 1.1845223986726712e-07, 'epoch': 15.67}
+
 98%|█████████▊| 363860/371472 [6:47:34<34:25,  3.69it/s]
 98%|█████████▊| 363861/371472 [6:47:35<34:30,  3.68it/s]
 98%|█████████▊| 363862/371472 [6:47:35<34:06,  3.72it/s]
 98%|█████████▊| 363863/371472 [6:47:35<33:49,  3.75it/s]
 98%|█████████▊| 363864/371472 [6:47:35<35:30,  3.57it/s]
 98%|█████████▊| 363865/371472 [6:47:36<35:44,  3.55it/s]
 98%|█████████▊| 363866/371472 [6:47:36<35:25,  3.58it/s]
 98%|█████████▊| 363867/371472 [6:47:36<35:13,  3.60it/s]
 98%|█████████▊| 363868/371472 [6:47:37<34:51,  3.64it/s]
 98%|█████████▊| 363869/371472 [6:47:37<34:45,  3.65it/s]
 98%|█████████▊| 363870/371472 [6:47:37<35:23,  3.58it/s]
 98%|█████████▊| 363871/371472 [6:47:37<35:24,  3.58it/s]
 98%|█████████▊| 363872/371472 [6:47:38<34:48,  3.64it/s]
 98%|█████████▊| 363873/371472 [6:47:38<34:05,  3.72it/s]
 98%|█████████▊| 363874/371472 [6:47:38<35:58,  3.52it/s]
 98%|█████████▊| 363875/371472 [6:47:39<35:19,  3.58it/s]
 98%|█████████▊| 363876/371472 [6:47:39<35:08,  3.60it/s]
 98%|█████████▊| 363877/371472 [6:47:39<35:18,  3.58it/s]
 98%|█████████▊| 363878/371472 [6:47:39<34:16,  3.69it/s]
 98%|█████████▊| 363879/371472 [6:47:40<36:01,  3.51it/s]
 98%|█████████▊| 363880/371472 [6:47:40<35:39,  3.55it/s]
                                                         
{'loss': 2.4845, 'learning_rate': 1.1840375789178818e-07, 'epoch': 15.67}
+
 98%|█████████▊| 363880/371472 [6:47:40<35:39,  3.55it/s]
 98%|█████████▊| 363881/371472 [6:47:40<34:46,  3.64it/s]
 98%|█████████▊| 363882/371472 [6:47:40<35:16,  3.59it/s]
 98%|█████████▊| 363883/371472 [6:47:41<34:53,  3.62it/s]
 98%|█████████▊| 363884/371472 [6:47:41<33:34,  3.77it/s]
 98%|█████████▊| 363885/371472 [6:47:41<32:55,  3.84it/s]
 98%|█████████▊| 363886/371472 [6:47:42<35:06,  3.60it/s]
 98%|█████████▊| 363887/371472 [6:47:42<35:46,  3.53it/s]
 98%|█████████▊| 363888/371472 [6:47:42<35:37,  3.55it/s]
 98%|█████████▊| 363889/371472 [6:47:42<36:06,  3.50it/s]
 98%|█████████▊| 363890/371472 [6:47:43<34:03,  3.71it/s]
 98%|█████████▊| 363891/371472 [6:47:43<35:21,  3.57it/s]
 98%|█████████▊| 363892/371472 [6:47:43<34:22,  3.67it/s]
 98%|█████████▊| 363893/371472 [6:47:44<35:16,  3.58it/s]
 98%|█████████▊| 363894/371472 [6:47:44<36:55,  3.42it/s]
 98%|█████████▊| 363895/371472 [6:47:44<35:35,  3.55it/s]
 98%|█████████▊| 363896/371472 [6:47:44<34:49,  3.63it/s]
 98%|█████████▊| 363897/371472 [6:47:45<34:36,  3.65it/s]
 98%|█████████▊| 363898/371472 [6:47:45<34:14,  3.69it/s]
 98%|█████████▊| 363899/371472 [6:47:45<33:46,  3.74it/s]
 98%|█████████▊| 363900/371472 [6:47:45<33:47,  3.73it/s]
                                                         
{'loss': 2.648, 'learning_rate': 1.1835527591630933e-07, 'epoch': 15.67}
+
 98%|█████████▊| 363900/371472 [6:47:45<33:47,  3.73it/s]
 98%|█████████▊| 363901/371472 [6:47:46<33:35,  3.76it/s]
 98%|█████████▊| 363902/371472 [6:47:46<33:18,  3.79it/s]
 98%|█████████▊| 363903/371472 [6:47:46<34:02,  3.71it/s]
 98%|█████████▊| 363904/371472 [6:47:46<33:36,  3.75it/s]
 98%|█████████▊| 363905/371472 [6:47:47<32:54,  3.83it/s]
 98%|█████████▊| 363906/371472 [6:47:47<32:10,  3.92it/s]
 98%|█████████▊| 363907/371472 [6:47:47<33:33,  3.76it/s]
 98%|█████████▊| 363908/371472 [6:47:48<33:13,  3.79it/s]
 98%|█████████▊| 363909/371472 [6:47:48<34:25,  3.66it/s]
 98%|█████████▊| 363910/371472 [6:47:48<38:39,  3.26it/s]
 98%|█████████▊| 363911/371472 [6:47:48<37:16,  3.38it/s]
 98%|█████████▊| 363912/371472 [6:47:49<38:14,  3.29it/s]
 98%|█████████▊| 363913/371472 [6:47:49<37:15,  3.38it/s]
 98%|█████████▊| 363914/371472 [6:47:49<37:30,  3.36it/s]
 98%|█████████▊| 363915/371472 [6:47:50<38:21,  3.28it/s]
 98%|█████████▊| 363916/371472 [6:47:50<37:39,  3.34it/s]
 98%|█████████▊| 363917/371472 [6:47:50<37:28,  3.36it/s]
 98%|█████████▊| 363918/371472 [6:47:51<36:41,  3.43it/s]
 98%|█████████▊| 363919/371472 [6:47:51<35:47,  3.52it/s]
 98%|█████████▊| 363920/371472 [6:47:51<37:24,  3.36it/s]
                                                         
{'loss': 2.6715, 'learning_rate': 1.183067939408304e-07, 'epoch': 15.67}
+
 98%|█████████▊| 363920/371472 [6:47:51<37:24,  3.36it/s]
 98%|█████████▊| 363921/371472 [6:47:51<37:30,  3.36it/s]
 98%|█████████▊| 363922/371472 [6:47:52<38:16,  3.29it/s]
 98%|█████████▊| 363923/371472 [6:47:52<38:37,  3.26it/s]
 98%|█████████▊| 363924/371472 [6:47:52<39:59,  3.15it/s]
 98%|█████████▊| 363925/371472 [6:47:53<39:08,  3.21it/s]
 98%|█████████▊| 363926/371472 [6:47:53<41:45,  3.01it/s]
 98%|█████████▊| 363927/371472 [6:47:53<40:07,  3.13it/s]
 98%|█████████▊| 363928/371472 [6:47:54<37:40,  3.34it/s]
 98%|█████████▊| 363929/371472 [6:47:54<35:36,  3.53it/s]
 98%|█████████▊| 363930/371472 [6:47:54<37:57,  3.31it/s]
 98%|█████████▊| 363931/371472 [6:47:54<36:12,  3.47it/s]
 98%|█████████▊| 363932/371472 [6:47:55<34:18,  3.66it/s]
 98%|█████████▊| 363933/371472 [6:47:55<33:53,  3.71it/s]
 98%|█████████▊| 363934/371472 [6:47:55<34:34,  3.63it/s]
 98%|█████████▊| 363935/371472 [6:47:56<33:24,  3.76it/s]
 98%|█████████▊| 363936/371472 [6:47:56<34:52,  3.60it/s]
 98%|█████████▊| 363937/371472 [6:47:56<35:18,  3.56it/s]
 98%|█████████▊| 363938/371472 [6:47:56<35:39,  3.52it/s]
 98%|█████████▊| 363939/371472 [6:47:57<39:49,  3.15it/s]
 98%|█████████▊| 363940/371472 [6:47:57<39:14,  3.20it/s]
                                                         
{'loss': 2.5261, 'learning_rate': 1.1825831196535155e-07, 'epoch': 15.68}
+
 98%|█████████▊| 363940/371472 [6:47:57<39:14,  3.20it/s]
 98%|█████████▊| 363941/371472 [6:47:57<37:13,  3.37it/s]
 98%|█████████▊| 363942/371472 [6:47:58<36:39,  3.42it/s]
 98%|█████████▊| 363943/371472 [6:47:58<35:26,  3.54it/s]
 98%|█████████▊| 363944/371472 [6:47:58<35:40,  3.52it/s]
 98%|█████████▊| 363945/371472 [6:47:58<35:54,  3.49it/s]
 98%|█████████▊| 363946/371472 [6:47:59<35:15,  3.56it/s]
 98%|█████████▊| 363947/371472 [6:47:59<34:05,  3.68it/s]
 98%|█████████▊| 363948/371472 [6:47:59<34:11,  3.67it/s]
 98%|█████████▊| 363949/371472 [6:48:00<33:57,  3.69it/s]
 98%|█████████▊| 363950/371472 [6:48:00<34:13,  3.66it/s]
 98%|█████████▊| 363951/371472 [6:48:00<33:43,  3.72it/s]
 98%|█████████▊| 363952/371472 [6:48:00<32:58,  3.80it/s]
 98%|█████████▊| 363953/371472 [6:48:01<39:24,  3.18it/s]
 98%|█████████▊| 363954/371472 [6:48:01<37:22,  3.35it/s]
 98%|█████████▊| 363955/371472 [6:48:01<37:08,  3.37it/s]
 98%|█████████▊| 363956/371472 [6:48:02<35:20,  3.54it/s]
 98%|█████████▊| 363957/371472 [6:48:02<35:20,  3.54it/s]
 98%|█████████▊| 363958/371472 [6:48:02<34:43,  3.61it/s]
 98%|█████████▊| 363959/371472 [6:48:02<35:15,  3.55it/s]
 98%|█████████▊| 363960/371472 [6:48:03<34:03,  3.68it/s]
                                                         
{'loss': 2.5999, 'learning_rate': 1.1820982998987261e-07, 'epoch': 15.68}
+
 98%|█████████▊| 363960/371472 [6:48:03<34:03,  3.68it/s]
 98%|█████████▊| 363961/371472 [6:48:03<35:43,  3.50it/s]
 98%|█████████▊| 363962/371472 [6:48:03<35:41,  3.51it/s]
 98%|█████████▊| 363963/371472 [6:48:04<35:00,  3.57it/s]
 98%|█████████▊| 363964/371472 [6:48:04<37:35,  3.33it/s]
 98%|█████████▊| 363965/371472 [6:48:04<35:39,  3.51it/s]
 98%|█████████▊| 363966/371472 [6:48:04<34:55,  3.58it/s]
 98%|█████████▊| 363967/371472 [6:48:05<33:50,  3.70it/s]
 98%|█████████▊| 363968/371472 [6:48:05<33:59,  3.68it/s]
 98%|█████████▊| 363969/371472 [6:48:05<33:16,  3.76it/s]
 98%|█████████▊| 363970/371472 [6:48:05<34:28,  3.63it/s]
 98%|█████████▊| 363971/371472 [6:48:06<33:57,  3.68it/s]
 98%|█████████▊| 363972/371472 [6:48:06<36:13,  3.45it/s]
 98%|█████████▊| 363973/371472 [6:48:06<38:17,  3.26it/s]
 98%|█████████▊| 363974/371472 [6:48:07<36:04,  3.46it/s]
 98%|█████████▊| 363975/371472 [6:48:07<37:00,  3.38it/s]
 98%|█████████▊| 363976/371472 [6:48:07<35:06,  3.56it/s]
 98%|█████████▊| 363977/371472 [6:48:07<33:59,  3.67it/s]
 98%|█████████▊| 363978/371472 [6:48:08<34:53,  3.58it/s]
 98%|█████████▊| 363979/371472 [6:48:08<34:44,  3.59it/s]
 98%|█████████▊| 363980/371472 [6:48:08<33:48,  3.69it/s]
                                                         
{'loss': 2.6795, 'learning_rate': 1.1816134801439378e-07, 'epoch': 15.68}
+
 98%|█████████▊| 363980/371472 [6:48:08<33:48,  3.69it/s]
 98%|█████████▊| 363981/371472 [6:48:09<34:49,  3.59it/s]
 98%|█████████▊| 363982/371472 [6:48:09<35:20,  3.53it/s]
 98%|█████████▊| 363983/371472 [6:48:09<39:22,  3.17it/s]
 98%|█████████▊| 363984/371472 [6:48:10<40:57,  3.05it/s]
 98%|█████████▊| 363985/371472 [6:48:10<39:21,  3.17it/s]
 98%|█████████▊| 363986/371472 [6:48:10<39:38,  3.15it/s]
 98%|█████████▊| 363987/371472 [6:48:10<37:33,  3.32it/s]
 98%|█████████▊| 363988/371472 [6:48:11<38:05,  3.27it/s]
 98%|█████████▊| 363989/371472 [6:48:11<38:27,  3.24it/s]
 98%|█████████▊| 363990/371472 [6:48:11<38:02,  3.28it/s]
 98%|█████████▊| 363991/371472 [6:48:12<36:12,  3.44it/s]
 98%|█████████▊| 363992/371472 [6:48:12<34:35,  3.60it/s]
 98%|█████████▊| 363993/371472 [6:48:12<37:47,  3.30it/s]
 98%|█████████▊| 363994/371472 [6:48:13<37:08,  3.36it/s]
 98%|█████████▊| 363995/371472 [6:48:13<35:38,  3.50it/s]
 98%|█████████▊| 363996/371472 [6:48:13<35:38,  3.50it/s]
 98%|█████████▊| 363997/371472 [6:48:13<35:11,  3.54it/s]
 98%|█████████▊| 363998/371472 [6:48:14<34:47,  3.58it/s]
 98%|█████████▊| 363999/371472 [6:48:14<35:06,  3.55it/s]
 98%|█████████▊| 364000/371472 [6:48:14<34:38,  3.59it/s]
                                                         
{'loss': 2.4338, 'learning_rate': 1.1811286603891482e-07, 'epoch': 15.68}
+
 98%|█████████▊| 364000/371472 [6:48:14<34:38,  3.59it/s]
 98%|█████████▊| 364001/371472 [6:48:14<34:02,  3.66it/s]
 98%|█████████▊| 364002/371472 [6:48:15<34:09,  3.64it/s]
 98%|█████████▊| 364003/371472 [6:48:15<33:05,  3.76it/s]
 98%|█████████▊| 364004/371472 [6:48:15<33:40,  3.70it/s]
 98%|█████████▊| 364005/371472 [6:48:16<34:02,  3.66it/s]
 98%|█████████▊| 364006/371472 [6:48:16<33:00,  3.77it/s]
 98%|█████████▊| 364007/371472 [6:48:16<32:56,  3.78it/s]
 98%|█████████▊| 364008/371472 [6:48:16<33:39,  3.70it/s]
 98%|█████████▊| 364009/371472 [6:48:17<33:15,  3.74it/s]
 98%|█████████▊| 364010/371472 [6:48:17<32:21,  3.84it/s]
 98%|█████████▊| 364011/371472 [6:48:17<32:00,  3.88it/s]
 98%|█████████▊| 364012/371472 [6:48:17<31:53,  3.90it/s]
 98%|█████████▊| 364013/371472 [6:48:18<36:07,  3.44it/s]
 98%|█████████▊| 364014/371472 [6:48:18<37:02,  3.36it/s]
 98%|█████████▊| 364015/371472 [6:48:18<35:44,  3.48it/s]
 98%|█████████▊| 364016/371472 [6:48:19<34:57,  3.56it/s]
 98%|█████████▊| 364017/371472 [6:48:19<34:36,  3.59it/s]
 98%|█████████▊| 364018/371472 [6:48:19<35:31,  3.50it/s]
 98%|█████████▊| 364019/371472 [6:48:19<35:46,  3.47it/s]
 98%|█████████▊| 364020/371472 [6:48:20<34:41,  3.58it/s]
                                                         
{'loss': 2.5579, 'learning_rate': 1.1806438406343599e-07, 'epoch': 15.68}
+
 98%|█████████▊| 364020/371472 [6:48:20<34:41,  3.58it/s]
 98%|█████████▊| 364021/371472 [6:48:20<35:26,  3.50it/s]
 98%|█████████▊| 364022/371472 [6:48:20<33:34,  3.70it/s]
 98%|█████████▊| 364023/371472 [6:48:21<33:47,  3.67it/s]
 98%|█████████▊| 364024/371472 [6:48:21<32:51,  3.78it/s]
 98%|█████████▊| 364025/371472 [6:48:21<32:24,  3.83it/s]
 98%|█████████▊| 364026/371472 [6:48:21<35:01,  3.54it/s]
 98%|█████████▊| 364027/371472 [6:48:22<35:10,  3.53it/s]
 98%|█████████▊| 364028/371472 [6:48:22<34:20,  3.61it/s]
 98%|█████████▊| 364029/371472 [6:48:22<35:04,  3.54it/s]
 98%|█████████▊| 364030/371472 [6:48:22<33:51,  3.66it/s]
 98%|█████████▊| 364031/371472 [6:48:23<32:55,  3.77it/s]
 98%|█████████▊| 364032/371472 [6:48:23<36:35,  3.39it/s]
 98%|█████████▊| 364033/371472 [6:48:23<35:49,  3.46it/s]
 98%|█████████▊| 364034/371472 [6:48:24<37:11,  3.33it/s]
 98%|█████████▊| 364035/371472 [6:48:24<35:42,  3.47it/s]
 98%|█████████▊| 364036/371472 [6:48:24<36:31,  3.39it/s]
 98%|█████████▊| 364037/371472 [6:48:24<34:57,  3.54it/s]
 98%|█████████▊| 364038/371472 [6:48:25<34:53,  3.55it/s]
 98%|█████████▊| 364039/371472 [6:48:25<34:53,  3.55it/s]
 98%|█████████▊| 364040/371472 [6:48:25<33:49,  3.66it/s]
                                                         
{'loss': 2.3911, 'learning_rate': 1.1801590208795704e-07, 'epoch': 15.68}
+
 98%|█████████▊| 364040/371472 [6:48:25<33:49,  3.66it/s]
 98%|█████████▊| 364041/371472 [6:48:26<32:51,  3.77it/s]
 98%|█████████▊| 364042/371472 [6:48:26<33:00,  3.75it/s]
 98%|█████████▊| 364043/371472 [6:48:26<35:23,  3.50it/s]
 98%|█████████▊| 364044/371472 [6:48:26<34:55,  3.54it/s]
 98%|█████████▊| 364045/371472 [6:48:27<38:41,  3.20it/s]
 98%|█████████▊| 364046/371472 [6:48:27<37:32,  3.30it/s]
 98%|█████████▊| 364047/371472 [6:48:27<35:19,  3.50it/s]
 98%|█████████▊| 364048/371472 [6:48:28<34:50,  3.55it/s]
 98%|█████████▊| 364049/371472 [6:48:28<34:32,  3.58it/s]
 98%|█████████▊| 364050/371472 [6:48:28<33:57,  3.64it/s]
 98%|█████████▊| 364051/371472 [6:48:28<35:24,  3.49it/s]
 98%|█████████▊| 364052/371472 [6:48:29<33:42,  3.67it/s]
 98%|█████████▊| 364053/371472 [6:48:29<35:07,  3.52it/s]
 98%|█████████▊| 364054/371472 [6:48:29<34:43,  3.56it/s]
 98%|█████████▊| 364055/371472 [6:48:30<34:07,  3.62it/s]
 98%|█████████▊| 364056/371472 [6:48:30<33:24,  3.70it/s]
 98%|█████████▊| 364057/371472 [6:48:30<33:06,  3.73it/s]
 98%|█████████▊| 364058/371472 [6:48:30<34:35,  3.57it/s]
 98%|█████████▊| 364059/371472 [6:48:31<33:35,  3.68it/s]
 98%|█████████▊| 364060/371472 [6:48:31<37:38,  3.28it/s]
                                                         
{'loss': 2.5988, 'learning_rate': 1.179674201124782e-07, 'epoch': 15.68}
+
 98%|█████████▊| 364060/371472 [6:48:31<37:38,  3.28it/s]
 98%|█████████▊| 364061/371472 [6:48:31<36:04,  3.42it/s]
 98%|█████████▊| 364062/371472 [6:48:32<34:34,  3.57it/s]
 98%|█████████▊| 364063/371472 [6:48:32<34:17,  3.60it/s]
 98%|█████████▊| 364064/371472 [6:48:32<33:56,  3.64it/s]
 98%|█████████▊| 364065/371472 [6:48:32<34:10,  3.61it/s]
 98%|█████████▊| 364066/371472 [6:48:33<33:08,  3.72it/s]
 98%|█████████▊| 364067/371472 [6:48:33<33:17,  3.71it/s]
 98%|█████████▊| 364068/371472 [6:48:33<34:01,  3.63it/s]
 98%|█████████▊| 364069/371472 [6:48:33<33:34,  3.67it/s]
 98%|█████████▊| 364070/371472 [6:48:34<35:19,  3.49it/s]
 98%|█████████▊| 364071/371472 [6:48:34<36:31,  3.38it/s]
 98%|█████████▊| 364072/371472 [6:48:34<36:13,  3.41it/s]
 98%|█████████▊| 364073/371472 [6:48:35<37:28,  3.29it/s]
 98%|█████████▊| 364074/371472 [6:48:35<34:58,  3.53it/s]
 98%|█████████▊| 364075/371472 [6:48:35<33:43,  3.66it/s]
 98%|█████████▊| 364076/371472 [6:48:35<34:23,  3.58it/s]
 98%|█████████▊| 364077/371472 [6:48:36<34:43,  3.55it/s]
 98%|█████████▊| 364078/371472 [6:48:36<34:31,  3.57it/s]
 98%|█████████▊| 364079/371472 [6:48:36<35:43,  3.45it/s]
 98%|█████████▊| 364080/371472 [6:48:37<35:40,  3.45it/s]
                                                         
{'loss': 2.6504, 'learning_rate': 1.1791893813699925e-07, 'epoch': 15.68}
+
 98%|█████████▊| 364080/371472 [6:48:37<35:40,  3.45it/s]
 98%|█████████▊| 364081/371472 [6:48:37<33:32,  3.67it/s]
 98%|█████████▊| 364082/371472 [6:48:37<32:07,  3.83it/s]
 98%|█████████▊| 364083/371472 [6:48:37<32:19,  3.81it/s]
 98%|█████████▊| 364084/371472 [6:48:38<32:46,  3.76it/s]
 98%|█████████▊| 364085/371472 [6:48:38<33:13,  3.71it/s]
 98%|█████████▊| 364086/371472 [6:48:38<35:36,  3.46it/s]
 98%|█████████▊| 364087/371472 [6:48:38<34:17,  3.59it/s]
 98%|█████████▊| 364088/371472 [6:48:39<34:29,  3.57it/s]
 98%|█████████▊| 364089/371472 [6:48:39<34:07,  3.61it/s]
 98%|█████████▊| 364090/371472 [6:48:39<33:35,  3.66it/s]
 98%|█████████▊| 364091/371472 [6:48:40<33:37,  3.66it/s]
 98%|█████████▊| 364092/371472 [6:48:40<33:34,  3.66it/s]
 98%|█████████▊| 364093/371472 [6:48:40<37:23,  3.29it/s]
 98%|█████████▊| 364094/371472 [6:48:41<36:09,  3.40it/s]
 98%|█████████▊| 364095/371472 [6:48:41<38:17,  3.21it/s]
 98%|█████████▊| 364096/371472 [6:48:41<38:34,  3.19it/s]
 98%|█████████▊| 364097/371472 [6:48:41<38:18,  3.21it/s]
 98%|█████████▊| 364098/371472 [6:48:42<37:11,  3.30it/s]
 98%|█████████▊| 364099/371472 [6:48:42<38:17,  3.21it/s]
 98%|█████████▊| 364100/371472 [6:48:42<36:59,  3.32it/s]
                                                         
{'loss': 2.5545, 'learning_rate': 1.1787045616152042e-07, 'epoch': 15.68}
+
 98%|█████████▊| 364100/371472 [6:48:42<36:59,  3.32it/s]
 98%|█████████▊| 364101/371472 [6:48:43<35:14,  3.49it/s]
 98%|█████████▊| 364102/371472 [6:48:43<34:36,  3.55it/s]
 98%|█████████▊| 364103/371472 [6:48:43<34:21,  3.57it/s]
 98%|█████████▊| 364104/371472 [6:48:43<35:13,  3.49it/s]
 98%|█████████▊| 364105/371472 [6:48:44<34:24,  3.57it/s]
 98%|█████████▊| 364106/371472 [6:48:44<33:15,  3.69it/s]
 98%|█████████▊| 364107/371472 [6:48:44<32:16,  3.80it/s]
 98%|█████████▊| 364108/371472 [6:48:44<32:05,  3.82it/s]
 98%|█████████▊| 364109/371472 [6:48:45<34:32,  3.55it/s]
 98%|█████████▊| 364110/371472 [6:48:45<34:15,  3.58it/s]
 98%|█████████▊| 364111/371472 [6:48:45<33:48,  3.63it/s]
 98%|█████████▊| 364112/371472 [6:48:46<35:59,  3.41it/s]
 98%|█████████▊| 364113/371472 [6:48:46<34:22,  3.57it/s]
 98%|█████████▊| 364114/371472 [6:48:46<33:34,  3.65it/s]
 98%|█████████▊| 364115/371472 [6:48:47<37:00,  3.31it/s]
 98%|█████████▊| 364116/371472 [6:48:47<35:39,  3.44it/s]
 98%|█████████▊| 364117/371472 [6:48:47<34:29,  3.55it/s]
 98%|█████████▊| 364118/371472 [6:48:47<34:27,  3.56it/s]
 98%|█████████▊| 364119/371472 [6:48:48<33:31,  3.66it/s]
 98%|█████████▊| 364120/371472 [6:48:48<32:32,  3.76it/s]
                                                         
{'loss': 2.7996, 'learning_rate': 1.1782197418604148e-07, 'epoch': 15.68}
+
 98%|█████████▊| 364120/371472 [6:48:48<32:32,  3.76it/s]
 98%|█████████▊| 364121/371472 [6:48:48<32:18,  3.79it/s]
 98%|█████████▊| 364122/371472 [6:48:48<31:57,  3.83it/s]
 98%|█████████▊| 364123/371472 [6:48:49<31:54,  3.84it/s]
 98%|█████████▊| 364124/371472 [6:48:49<36:27,  3.36it/s]
 98%|█████████▊| 364125/371472 [6:48:49<35:12,  3.48it/s]
 98%|█████████▊| 364126/371472 [6:48:50<35:48,  3.42it/s]
 98%|█████████▊| 364127/371472 [6:48:50<35:41,  3.43it/s]
 98%|█████████▊| 364128/371472 [6:48:50<36:00,  3.40it/s]
 98%|█████████▊| 364129/371472 [6:48:50<34:43,  3.52it/s]
 98%|█████████▊| 364130/371472 [6:48:51<34:10,  3.58it/s]
 98%|█████████▊| 364131/371472 [6:48:51<35:21,  3.46it/s]
 98%|█████████▊| 364132/371472 [6:48:51<34:11,  3.58it/s]
 98%|█████████▊| 364133/371472 [6:48:52<33:51,  3.61it/s]
 98%|█████████▊| 364134/371472 [6:48:52<33:28,  3.65it/s]
 98%|█████████▊| 364135/371472 [6:48:52<34:16,  3.57it/s]
 98%|█████████▊| 364136/371472 [6:48:52<33:05,  3.69it/s]
 98%|█████████▊| 364137/371472 [6:48:53<33:04,  3.70it/s]
 98%|█████████▊| 364138/371472 [6:48:53<33:41,  3.63it/s]
 98%|█████████▊| 364139/371472 [6:48:53<34:03,  3.59it/s]
 98%|█████████▊| 364140/371472 [6:48:53<32:59,  3.70it/s]
                                                         
{'loss': 2.5878, 'learning_rate': 1.1777349221056263e-07, 'epoch': 15.68}
+
 98%|█████████▊| 364140/371472 [6:48:53<32:59,  3.70it/s]
 98%|█████████▊| 364141/371472 [6:48:54<32:27,  3.76it/s]
 98%|█████████▊| 364142/371472 [6:48:54<32:52,  3.72it/s]
 98%|█████████▊| 364143/371472 [6:48:54<33:08,  3.69it/s]
 98%|█████████▊| 364144/371472 [6:48:55<33:38,  3.63it/s]
 98%|█████████▊| 364145/371472 [6:48:55<33:33,  3.64it/s]
 98%|█████████▊| 364146/371472 [6:48:55<35:24,  3.45it/s]
 98%|█████████▊| 364147/371472 [6:48:55<34:51,  3.50it/s]
 98%|█████████▊| 364148/371472 [6:48:56<34:23,  3.55it/s]
 98%|█████████▊| 364149/371472 [6:48:56<33:34,  3.63it/s]
 98%|█████████▊| 364150/371472 [6:48:56<36:15,  3.36it/s]
 98%|█████████▊| 364151/371472 [6:48:57<35:20,  3.45it/s]
 98%|█████████▊| 364152/371472 [6:48:57<37:56,  3.22it/s]
 98%|█████████▊| 364153/371472 [6:48:57<37:13,  3.28it/s]
 98%|█████████▊| 364154/371472 [6:48:58<36:45,  3.32it/s]
 98%|█████████▊| 364155/371472 [6:48:58<35:50,  3.40it/s]
 98%|█████████▊| 364156/371472 [6:48:58<36:00,  3.39it/s]
 98%|█████████▊| 364157/371472 [6:48:58<37:00,  3.29it/s]
 98%|█████████▊| 364158/371472 [6:48:59<36:07,  3.37it/s]
 98%|█████████▊| 364159/371472 [6:48:59<36:16,  3.36it/s]
 98%|█████████▊| 364160/371472 [6:48:59<34:56,  3.49it/s]
                                                         
{'loss': 2.4243, 'learning_rate': 1.1772501023508368e-07, 'epoch': 15.69}
+
 98%|█████████▊| 364160/371472 [6:48:59<34:56,  3.49it/s]
 98%|█████████▊| 364161/371472 [6:49:00<33:51,  3.60it/s]
 98%|█████████▊| 364162/371472 [6:49:00<35:06,  3.47it/s]
 98%|█████████▊| 364163/371472 [6:49:00<34:36,  3.52it/s]
 98%|█████████▊| 364164/371472 [6:49:00<35:44,  3.41it/s]
 98%|█████████▊| 364165/371472 [6:49:01<36:14,  3.36it/s]
 98%|█████████▊| 364166/371472 [6:49:01<34:34,  3.52it/s]
 98%|█████████▊| 364167/371472 [6:49:01<34:48,  3.50it/s]
 98%|█████████▊| 364168/371472 [6:49:02<33:22,  3.65it/s]
 98%|█████████▊| 364169/371472 [6:49:02<36:10,  3.36it/s]
 98%|█████████▊| 364170/371472 [6:49:02<37:54,  3.21it/s]
 98%|█████████▊| 364171/371472 [6:49:02<35:01,  3.47it/s]
 98%|█████████▊| 364172/371472 [6:49:03<33:23,  3.64it/s]
 98%|█████████▊| 364173/371472 [6:49:03<35:17,  3.45it/s]
 98%|█████████▊| 364174/371472 [6:49:03<35:47,  3.40it/s]
 98%|█████████▊| 364175/371472 [6:49:04<36:34,  3.33it/s]
 98%|█████████▊| 364176/371472 [6:49:04<35:18,  3.44it/s]
 98%|█████████▊| 364177/371472 [6:49:04<35:01,  3.47it/s]
 98%|█████████▊| 364178/371472 [6:49:05<39:12,  3.10it/s]
 98%|█████████▊| 364179/371472 [6:49:05<37:34,  3.23it/s]
 98%|█████████▊| 364180/371472 [6:49:05<35:12,  3.45it/s]
                                                         
{'loss': 2.533, 'learning_rate': 1.1767652825960485e-07, 'epoch': 15.69}
+
 98%|█████████▊| 364180/371472 [6:49:05<35:12,  3.45it/s]
 98%|█████████▊| 364181/371472 [6:49:05<34:51,  3.49it/s]
 98%|█████████▊| 364182/371472 [6:49:06<34:04,  3.57it/s]
 98%|█████████▊| 364183/371472 [6:49:06<34:48,  3.49it/s]
 98%|█████████▊| 364184/371472 [6:49:06<35:44,  3.40it/s]
 98%|█████████▊| 364185/371472 [6:49:07<35:45,  3.40it/s]
 98%|█████████▊| 364186/371472 [6:49:07<35:30,  3.42it/s]
 98%|█████████▊| 364187/371472 [6:49:07<33:45,  3.60it/s]
 98%|█████████▊| 364188/371472 [6:49:07<37:02,  3.28it/s]
 98%|█████████▊| 364189/371472 [6:49:08<35:52,  3.38it/s]
 98%|█████████▊| 364190/371472 [6:49:08<37:11,  3.26it/s]
 98%|█████████▊| 364191/371472 [6:49:08<35:30,  3.42it/s]
 98%|█████████▊| 364192/371472 [6:49:09<36:52,  3.29it/s]
 98%|█████████▊| 364193/371472 [6:49:09<35:10,  3.45it/s]
 98%|█████████▊| 364194/371472 [6:49:09<34:18,  3.54it/s]
 98%|█████████▊| 364195/371472 [6:49:10<35:31,  3.41it/s]
 98%|█████████▊| 364196/371472 [6:49:10<35:47,  3.39it/s]
 98%|█████████▊| 364197/371472 [6:49:10<33:31,  3.62it/s]
 98%|█████████▊| 364198/371472 [6:49:10<33:56,  3.57it/s]
 98%|█████████▊| 364199/371472 [6:49:11<35:15,  3.44it/s]
 98%|█████████▊| 364200/371472 [6:49:11<36:06,  3.36it/s]
                                                         
{'loss': 2.6663, 'learning_rate': 1.1762804628412591e-07, 'epoch': 15.69}
+
 98%|█████████▊| 364200/371472 [6:49:11<36:06,  3.36it/s]
 98%|█████████▊| 364201/371472 [6:49:11<35:58,  3.37it/s]
 98%|█████████▊| 364202/371472 [6:49:12<34:58,  3.46it/s]
 98%|█████████▊| 364203/371472 [6:49:12<35:14,  3.44it/s]
 98%|█████████▊| 364204/371472 [6:49:12<33:36,  3.60it/s]
 98%|█████████▊| 364205/371472 [6:49:12<34:58,  3.46it/s]
 98%|█████████▊| 364206/371472 [6:49:13<34:02,  3.56it/s]
 98%|█████████▊| 364207/371472 [6:49:13<33:08,  3.65it/s]
 98%|█████████▊| 364208/371472 [6:49:13<34:17,  3.53it/s]
 98%|█████████▊| 364209/371472 [6:49:14<35:33,  3.40it/s]
 98%|█████████▊| 364210/371472 [6:49:14<34:46,  3.48it/s]
 98%|█████████▊| 364211/371472 [6:49:14<35:02,  3.45it/s]
 98%|█████████▊| 364212/371472 [6:49:14<34:32,  3.50it/s]
 98%|█████████▊| 364213/371472 [6:49:15<33:12,  3.64it/s]
 98%|█████████▊| 364214/371472 [6:49:15<32:02,  3.77it/s]
 98%|█████████▊| 364215/371472 [6:49:15<31:27,  3.84it/s]
 98%|█████████▊| 364216/371472 [6:49:15<31:58,  3.78it/s]
 98%|█████████▊| 364217/371472 [6:49:16<32:09,  3.76it/s]
 98%|█████████▊| 364218/371472 [6:49:16<33:10,  3.64it/s]
 98%|█████████▊| 364219/371472 [6:49:16<34:33,  3.50it/s]
 98%|█████████▊| 364220/371472 [6:49:17<35:29,  3.41it/s]
                                                         
{'loss': 2.6595, 'learning_rate': 1.1757956430864706e-07, 'epoch': 15.69}
+
 98%|█████████▊| 364220/371472 [6:49:17<35:29,  3.41it/s]
 98%|█████████▊| 364221/371472 [6:49:17<34:10,  3.54it/s]
 98%|█████████▊| 364222/371472 [6:49:17<33:24,  3.62it/s]
 98%|█████████▊| 364223/371472 [6:49:17<32:50,  3.68it/s]
 98%|█████████▊| 364224/371472 [6:49:18<31:17,  3.86it/s]
 98%|█████████▊| 364225/371472 [6:49:18<31:15,  3.86it/s]
 98%|█████████▊| 364226/371472 [6:49:18<31:33,  3.83it/s]
 98%|█████████▊| 364227/371472 [6:49:18<32:15,  3.74it/s]
 98%|█████████▊| 364228/371472 [6:49:19<35:03,  3.44it/s]
 98%|█████████▊| 364229/371472 [6:49:19<37:25,  3.23it/s]
 98%|█████████▊| 364230/371472 [6:49:19<36:59,  3.26it/s]
 98%|█████████▊| 364231/371472 [6:49:20<35:35,  3.39it/s]
 98%|█████████▊| 364232/371472 [6:49:20<33:44,  3.58it/s]
 98%|█████████▊| 364233/371472 [6:49:20<33:10,  3.64it/s]
 98%|█████████▊| 364234/371472 [6:49:20<32:43,  3.69it/s]
 98%|█████████▊| 364235/371472 [6:49:21<33:42,  3.58it/s]
 98%|█████████▊| 364236/371472 [6:49:21<32:18,  3.73it/s]
 98%|█████████▊| 364237/371472 [6:49:21<32:29,  3.71it/s]
 98%|█████████▊| 364238/371472 [6:49:22<32:35,  3.70it/s]
 98%|█████████▊| 364239/371472 [6:49:22<32:54,  3.66it/s]
 98%|█████████▊| 364240/371472 [6:49:22<32:58,  3.66it/s]
                                                         
{'loss': 2.6815, 'learning_rate': 1.1753108233316812e-07, 'epoch': 15.69}
+
 98%|█████████▊| 364240/371472 [6:49:22<32:58,  3.66it/s]
 98%|█████████▊| 364241/371472 [6:49:22<33:00,  3.65it/s]
 98%|█████████▊| 364242/371472 [6:49:23<32:35,  3.70it/s]
 98%|█████████▊| 364243/371472 [6:49:23<31:33,  3.82it/s]
 98%|█████████▊| 364244/371472 [6:49:23<31:03,  3.88it/s]
 98%|█████████▊| 364245/371472 [6:49:23<33:16,  3.62it/s]
 98%|█████████▊| 364246/371472 [6:49:24<34:05,  3.53it/s]
 98%|█████████▊| 364247/371472 [6:49:24<33:27,  3.60it/s]
 98%|█████████▊| 364248/371472 [6:49:24<33:33,  3.59it/s]
 98%|█████████▊| 364249/371472 [6:49:25<35:08,  3.43it/s]
 98%|█████████▊| 364250/371472 [6:49:25<36:37,  3.29it/s]
 98%|█████████▊| 364251/371472 [6:49:25<36:05,  3.33it/s]
 98%|█████████▊| 364252/371472 [6:49:25<35:17,  3.41it/s]
 98%|█████████▊| 364253/371472 [6:49:26<33:53,  3.55it/s]
 98%|█████████▊| 364254/371472 [6:49:26<35:01,  3.43it/s]
 98%|█████████▊| 364255/371472 [6:49:26<37:05,  3.24it/s]
 98%|█████████▊| 364256/371472 [6:49:27<35:30,  3.39it/s]
 98%|█████████▊| 364257/371472 [6:49:27<36:21,  3.31it/s]
 98%|█████████▊| 364258/371472 [6:49:27<36:02,  3.34it/s]
 98%|█████████▊| 364259/371472 [6:49:28<35:24,  3.40it/s]
 98%|█████████▊| 364260/371472 [6:49:28<35:01,  3.43it/s]
                                                         
{'loss': 2.5289, 'learning_rate': 1.1748260035768928e-07, 'epoch': 15.69}
+
 98%|█████████▊| 364260/371472 [6:49:28<35:01,  3.43it/s]
 98%|█████████▊| 364261/371472 [6:49:28<33:36,  3.58it/s]
 98%|█████████▊| 364262/371472 [6:49:28<32:16,  3.72it/s]
 98%|█████████▊| 364263/371472 [6:49:29<36:25,  3.30it/s]
 98%|█████████▊| 364264/371472 [6:49:29<35:24,  3.39it/s]
 98%|█████████▊| 364265/371472 [6:49:29<33:36,  3.57it/s]
 98%|█████████▊| 364266/371472 [6:49:29<32:55,  3.65it/s]
 98%|█████████▊| 364267/371472 [6:49:30<34:27,  3.49it/s]
 98%|█████████▊| 364268/371472 [6:49:30<33:17,  3.61it/s]
 98%|█████████▊| 364269/371472 [6:49:30<34:05,  3.52it/s]
 98%|█████████▊| 364270/371472 [6:49:31<36:38,  3.28it/s]
 98%|█████████▊| 364271/371472 [6:49:31<34:35,  3.47it/s]
 98%|█████████▊| 364272/371472 [6:49:31<34:37,  3.47it/s]
 98%|█████████▊| 364273/371472 [6:49:32<34:27,  3.48it/s]
 98%|█████████▊| 364274/371472 [6:49:32<33:12,  3.61it/s]
 98%|█████████▊| 364275/371472 [6:49:32<32:34,  3.68it/s]
 98%|█████████▊| 364276/371472 [6:49:32<31:42,  3.78it/s]
 98%|█████████▊| 364277/371472 [6:49:33<30:47,  3.89it/s]
 98%|█████████▊| 364278/371472 [6:49:33<31:00,  3.87it/s]
 98%|█████████▊| 364279/371472 [6:49:33<31:25,  3.82it/s]
 98%|█████████▊| 364280/371472 [6:49:33<32:51,  3.65it/s]
                                                         
{'loss': 2.5646, 'learning_rate': 1.1743411838221035e-07, 'epoch': 15.69}
+
 98%|█████████▊| 364280/371472 [6:49:33<32:51,  3.65it/s]
 98%|█████████▊| 364281/371472 [6:49:34<33:01,  3.63it/s]
 98%|█████████▊| 364282/371472 [6:49:34<33:58,  3.53it/s]
 98%|█████████▊| 364283/371472 [6:49:34<35:27,  3.38it/s]
 98%|█████████▊| 364284/371472 [6:49:35<35:50,  3.34it/s]
 98%|█████████▊| 364285/371472 [6:49:35<34:40,  3.45it/s]
 98%|█████████▊| 364286/371472 [6:49:35<34:01,  3.52it/s]
 98%|█████████▊| 364287/371472 [6:49:35<32:32,  3.68it/s]
 98%|█████████▊| 364288/371472 [6:49:36<33:07,  3.62it/s]
 98%|█████████▊| 364289/371472 [6:49:36<32:36,  3.67it/s]
 98%|█████████▊| 364290/371472 [6:49:36<32:53,  3.64it/s]
 98%|█████████▊| 364291/371472 [6:49:36<31:41,  3.78it/s]
 98%|█████████▊| 364292/371472 [6:49:37<33:23,  3.58it/s]
 98%|█████████▊| 364293/371472 [6:49:37<32:24,  3.69it/s]
 98%|█████████▊| 364294/371472 [6:49:37<31:56,  3.75it/s]
 98%|█████████▊| 364295/371472 [6:49:38<32:50,  3.64it/s]
 98%|█████████▊| 364296/371472 [6:49:38<37:02,  3.23it/s]
 98%|██���██████▊| 364297/371472 [6:49:38<35:42,  3.35it/s]
 98%|█████████▊| 364298/371472 [6:49:39<35:08,  3.40it/s]
 98%|█████████▊| 364299/371472 [6:49:39<33:49,  3.53it/s]
 98%|█████████▊| 364300/371472 [6:49:39<33:09,  3.60it/s]
                                                         
{'loss': 2.5328, 'learning_rate': 1.1738563640673149e-07, 'epoch': 15.69}
+
 98%|█████████▊| 364300/371472 [6:49:39<33:09,  3.60it/s]
 98%|█████████▊| 364301/371472 [6:49:39<32:02,  3.73it/s]
 98%|█████████▊| 364302/371472 [6:49:40<34:02,  3.51it/s]
 98%|█████████▊| 364303/371472 [6:49:40<37:36,  3.18it/s]
 98%|█████████▊| 364304/371472 [6:49:40<35:51,  3.33it/s]
 98%|█████████▊| 364305/371472 [6:49:40<34:05,  3.50it/s]
 98%|█████████▊| 364306/371472 [6:49:41<33:15,  3.59it/s]
 98%|█████████▊| 364307/371472 [6:49:41<31:47,  3.76it/s]
 98%|█████████▊| 364308/371472 [6:49:41<32:47,  3.64it/s]
 98%|█████████▊| 364309/371472 [6:49:42<34:52,  3.42it/s]
 98%|█████████▊| 364310/371472 [6:49:42<34:17,  3.48it/s]
 98%|█████████▊| 364311/371472 [6:49:42<32:15,  3.70it/s]
 98%|█████████▊| 364312/371472 [6:49:42<33:35,  3.55it/s]
 98%|█████████▊| 364313/371472 [6:49:43<33:57,  3.51it/s]
 98%|█████████▊| 364314/371472 [6:49:43<34:57,  3.41it/s]
 98%|█████████▊| 364315/371472 [6:49:43<35:37,  3.35it/s]
 98%|█████████▊| 364316/371472 [6:49:44<35:11,  3.39it/s]
 98%|█████████▊| 364317/371472 [6:49:44<36:03,  3.31it/s]
 98%|█████████▊| 364318/371472 [6:49:44<34:31,  3.45it/s]
 98%|█████████▊| 364319/371472 [6:49:45<34:32,  3.45it/s]
 98%|█████████▊| 364320/371472 [6:49:45<33:22,  3.57it/s]
                                                         
{'loss': 2.4993, 'learning_rate': 1.1733715443125255e-07, 'epoch': 15.69}
+
 98%|█████████▊| 364320/371472 [6:49:45<33:22,  3.57it/s]
 98%|█████████▊| 364321/371472 [6:49:45<32:44,  3.64it/s]
 98%|█████████▊| 364322/371472 [6:49:45<32:10,  3.70it/s]
 98%|█████████▊| 364323/371472 [6:49:46<32:11,  3.70it/s]
 98%|█████████▊| 364324/371472 [6:49:46<31:23,  3.80it/s]
 98%|█████████▊| 364325/371472 [6:49:46<31:00,  3.84it/s]
 98%|█████████▊| 364326/371472 [6:49:46<31:11,  3.82it/s]
 98%|█████████▊| 364327/371472 [6:49:47<32:12,  3.70it/s]
 98%|█████████▊| 364328/371472 [6:49:47<32:49,  3.63it/s]
 98%|█████████▊| 364329/371472 [6:49:47<32:09,  3.70it/s]
 98%|█████████▊| 364330/371472 [6:49:47<31:28,  3.78it/s]
 98%|█████████▊| 364331/371472 [6:49:48<34:12,  3.48it/s]
 98%|█████████▊| 364332/371472 [6:49:48<32:31,  3.66it/s]
 98%|█████████▊| 364333/371472 [6:49:48<34:53,  3.41it/s]
 98%|█████████▊| 364334/371472 [6:49:49<33:26,  3.56it/s]
 98%|█████████▊| 364335/371472 [6:49:49<32:28,  3.66it/s]
 98%|█████████▊| 364336/371472 [6:49:49<34:42,  3.43it/s]
 98%|█████████▊| 364337/371472 [6:49:49<34:05,  3.49it/s]
 98%|█████████▊| 364338/371472 [6:49:50<32:18,  3.68it/s]
 98%|█████████▊| 364339/371472 [6:49:50<31:50,  3.73it/s]
 98%|█████████▊| 364340/371472 [6:49:50<30:39,  3.88it/s]
                                                         
{'loss': 2.6201, 'learning_rate': 1.1728867245577371e-07, 'epoch': 15.69}
+
 98%|█████████▊| 364340/371472 [6:49:50<30:39,  3.88it/s]
 98%|█████████▊| 364341/371472 [6:49:50<31:35,  3.76it/s]
 98%|█████████▊| 364342/371472 [6:49:51<31:09,  3.81it/s]
 98%|█████████▊| 364343/371472 [6:49:51<30:56,  3.84it/s]
 98%|█████████▊| 364344/371472 [6:49:51<30:32,  3.89it/s]
 98%|█████████▊| 364345/371472 [6:49:52<31:08,  3.81it/s]
 98%|█████████▊| 364346/371472 [6:49:52<32:30,  3.65it/s]
 98%|█████████▊| 364347/371472 [6:49:52<32:49,  3.62it/s]
 98%|█████████▊| 364348/371472 [6:49:52<31:36,  3.76it/s]
 98%|█████████▊| 364349/371472 [6:49:53<31:23,  3.78it/s]
 98%|█████████▊| 364350/371472 [6:49:53<30:37,  3.88it/s]
 98%|█████████▊| 364351/371472 [6:49:53<31:55,  3.72it/s]
 98%|█████████▊| 364352/371472 [6:49:53<31:23,  3.78it/s]
 98%|█████████▊| 364353/371472 [6:49:54<31:39,  3.75it/s]
 98%|█████████▊| 364354/371472 [6:49:54<31:32,  3.76it/s]
 98%|█████████▊| 364355/371472 [6:49:54<30:43,  3.86it/s]
 98%|█████████▊| 364356/371472 [6:49:54<30:30,  3.89it/s]
 98%|█████████▊| 364357/371472 [6:49:55<33:13,  3.57it/s]
 98%|█████████▊| 364358/371472 [6:49:55<33:28,  3.54it/s]
 98%|█████████▊| 364359/371472 [6:49:55<32:17,  3.67it/s]
 98%|█████████▊| 364360/371472 [6:49:56<31:24,  3.77it/s]
                                                         
{'loss': 2.7352, 'learning_rate': 1.1724019048029477e-07, 'epoch': 15.69}
+
 98%|█████████▊| 364360/371472 [6:49:56<31:24,  3.77it/s]
 98%|█████████▊| 364361/371472 [6:49:56<33:00,  3.59it/s]
 98%|█████████▊| 364362/371472 [6:49:56<32:00,  3.70it/s]
 98%|█████████▊| 364363/371472 [6:49:56<31:38,  3.75it/s]
 98%|█████████▊| 364364/371472 [6:49:57<32:13,  3.68it/s]
 98%|█████████▊| 364365/371472 [6:49:57<31:52,  3.72it/s]
 98%|█████████▊| 364366/371472 [6:49:57<31:30,  3.76it/s]
 98%|█████████▊| 364367/371472 [6:49:57<32:53,  3.60it/s]
 98%|█████████▊| 364368/371472 [6:49:58<31:47,  3.72it/s]
 98%|█████████▊| 364369/371472 [6:49:58<30:55,  3.83it/s]
 98%|█████████▊| 364370/371472 [6:49:58<30:59,  3.82it/s]
 98%|█████████▊| 364371/371472 [6:49:58<30:11,  3.92it/s]
 98%|█████████▊| 364372/371472 [6:49:59<30:10,  3.92it/s]
 98%|█████████▊| 364373/371472 [6:49:59<29:46,  3.97it/s]
 98%|█████████▊| 364374/371472 [6:49:59<29:57,  3.95it/s]
 98%|█████████▊| 364375/371472 [6:50:00<31:32,  3.75it/s]
 98%|█████████▊| 364376/371472 [6:50:00<30:32,  3.87it/s]
 98%|█████████▊| 364377/371472 [6:50:00<30:14,  3.91it/s]
 98%|█████████▊| 364378/371472 [6:50:00<29:22,  4.03it/s]
 98%|█████████▊| 364379/371472 [6:50:00<29:58,  3.94it/s]
 98%|█████████▊| 364380/371472 [6:50:01<31:08,  3.80it/s]
                                                         
{'loss': 2.7911, 'learning_rate': 1.1719170850481582e-07, 'epoch': 15.69}
+
 98%|█████████▊| 364380/371472 [6:50:01<31:08,  3.80it/s]
 98%|█████████▊| 364381/371472 [6:50:01<31:38,  3.73it/s]
 98%|█████████▊| 364382/371472 [6:50:01<31:15,  3.78it/s]
 98%|█████████▊| 364383/371472 [6:50:02<31:54,  3.70it/s]
 98%|█████████▊| 364384/371472 [6:50:02<35:55,  3.29it/s]
 98%|█████████▊| 364385/371472 [6:50:02<35:49,  3.30it/s]
 98%|█████████▊| 364386/371472 [6:50:03<34:29,  3.42it/s]
 98%|█████████▊| 364387/371472 [6:50:03<34:34,  3.42it/s]
 98%|█████████▊| 364388/371472 [6:50:03<40:41,  2.90it/s]
 98%|█████████▊| 364389/371472 [6:50:04<37:34,  3.14it/s]
 98%|█████████▊| 364390/371472 [6:50:04<35:20,  3.34it/s]
 98%|█████████▊| 364391/371472 [6:50:04<34:01,  3.47it/s]
 98%|█████████▊| 364392/371472 [6:50:04<35:07,  3.36it/s]
 98%|█████████▊| 364393/371472 [6:50:05<34:51,  3.38it/s]
 98%|█████████▊| 364394/371472 [6:50:05<32:44,  3.60it/s]
 98%|█████████▊| 364395/371472 [6:50:05<36:01,  3.27it/s]
 98%|█████████▊| 364396/371472 [6:50:06<34:46,  3.39it/s]
 98%|█████████▊| 364397/371472 [6:50:06<33:12,  3.55it/s]
 98%|█████████▊| 364398/371472 [6:50:06<33:35,  3.51it/s]
 98%|█████████▊| 364399/371472 [6:50:06<32:40,  3.61it/s]
 98%|█████████▊| 364400/371472 [6:50:07<34:13,  3.44it/s]
                                                         
{'loss': 2.4854, 'learning_rate': 1.1714322652933698e-07, 'epoch': 15.7}
+
 98%|█████████▊| 364400/371472 [6:50:07<34:13,  3.44it/s]
 98%|█████████▊| 364401/371472 [6:50:07<35:09,  3.35it/s]
 98%|█████████▊| 364402/371472 [6:50:07<33:28,  3.52it/s]
 98%|█████████▊| 364403/371472 [6:50:08<32:45,  3.60it/s]
 98%|█████████▊| 364404/371472 [6:50:08<31:18,  3.76it/s]
 98%|█████████▊| 364405/371472 [6:50:08<31:11,  3.78it/s]
 98%|█████████▊| 364406/371472 [6:50:08<30:35,  3.85it/s]
 98%|█████████▊| 364407/371472 [6:50:09<31:45,  3.71it/s]
 98%|█████████▊| 364408/371472 [6:50:09<30:49,  3.82it/s]
 98%|█████████▊| 364409/371472 [6:50:09<30:19,  3.88it/s]
 98%|█████████▊| 364410/371472 [6:50:09<30:32,  3.85it/s]
 98%|█████████▊| 364411/371472 [6:50:10<32:14,  3.65it/s]
 98%|█████████▊| 364412/371472 [6:50:10<31:18,  3.76it/s]
 98%|█████████▊| 364413/371472 [6:50:10<30:41,  3.83it/s]
 98%|█████████▊| 364414/371472 [6:50:10<30:49,  3.82it/s]
 98%|█████████▊| 364415/371472 [6:50:11<32:26,  3.63it/s]
 98%|█████████▊| 364416/371472 [6:50:11<33:51,  3.47it/s]
 98%|█████████▊| 364417/371472 [6:50:11<33:13,  3.54it/s]
 98%|█████████▊| 364418/371472 [6:50:12<32:36,  3.60it/s]
 98%|█████████▊| 364419/371472 [6:50:12<32:37,  3.60it/s]
 98%|█████████▊| 364420/371472 [6:50:12<32:29,  3.62it/s]
                                                         
{'loss': 2.7192, 'learning_rate': 1.1709474455385804e-07, 'epoch': 15.7}
+
 98%|█████████▊| 364420/371472 [6:50:12<32:29,  3.62it/s]
 98%|█████████▊| 364421/371472 [6:50:12<31:46,  3.70it/s]
 98%|█████████▊| 364422/371472 [6:50:13<33:06,  3.55it/s]
 98%|█████████▊| 364423/371472 [6:50:13<32:44,  3.59it/s]
 98%|█████████▊| 364424/371472 [6:50:13<32:14,  3.64it/s]
 98%|█████████▊| 364425/371472 [6:50:14<33:06,  3.55it/s]
 98%|█████████▊| 364426/371472 [6:50:14<35:26,  3.31it/s]
 98%|█████████▊| 364427/371472 [6:50:14<36:15,  3.24it/s]
 98%|█████████▊| 364428/371472 [6:50:14<34:29,  3.40it/s]
 98%|█████████▊| 364429/371472 [6:50:15<33:27,  3.51it/s]
 98%|█████████▊| 364430/371472 [6:50:15<32:51,  3.57it/s]
 98%|█████████▊| 364431/371472 [6:50:15<32:08,  3.65it/s]
 98%|█████████▊| 364432/371472 [6:50:16<33:36,  3.49it/s]
 98%|█████████▊| 364433/371472 [6:50:16<33:57,  3.46it/s]
 98%|█████████▊| 364434/371472 [6:50:16<32:39,  3.59it/s]
 98%|█████████▊| 364435/371472 [6:50:16<33:33,  3.49it/s]
 98%|█████████▊| 364436/371472 [6:50:17<33:12,  3.53it/s]
 98%|█████████▊| 364437/371472 [6:50:17<32:24,  3.62it/s]
 98%|█████████▊| 364438/371472 [6:50:17<31:21,  3.74it/s]
 98%|█████████▊| 364439/371472 [6:50:17<31:34,  3.71it/s]
 98%|█████████▊| 364440/371472 [6:50:18<31:41,  3.70it/s]
                                                         
{'loss': 2.4633, 'learning_rate': 1.1704626257837919e-07, 'epoch': 15.7}
+
 98%|█████████▊| 364440/371472 [6:50:18<31:41,  3.70it/s]
 98%|█████████▊| 364441/371472 [6:50:18<31:21,  3.74it/s]
 98%|█████████▊| 364442/371472 [6:50:18<31:15,  3.75it/s]
 98%|█████████▊| 364443/371472 [6:50:19<32:52,  3.56it/s]
 98%|█████████▊| 364444/371472 [6:50:19<33:18,  3.52it/s]
 98%|█████████▊| 364445/371472 [6:50:19<33:11,  3.53it/s]
 98%|█████████▊| 364446/371472 [6:50:19<33:23,  3.51it/s]
 98%|█████████▊| 364447/371472 [6:50:20<34:36,  3.38it/s]
 98%|█████████▊| 364448/371472 [6:50:20<35:58,  3.25it/s]
 98%|█████████▊| 364449/371472 [6:50:20<36:17,  3.23it/s]
 98%|█████████▊| 364450/371472 [6:50:21<35:01,  3.34it/s]
 98%|█████████▊| 364451/371472 [6:50:21<33:38,  3.48it/s]
 98%|█████████▊| 364452/371472 [6:50:21<32:37,  3.59it/s]
 98%|█████████▊| 364453/371472 [6:50:21<32:12,  3.63it/s]
 98%|█████████▊| 364454/371472 [6:50:22<31:59,  3.66it/s]
 98%|█████████▊| 364455/371472 [6:50:22<34:04,  3.43it/s]
 98%|█████████▊| 364456/371472 [6:50:22<33:07,  3.53it/s]
 98%|█████████▊| 364457/371472 [6:50:23<31:54,  3.66it/s]
 98%|█████████▊| 364458/371472 [6:50:23<32:56,  3.55it/s]
 98%|█████████▊| 364459/371472 [6:50:23<33:41,  3.47it/s]
 98%|█████████▊| 364460/371472 [6:50:23<32:17,  3.62it/s]
                                                         
{'loss': 2.6139, 'learning_rate': 1.1699778060290026e-07, 'epoch': 15.7}
+
 98%|█████████▊| 364460/371472 [6:50:23<32:17,  3.62it/s]
 98%|█████████▊| 364461/371472 [6:50:24<31:47,  3.67it/s]
 98%|█████████▊| 364462/371472 [6:50:24<39:07,  2.99it/s]
 98%|█████████▊| 364463/371472 [6:50:25<39:12,  2.98it/s]
 98%|█████████▊| 364464/371472 [6:50:25<36:13,  3.22it/s]
 98%|█████████▊| 364465/371472 [6:50:25<34:32,  3.38it/s]
 98%|█████████▊| 364466/371472 [6:50:25<33:11,  3.52it/s]
 98%|█████████▊| 364467/371472 [6:50:26<31:47,  3.67it/s]
 98%|█████████▊| 364468/371472 [6:50:26<32:49,  3.56it/s]
 98%|█████████▊| 364469/371472 [6:50:26<33:02,  3.53it/s]
 98%|█████████▊| 364470/371472 [6:50:26<31:45,  3.68it/s]
 98%|█████████▊| 364471/371472 [6:50:27<33:12,  3.51it/s]
 98%|█████████▊| 364472/371472 [6:50:27<32:12,  3.62it/s]
 98%|█████████▊| 364473/371472 [6:50:27<31:39,  3.68it/s]
 98%|█████████▊| 364474/371472 [6:50:27<30:28,  3.83it/s]
 98%|█████████▊| 364475/371472 [6:50:28<30:11,  3.86it/s]
 98%|█████████▊| 364476/371472 [6:50:28<31:17,  3.73it/s]
 98%|█████████▊| 364477/371472 [6:50:28<30:31,  3.82it/s]
 98%|█████████▊| 364478/371472 [6:50:29<33:41,  3.46it/s]
 98%|█████████▊| 364479/371472 [6:50:29<33:56,  3.43it/s]
 98%|█████████▊| 364480/371472 [6:50:29<33:29,  3.48it/s]
                                                         
{'loss': 2.4743, 'learning_rate': 1.1694929862742141e-07, 'epoch': 15.7}
+
 98%|█████████▊| 364480/371472 [6:50:29<33:29,  3.48it/s]
 98%|█████████▊| 364481/371472 [6:50:29<32:19,  3.61it/s]
 98%|█████████▊| 364482/371472 [6:50:30<33:45,  3.45it/s]
 98%|█████████▊| 364483/371472 [6:50:30<33:12,  3.51it/s]
 98%|█████████▊| 364484/371472 [6:50:30<34:09,  3.41it/s]
 98%|█████████▊| 364485/371472 [6:50:31<33:56,  3.43it/s]
 98%|█████████▊| 364486/371472 [6:50:31<32:17,  3.60it/s]
 98%|█████████▊| 364487/371472 [6:50:31<33:12,  3.51it/s]
 98%|█████████▊| 364488/371472 [6:50:31<32:59,  3.53it/s]
 98%|█████████▊| 364489/371472 [6:50:32<32:52,  3.54it/s]
 98%|█████████▊| 364490/371472 [6:50:32<33:01,  3.52it/s]
 98%|█████████▊| 364491/371472 [6:50:32<32:24,  3.59it/s]
 98%|█████████▊| 364492/371472 [6:50:32<30:26,  3.82it/s]
 98%|█████████▊| 364493/371472 [6:50:33<30:53,  3.76it/s]
 98%|█████████▊| 364494/371472 [6:50:33<30:03,  3.87it/s]
 98%|█████████▊| 364495/371472 [6:50:33<30:42,  3.79it/s]
 98%|█████████▊| 364496/371472 [6:50:34<31:33,  3.68it/s]
 98%|█████████▊| 364497/371472 [6:50:34<33:28,  3.47it/s]
 98%|█████████▊| 364498/371472 [6:50:34<35:59,  3.23it/s]
 98%|█████████▊| 364499/371472 [6:50:35<34:25,  3.38it/s]
 98%|█████████▊| 364500/371472 [6:50:35<33:39,  3.45it/s]
                                                         
{'loss': 2.6811, 'learning_rate': 1.1690081665194247e-07, 'epoch': 15.7}
+
 98%|█████████▊| 364500/371472 [6:50:35<33:39,  3.45it/s]
 98%|█████████▊| 364501/371472 [6:50:35<32:36,  3.56it/s]
 98%|█████████▊| 364502/371472 [6:50:35<32:17,  3.60it/s]
 98%|█████████▊| 364503/371472 [6:50:36<30:50,  3.77it/s]
 98%|█████████▊| 364504/371472 [6:50:36<30:21,  3.83it/s]
 98%|█████████▊| 364505/371472 [6:50:36<29:57,  3.88it/s]
 98%|█████████▊| 364506/371472 [6:50:36<30:40,  3.78it/s]
 98%|█████████▊| 364507/371472 [6:50:37<30:06,  3.86it/s]
 98%|█████████▊| 364508/371472 [6:50:37<31:14,  3.72it/s]
 98%|█████████▊| 364509/371472 [6:50:37<30:49,  3.76it/s]
 98%|█████████▊| 364510/371472 [6:50:37<32:18,  3.59it/s]
 98%|█████████▊| 364511/371472 [6:50:38<32:34,  3.56it/s]
 98%|█████████▊| 364512/371472 [6:50:38<32:23,  3.58it/s]
 98%|█████████▊| 364513/371472 [6:50:38<32:43,  3.54it/s]
 98%|█████████▊| 364514/371472 [6:50:39<30:52,  3.76it/s]
 98%|█████████▊| 364515/371472 [6:50:39<32:14,  3.60it/s]
 98%|█████████▊| 364516/371472 [6:50:39<31:38,  3.66it/s]
 98%|█████████▊| 364517/371472 [6:50:39<32:58,  3.51it/s]
 98%|█████████▊| 364518/371472 [6:50:40<32:31,  3.56it/s]
 98%|█████████▊| 364519/371472 [6:50:40<31:10,  3.72it/s]
 98%|█████████▊| 364520/371472 [6:50:41<42:28,  2.73it/s]
                                                         
{'loss': 2.6283, 'learning_rate': 1.1685233467646363e-07, 'epoch': 15.7}
+
 98%|█████████▊| 364520/371472 [6:50:41<42:28,  2.73it/s]
 98%|█████████▊| 364521/371472 [6:50:41<39:18,  2.95it/s]
 98%|█████████▊| 364522/371472 [6:50:41<37:55,  3.05it/s]
 98%|█████████▊| 364523/371472 [6:50:41<35:34,  3.25it/s]
 98%|█████████▊| 364524/371472 [6:50:42<33:17,  3.48it/s]
 98%|█████████▊| 364525/371472 [6:50:42<32:00,  3.62it/s]
 98%|█████████▊| 364526/371472 [6:50:42<32:31,  3.56it/s]
 98%|███���█████▊| 364527/371472 [6:50:42<32:30,  3.56it/s]
 98%|█████████▊| 364528/371472 [6:50:43<32:35,  3.55it/s]
 98%|█████████▊| 364529/371472 [6:50:43<31:19,  3.69it/s]
 98%|█████████▊| 364530/371472 [6:50:43<30:59,  3.73it/s]
 98%|█████████▊| 364531/371472 [6:50:43<30:20,  3.81it/s]
 98%|█████████▊| 364532/371472 [6:50:44<31:07,  3.72it/s]
 98%|█████████▊| 364533/371472 [6:50:44<31:29,  3.67it/s]
 98%|█████████▊| 364534/371472 [6:50:44<30:48,  3.75it/s]
 98%|█████████▊| 364535/371472 [6:50:45<30:07,  3.84it/s]
 98%|█████████▊| 364536/371472 [6:50:45<30:51,  3.75it/s]
 98%|█████████▊| 364537/371472 [6:50:45<30:13,  3.82it/s]
 98%|█████████▊| 364538/371472 [6:50:45<30:59,  3.73it/s]
 98%|█████████▊| 364539/371472 [6:50:46<30:51,  3.74it/s]
 98%|█████████▊| 364540/371472 [6:50:46<32:56,  3.51it/s]
                                                         
{'loss': 2.5899, 'learning_rate': 1.1680385270098468e-07, 'epoch': 15.7}
+
 98%|█████████▊| 364540/371472 [6:50:46<32:56,  3.51it/s]
 98%|█████████▊| 364541/371472 [6:50:46<33:49,  3.42it/s]
 98%|█████████▊| 364542/371472 [6:50:47<33:39,  3.43it/s]
 98%|█████████▊| 364543/371472 [6:50:47<34:29,  3.35it/s]
 98%|█████████▊| 364544/371472 [6:50:47<34:50,  3.31it/s]
 98%|█████████▊| 364545/371472 [6:50:47<33:30,  3.45it/s]
 98%|█████████▊| 364546/371472 [6:50:48<32:36,  3.54it/s]
 98%|█████████▊| 364547/371472 [6:50:48<36:29,  3.16it/s]
 98%|█████████▊| 364548/371472 [6:50:48<35:51,  3.22it/s]
 98%|█████████▊| 364549/371472 [6:50:49<36:16,  3.18it/s]
 98%|█████████▊| 364550/371472 [6:50:49<35:10,  3.28it/s]
 98%|█████████▊| 364551/371472 [6:50:49<33:40,  3.43it/s]
 98%|█████████▊| 364552/371472 [6:50:50<33:24,  3.45it/s]
 98%|█████████▊| 364553/371472 [6:50:50<32:14,  3.58it/s]
 98%|█████████▊| 364554/371472 [6:50:50<32:04,  3.60it/s]
 98%|█████████▊| 364555/371472 [6:50:50<31:28,  3.66it/s]
 98%|█████████▊| 364556/371472 [6:50:51<31:41,  3.64it/s]
 98%|█████████▊| 364557/371472 [6:50:51<31:01,  3.71it/s]
 98%|█████████▊| 364558/371472 [6:50:51<30:13,  3.81it/s]
 98%|█████████▊| 364559/371472 [6:50:51<32:14,  3.57it/s]
 98%|█████████▊| 364560/371472 [6:50:52<34:31,  3.34it/s]
                                                         
{'loss': 2.6301, 'learning_rate': 1.1675537072550584e-07, 'epoch': 15.7}
+
 98%|█████████▊| 364560/371472 [6:50:52<34:31,  3.34it/s]
 98%|█████████▊| 364561/371472 [6:50:52<33:38,  3.42it/s]
 98%|█████████▊| 364562/371472 [6:50:52<32:50,  3.51it/s]
 98%|█████████▊| 364563/371472 [6:50:53<35:30,  3.24it/s]
 98%|█████████▊| 364564/371472 [6:50:53<32:50,  3.51it/s]
 98%|█████████▊| 364565/371472 [6:50:53<33:10,  3.47it/s]
 98%|█████████▊| 364566/371472 [6:50:53<31:29,  3.65it/s]
 98%|█████████▊| 364567/371472 [6:50:54<30:44,  3.74it/s]
 98%|█████████▊| 364568/371472 [6:50:54<30:06,  3.82it/s]
 98%|█████████▊| 364569/371472 [6:50:54<29:54,  3.85it/s]
 98%|█████████▊| 364570/371472 [6:50:54<30:42,  3.75it/s]
 98%|█████████▊| 364571/371472 [6:50:55<30:32,  3.77it/s]
 98%|█████████▊| 364572/371472 [6:50:55<29:53,  3.85it/s]
 98%|█████████▊| 364573/371472 [6:50:55<30:09,  3.81it/s]
 98%|█████████▊| 364574/371472 [6:50:55<29:55,  3.84it/s]
 98%|█████████▊| 364575/371472 [6:50:56<31:16,  3.67it/s]
 98%|█████████▊| 364576/371472 [6:50:56<32:07,  3.58it/s]
 98%|█████████▊| 364577/371472 [6:50:56<34:25,  3.34it/s]
 98%|█████████▊| 364578/371472 [6:50:57<34:14,  3.36it/s]
 98%|█████████▊| 364579/371472 [6:50:57<32:25,  3.54it/s]
 98%|█████████▊| 364580/371472 [6:50:57<33:54,  3.39it/s]
                                                         
{'loss': 2.6389, 'learning_rate': 1.167068887500269e-07, 'epoch': 15.7}
+
 98%|█████████▊| 364580/371472 [6:50:57<33:54,  3.39it/s]
 98%|█████████▊| 364581/371472 [6:50:58<34:38,  3.32it/s]
 98%|█████████▊| 364582/371472 [6:50:58<34:29,  3.33it/s]
 98%|█████████▊| 364583/371472 [6:50:58<35:17,  3.25it/s]
 98%|█████████▊| 364584/371472 [6:50:59<33:45,  3.40it/s]
 98%|█████████▊| 364585/371472 [6:50:59<34:04,  3.37it/s]
 98%|█████████▊| 364586/371472 [6:50:59<33:38,  3.41it/s]
 98%|█████████▊| 364587/371472 [6:50:59<33:56,  3.38it/s]
 98%|█████████▊| 364588/371472 [6:51:00<33:33,  3.42it/s]
 98%|█████████▊| 364589/371472 [6:51:00<33:17,  3.45it/s]
 98%|█████████▊| 364590/371472 [6:51:00<31:15,  3.67it/s]
 98%|█████████▊| 364591/371472 [6:51:01<32:59,  3.48it/s]
 98%|█████████▊| 364592/371472 [6:51:01<33:55,  3.38it/s]
 98%|█████████▊| 364593/371472 [6:51:01<33:06,  3.46it/s]
 98%|█████████▊| 364594/371472 [6:51:01<32:56,  3.48it/s]
 98%|█████████▊| 364595/371472 [6:51:02<31:02,  3.69it/s]
 98%|█████████▊| 364596/371472 [6:51:02<30:43,  3.73it/s]
 98%|█████████▊| 364597/371472 [6:51:02<29:52,  3.83it/s]
 98%|█████████▊| 364598/371472 [6:51:02<30:19,  3.78it/s]
 98%|█████████▊| 364599/371472 [6:51:03<31:30,  3.64it/s]
 98%|█████████▊| 364600/371472 [6:51:03<31:27,  3.64it/s]
                                                         
{'loss': 2.6371, 'learning_rate': 1.1665840677454805e-07, 'epoch': 15.7}
+
 98%|█████████▊| 364600/371472 [6:51:03<31:27,  3.64it/s]
 98%|█████████▊| 364601/371472 [6:51:03<30:33,  3.75it/s]
 98%|█████████▊| 364602/371472 [6:51:03<30:57,  3.70it/s]
 98%|█████████▊| 364603/371472 [6:51:04<33:44,  3.39it/s]
 98%|█████████▊| 364604/371472 [6:51:04<33:58,  3.37it/s]
 98%|█████████▊| 364605/371472 [6:51:04<32:14,  3.55it/s]
 98%|█████████▊| 364606/371472 [6:51:05<32:15,  3.55it/s]
 98%|█████████▊| 364607/371472 [6:51:05<32:41,  3.50it/s]
 98%|█████████▊| 364608/371472 [6:51:05<31:08,  3.67it/s]
 98%|█████████▊| 364609/371472 [6:51:05<30:16,  3.78it/s]
 98%|█████████▊| 364610/371472 [6:51:06<31:13,  3.66it/s]
 98%|█████████▊| 364611/371472 [6:51:06<31:39,  3.61it/s]
 98%|█████████▊| 364612/371472 [6:51:06<30:54,  3.70it/s]
 98%|█████████▊| 364613/371472 [6:51:07<30:11,  3.79it/s]
 98%|█████████▊| 364614/371472 [6:51:07<30:43,  3.72it/s]
 98%|█████████▊| 364615/371472 [6:51:07<29:39,  3.85it/s]
 98%|█████████▊| 364616/371472 [6:51:07<30:18,  3.77it/s]
 98%|█████████▊| 364617/371472 [6:51:08<30:52,  3.70it/s]
 98%|█████████▊| 364618/371472 [6:51:08<30:58,  3.69it/s]
 98%|█████████▊| 364619/371472 [6:51:08<30:39,  3.73it/s]
 98%|█████████▊| 364620/371472 [6:51:08<29:50,  3.83it/s]
                                                         
{'loss': 2.6628, 'learning_rate': 1.1660992479906911e-07, 'epoch': 15.7}
+
 98%|█████████▊| 364620/371472 [6:51:08<29:50,  3.83it/s]
 98%|█████████▊| 364621/371472 [6:51:09<29:18,  3.90it/s]
 98%|█████████▊| 364622/371472 [6:51:09<29:33,  3.86it/s]
 98%|█████████▊| 364623/371472 [6:51:09<30:27,  3.75it/s]
 98%|█████████▊| 364624/371472 [6:51:10<31:56,  3.57it/s]
 98%|█████████▊| 364625/371472 [6:51:10<30:40,  3.72it/s]
 98%|█████████▊| 364626/371472 [6:51:10<30:21,  3.76it/s]
 98%|█████████▊| 364627/371472 [6:51:10<31:21,  3.64it/s]
 98%|█████████▊| 364628/371472 [6:51:11<31:39,  3.60it/s]
 98%|█████████▊| 364629/371472 [6:51:11<31:44,  3.59it/s]
 98%|█████████▊| 364630/371472 [6:51:11<31:51,  3.58it/s]
 98%|█████████▊| 364631/371472 [6:51:11<31:15,  3.65it/s]
 98%|█████████▊| 364632/371472 [6:51:12<31:10,  3.66it/s]
 98%|█████████▊| 364633/371472 [6:51:12<31:16,  3.64it/s]
 98%|█████████▊| 364634/371472 [6:51:12<31:00,  3.67it/s]
 98%|█████████▊| 364635/371472 [6:51:13<30:59,  3.68it/s]
 98%|█████████▊| 364636/371472 [6:51:13<31:09,  3.66it/s]
 98%|█████████▊| 364637/371472 [6:51:13<31:06,  3.66it/s]
 98%|█████████▊| 364638/371472 [6:51:13<30:27,  3.74it/s]
 98%|█████████▊| 364639/371472 [6:51:14<29:05,  3.91it/s]
 98%|█████████▊| 364640/371472 [6:51:14<30:43,  3.71it/s]
                                                         
{'loss': 2.6313, 'learning_rate': 1.1656144282359027e-07, 'epoch': 15.71}
+
 98%|█████████▊| 364640/371472 [6:51:14<30:43,  3.71it/s]
 98%|█████████▊| 364641/371472 [6:51:14<32:07,  3.54it/s]
 98%|████��████▊| 364642/371472 [6:51:14<32:26,  3.51it/s]
 98%|█████████▊| 364643/371472 [6:51:15<36:13,  3.14it/s]
 98%|█████████▊| 364644/371472 [6:51:15<34:48,  3.27it/s]
 98%|█████████▊| 364645/371472 [6:51:15<32:58,  3.45it/s]
 98%|█████████▊| 364646/371472 [6:51:16<31:40,  3.59it/s]
 98%|█████████▊| 364647/371472 [6:51:16<30:57,  3.67it/s]
 98%|█████████▊| 364648/371472 [6:51:16<33:48,  3.36it/s]
 98%|█████████▊| 364649/371472 [6:51:16<32:01,  3.55it/s]
 98%|█████████▊| 364650/371472 [6:51:17<32:28,  3.50it/s]
 98%|█████████▊| 364651/371472 [6:51:17<34:20,  3.31it/s]
 98%|█████████▊| 364652/371472 [6:51:17<34:20,  3.31it/s]
 98%|█████████▊| 364653/371472 [6:51:18<33:20,  3.41it/s]
 98%|█████████▊| 364654/371472 [6:51:18<32:26,  3.50it/s]
 98%|█████████▊| 364655/371472 [6:51:18<31:25,  3.61it/s]
 98%|█████████▊| 364656/371472 [6:51:18<31:01,  3.66it/s]
 98%|█████████▊| 364657/371472 [6:51:19<30:10,  3.76it/s]
 98%|█████████▊| 364658/371472 [6:51:19<29:54,  3.80it/s]
 98%|█████████▊| 364659/371472 [6:51:19<29:56,  3.79it/s]
 98%|█████████▊| 364660/371472 [6:51:19<29:39,  3.83it/s]
                                                         
{'loss': 2.5691, 'learning_rate': 1.1651296084811133e-07, 'epoch': 15.71}
+
 98%|█████████▊| 364660/371472 [6:51:20<29:39,  3.83it/s]
 98%|█████████▊| 364661/371472 [6:51:20<31:52,  3.56it/s]
 98%|█████████▊| 364662/371472 [6:51:20<32:28,  3.50it/s]
 98%|█████████▊| 364663/371472 [6:51:20<32:59,  3.44it/s]
 98%|█████████▊| 364664/371472 [6:51:21<35:24,  3.21it/s]
 98%|█████████▊| 364665/371472 [6:51:21<33:23,  3.40it/s]
 98%|█████████▊| 364666/371472 [6:51:21<32:08,  3.53it/s]
 98%|█████████▊| 364667/371472 [6:51:22<30:40,  3.70it/s]
 98%|█████████▊| 364668/371472 [6:51:22<31:36,  3.59it/s]
 98%|█████████▊| 364669/371472 [6:51:22<34:08,  3.32it/s]
 98%|█████████▊| 364670/371472 [6:51:22<32:13,  3.52it/s]
 98%|█████████▊| 364671/371472 [6:51:23<31:10,  3.64it/s]
 98%|█████████▊| 364672/371472 [6:51:23<31:11,  3.63it/s]
 98%|█████████▊| 364673/371472 [6:51:23<34:01,  3.33it/s]
 98%|█████████▊| 364674/371472 [6:51:24<32:20,  3.50it/s]
 98%|█████████▊| 364675/371472 [6:51:24<31:48,  3.56it/s]
 98%|█████████▊| 364676/371472 [6:51:24<31:59,  3.54it/s]
 98%|█████████▊| 364677/371472 [6:51:24<32:38,  3.47it/s]
 98%|█████████▊| 364678/371472 [6:51:25<31:43,  3.57it/s]
 98%|█████████▊| 364679/371472 [6:51:25<31:32,  3.59it/s]
 98%|█████████▊| 364680/371472 [6:51:25<32:25,  3.49it/s]
                                                         
{'loss': 2.723, 'learning_rate': 1.1646447887263248e-07, 'epoch': 15.71}
+
 98%|█████████▊| 364680/371472 [6:51:25<32:25,  3.49it/s]
 98%|█████████▊| 364681/371472 [6:51:26<32:00,  3.54it/s]
 98%|█████████▊| 364682/371472 [6:51:26<31:19,  3.61it/s]
 98%|█████████▊| 364683/371472 [6:51:26<31:31,  3.59it/s]
 98%|█████████▊| 364684/371472 [6:51:26<30:44,  3.68it/s]
 98%|█████████▊| 364685/371472 [6:51:27<32:54,  3.44it/s]
 98%|█████████▊| 364686/371472 [6:51:27<31:53,  3.55it/s]
 98%|█████████▊| 364687/371472 [6:51:27<32:34,  3.47it/s]
 98%|█████████▊| 364688/371472 [6:51:28<32:33,  3.47it/s]
 98%|█████████▊| 364689/371472 [6:51:28<32:41,  3.46it/s]
 98%|█████████▊| 364690/371472 [6:51:28<34:45,  3.25it/s]
 98%|█████████▊| 364691/371472 [6:51:28<33:11,  3.40it/s]
 98%|█████████▊| 364692/371472 [6:51:29<32:14,  3.50it/s]
 98%|█████████▊| 364693/371472 [6:51:29<32:00,  3.53it/s]
 98%|█████████▊| 364694/371472 [6:51:29<31:39,  3.57it/s]
 98%|█████████▊| 364695/371472 [6:51:30<32:15,  3.50it/s]
 98%|█████████▊| 364696/371472 [6:51:30<32:16,  3.50it/s]
 98%|█████████▊| 364697/371472 [6:51:30<32:04,  3.52it/s]
 98%|█████████▊| 364698/371472 [6:51:30<30:31,  3.70it/s]
 98%|█████████▊| 364699/371472 [6:51:31<29:43,  3.80it/s]
 98%|█████████▊| 364700/371472 [6:51:31<29:48,  3.79it/s]
                                                         
{'loss': 2.6446, 'learning_rate': 1.1641599689715355e-07, 'epoch': 15.71}
+
 98%|█████████▊| 364700/371472 [6:51:31<29:48,  3.79it/s]
 98%|█████████▊| 364701/371472 [6:51:31<30:44,  3.67it/s]
 98%|█████████▊| 364702/371472 [6:51:31<31:01,  3.64it/s]
 98%|█████████▊| 364703/371472 [6:51:32<31:47,  3.55it/s]
 98%|█████████▊| 364704/371472 [6:51:32<30:26,  3.71it/s]
 98%|█████████▊| 364705/371472 [6:51:32<31:35,  3.57it/s]
 98%|█████████▊| 364706/371472 [6:51:33<30:14,  3.73it/s]
 98%|█████████▊| 364707/371472 [6:51:33<30:53,  3.65it/s]
 98%|█████████▊| 364708/371472 [6:51:33<30:58,  3.64it/s]
 98%|█████████▊| 364709/371472 [6:51:33<30:58,  3.64it/s]
 98%|█████████▊| 364710/371472 [6:51:34<30:28,  3.70it/s]
 98%|█████████▊| 364711/371472 [6:51:34<30:28,  3.70it/s]
 98%|█████████▊| 364712/371472 [6:51:34<34:39,  3.25it/s]
 98%|█████████▊| 364713/371472 [6:51:35<33:13,  3.39it/s]
 98%|█████████▊| 364714/371472 [6:51:35<31:40,  3.56it/s]
 98%|█████████▊| 364715/371472 [6:51:35<33:18,  3.38it/s]
 98%|█████████▊| 364716/371472 [6:51:35<31:41,  3.55it/s]
 98%|█████████▊| 364717/371472 [6:51:36<30:54,  3.64it/s]
 98%|█████████▊| 364718/371472 [6:51:36<31:31,  3.57it/s]
 98%|█████████▊| 364719/371472 [6:51:36<32:34,  3.46it/s]
 98%|█████████▊| 364720/371472 [6:51:37<31:36,  3.56it/s]
                                                         
{'loss': 2.4994, 'learning_rate': 1.163675149216747e-07, 'epoch': 15.71}
+
 98%|█████████▊| 364720/371472 [6:51:37<31:36,  3.56it/s]
 98%|█████████▊| 364721/371472 [6:51:37<32:13,  3.49it/s]
 98%|█████████▊| 364722/371472 [6:51:37<31:18,  3.59it/s]
 98%|█████████▊| 364723/371472 [6:51:37<33:20,  3.37it/s]
 98%|█████████▊| 364724/371472 [6:51:38<36:45,  3.06it/s]
 98%|█████████▊| 364725/371472 [6:51:38<36:44,  3.06it/s]
 98%|█████████▊| 364726/371472 [6:51:38<36:05,  3.11it/s]
 98%|█████████▊| 364727/371472 [6:51:39<33:39,  3.34it/s]
 98%|█████████▊| 364728/371472 [6:51:39<36:54,  3.05it/s]
 98%|█████████▊| 364729/371472 [6:51:39<35:09,  3.20it/s]
 98%|█████████▊| 364730/371472 [6:51:40<33:49,  3.32it/s]
 98%|█████████▊| 364731/371472 [6:51:40<34:51,  3.22it/s]
 98%|█████████▊| 364732/371472 [6:51:40<34:51,  3.22it/s]
 98%|█████████▊| 364733/371472 [6:51:41<33:11,  3.38it/s]
 98%|█████████▊| 364734/371472 [6:51:41<32:08,  3.49it/s]
 98%|█████████▊| 364735/371472 [6:51:41<34:52,  3.22it/s]
 98%|█████████▊| 364736/371472 [6:51:42<36:52,  3.05it/s]
 98%|█████████▊| 364737/371472 [6:51:42<35:23,  3.17it/s]
 98%|█████████▊| 364738/371472 [6:51:42<33:33,  3.34it/s]
 98%|█████████▊| 364739/371472 [6:51:42<33:29,  3.35it/s]
 98%|█████████▊| 364740/371472 [6:51:43<31:50,  3.52it/s]
                                                         
{'loss': 2.4673, 'learning_rate': 1.1631903294619576e-07, 'epoch': 15.71}
+
 98%|█████████▊| 364740/371472 [6:51:43<31:50,  3.52it/s]
 98%|█████████▊| 364741/371472 [6:51:43<31:51,  3.52it/s]
 98%|█████████▊| 364742/371472 [6:51:43<30:51,  3.63it/s]
 98%|█████████▊| 364743/371472 [6:51:43<30:44,  3.65it/s]
 98%|█████████▊| 364744/371472 [6:51:44<29:29,  3.80it/s]
 98%|█████████▊| 364745/371472 [6:51:44<31:55,  3.51it/s]
 98%|█████████▊| 364746/371472 [6:51:44<32:07,  3.49it/s]
 98%|█████████▊| 364747/371472 [6:51:45<31:41,  3.54it/s]
 98%|█████████▊| 364748/371472 [6:51:45<32:04,  3.49it/s]
 98%|█████████▊| 364749/371472 [6:51:45<31:44,  3.53it/s]
 98%|█████████▊| 364750/371472 [6:51:45<31:11,  3.59it/s]
 98%|█████████▊| 364751/371472 [6:51:46<32:20,  3.46it/s]
 98%|█████████▊| 364752/371472 [6:51:46<31:12,  3.59it/s]
 98%|█████████▊| 364753/371472 [6:51:46<32:21,  3.46it/s]
 98%|█████████▊| 364754/371472 [6:51:47<31:48,  3.52it/s]
 98%|█████████▊| 364755/371472 [6:51:47<30:37,  3.66it/s]
 98%|█████████▊| 364756/371472 [6:51:47<30:10,  3.71it/s]
 98%|█████████▊| 364757/371472 [6:51:47<32:16,  3.47it/s]
 98%|█████████▊| 364758/371472 [6:51:48<31:45,  3.52it/s]
 98%|█████████▊| 364759/371472 [6:51:48<30:22,  3.68it/s]
 98%|█████████▊| 364760/371472 [6:51:48<30:12,  3.70it/s]
                                                         
{'loss': 2.5598, 'learning_rate': 1.1627055097071691e-07, 'epoch': 15.71}
+
 98%|█████████▊| 364760/371472 [6:51:48<30:12,  3.70it/s]
 98%|█████████▊| 364761/371472 [6:51:48<29:49,  3.75it/s]
 98%|█████████▊| 364762/371472 [6:51:49<30:13,  3.70it/s]
 98%|█████████▊| 364763/371472 [6:51:49<29:48,  3.75it/s]
 98%|█████████▊| 364764/371472 [6:51:49<29:25,  3.80it/s]
 98%|█████████▊| 364765/371472 [6:51:50<29:57,  3.73it/s]
 98%|█████████▊| 364766/371472 [6:51:50<29:31,  3.79it/s]
 98%|█████████▊| 364767/371472 [6:51:50<32:38,  3.42it/s]
 98%|█████████▊| 364768/371472 [6:51:50<30:59,  3.60it/s]
 98%|█████████▊| 364769/371472 [6:51:51<31:04,  3.60it/s]
 98%|█████████▊| 364770/371472 [6:51:51<30:05,  3.71it/s]
 98%|█████████▊| 364771/371472 [6:51:51<30:52,  3.62it/s]
 98%|█████████▊| 364772/371472 [6:51:51<30:27,  3.67it/s]
 98%|█████████▊| 364773/371472 [6:51:52<30:17,  3.69it/s]
 98%|█████████▊| 364774/371472 [6:51:52<31:20,  3.56it/s]
 98%|█████████▊| 364775/371472 [6:51:52<31:10,  3.58it/s]
 98%|█████████▊| 364776/371472 [6:51:53<30:48,  3.62it/s]
 98%|█████████▊| 364777/371472 [6:51:53<31:08,  3.58it/s]
 98%|█████████▊| 364778/371472 [6:51:53<32:51,  3.39it/s]
 98%|█████████▊| 364779/371472 [6:51:53<32:47,  3.40it/s]
 98%|█████████▊| 364780/371472 [6:51:54<31:15,  3.57it/s]
                                                         
{'loss': 2.6043, 'learning_rate': 1.1622206899523797e-07, 'epoch': 15.71}
+
 98%|█████████▊| 364780/371472 [6:51:54<31:15,  3.57it/s]
 98%|█████████▊| 364781/371472 [6:51:54<30:08,  3.70it/s]
 98%|█████████▊| 364782/371472 [6:51:54<29:30,  3.78it/s]
 98%|█████████▊| 364783/371472 [6:51:54<29:03,  3.84it/s]
 98%|█████████▊| 364784/371472 [6:51:55<28:44,  3.88it/s]
 98%|█████████▊| 364785/371472 [6:51:55<28:03,  3.97it/s]
 98%|█████████▊| 364786/371472 [6:51:55<30:10,  3.69it/s]
 98%|█████████▊| 364787/371472 [6:51:56<33:13,  3.35it/s]
 98%|█████████▊| 364788/371472 [6:51:56<31:31,  3.53it/s]
 98%|█████████▊| 364789/371472 [6:51:56<33:54,  3.29it/s]
 98%|█████████▊| 364790/371472 [6:51:57<32:39,  3.41it/s]
 98%|█████████▊| 364791/371472 [6:51:57<32:31,  3.42it/s]
 98%|█████████▊| 364792/371472 [6:51:57<31:21,  3.55it/s]
 98%|█████████▊| 364793/371472 [6:51:57<31:03,  3.58it/s]
 98%|█████████▊| 364794/371472 [6:51:58<30:41,  3.63it/s]
 98%|█████████▊| 364795/371472 [6:51:58<30:36,  3.63it/s]
 98%|█████████▊| 364796/371472 [6:51:58<30:08,  3.69it/s]
 98%|█████████▊| 364797/371472 [6:51:58<29:31,  3.77it/s]
 98%|█████████▊| 364798/371472 [6:51:59<32:14,  3.45it/s]
 98%|█████████▊| 364799/371472 [6:51:59<31:11,  3.57it/s]
 98%|█████████▊| 364800/371472 [6:51:59<34:28,  3.23it/s]
                                                         
{'loss': 2.6243, 'learning_rate': 1.1617358701975914e-07, 'epoch': 15.71}
+
 98%|█████████▊| 364800/371472 [6:51:59<34:28,  3.23it/s]
 98%|█████████▊| 364801/371472 [6:52:00<33:43,  3.30it/s]
 98%|█████████▊| 364802/371472 [6:52:00<32:35,  3.41it/s]
 98%|█████████▊| 364803/371472 [6:52:00<31:20,  3.55it/s]
 98%|█████████▊| 364804/371472 [6:52:00<30:16,  3.67it/s]
 98%|█████████▊| 364805/371472 [6:52:01<29:27,  3.77it/s]
 98%|█████████▊| 364806/371472 [6:52:01<28:56,  3.84it/s]
 98%|█████████▊| 364807/371472 [6:52:01<28:11,  3.94it/s]
 98%|█████████▊| 364808/371472 [6:52:01<29:43,  3.74it/s]
 98%|█████████▊| 364809/371472 [6:52:02<29:52,  3.72it/s]
 98%|█████████▊| 364810/371472 [6:52:02<29:03,  3.82it/s]
 98%|█████████▊| 364811/371472 [6:52:02<29:06,  3.81it/s]
 98%|█████████▊| 364812/371472 [6:52:03<31:02,  3.58it/s]
 98%|█████████▊| 364813/371472 [6:52:03<30:19,  3.66it/s]
 98%|█████████▊| 364814/371472 [6:52:03<29:30,  3.76it/s]
 98%|█████████▊| 364815/371472 [6:52:03<29:04,  3.82it/s]
 98%|█████████▊| 364816/371472 [6:52:04<30:02,  3.69it/s]
 98%|█████████▊| 364817/371472 [6:52:04<30:10,  3.68it/s]
 98%|█████████▊| 364818/371472 [6:52:04<29:55,  3.71it/s]
 98%|█████████▊| 364819/371472 [6:52:04<29:56,  3.70it/s]
 98%|█████████▊| 364820/371472 [6:52:05<29:22,  3.77it/s]
                                                         
{'loss': 2.5401, 'learning_rate': 1.161251050442802e-07, 'epoch': 15.71}
+
 98%|█████████▊| 364820/371472 [6:52:05<29:22,  3.77it/s]
 98%|█████████▊| 364821/371472 [6:52:05<29:10,  3.80it/s]
 98%|█████████▊| 364822/371472 [6:52:05<30:08,  3.68it/s]
 98%|█████████▊| 364823/371472 [6:52:06<31:23,  3.53it/s]
 98%|█████████▊| 364824/371472 [6:52:06<30:23,  3.65it/s]
 98%|█████████▊| 364825/371472 [6:52:06<29:38,  3.74it/s]
 98%|█████████▊| 364826/371472 [6:52:06<33:14,  3.33it/s]
 98%|█████████▊| 364827/371472 [6:52:07<32:18,  3.43it/s]
 98%|█████████▊| 364828/371472 [6:52:07<31:22,  3.53it/s]
 98%|█████████▊| 364829/371472 [6:52:07<30:34,  3.62it/s]
 98%|█████████▊| 364830/371472 [6:52:08<31:53,  3.47it/s]
 98%|█████████▊| 364831/371472 [6:52:08<30:40,  3.61it/s]
 98%|█████████▊| 364832/371472 [6:52:08<29:52,  3.70it/s]
 98%|█████████▊| 364833/371472 [6:52:08<30:36,  3.61it/s]
 98%|█████████▊| 364834/371472 [6:52:09<31:29,  3.51it/s]
 98%|█████████▊| 364835/371472 [6:52:09<32:15,  3.43it/s]
 98%|█████████▊| 364836/371472 [6:52:09<32:03,  3.45it/s]
 98%|█████████▊| 364837/371472 [6:52:10<31:26,  3.52it/s]
 98%|█████████▊| 364838/371472 [6:52:10<32:29,  3.40it/s]
 98%|█████████▊| 364839/371472 [6:52:10<31:02,  3.56it/s]
 98%|█████████▊| 364840/371472 [6:52:10<30:21,  3.64it/s]
                                                         
{'loss': 2.627, 'learning_rate': 1.1607662306880135e-07, 'epoch': 15.71}
+
 98%|█████████▊| 364840/371472 [6:52:10<30:21,  3.64it/s]
 98%|█████████▊| 364841/371472 [6:52:11<30:28,  3.63it/s]
 98%|█████████▊| 364842/371472 [6:52:11<29:39,  3.73it/s]
 98%|█████████▊| 364843/371472 [6:52:11<28:54,  3.82it/s]
 98%|█████████▊| 364844/371472 [6:52:11<28:36,  3.86it/s]
 98%|█████████▊| 364845/371472 [6:52:12<28:15,  3.91it/s]
 98%|█████████▊| 364846/371472 [6:52:12<31:15,  3.53it/s]
 98%|█████████▊| 364847/371472 [6:52:12<30:33,  3.61it/s]
 98%|█████████▊| 364848/371472 [6:52:12<29:34,  3.73it/s]
 98%|█████████▊| 364849/371472 [6:52:13<30:11,  3.66it/s]
 98%|█████████▊| 364850/371472 [6:52:13<32:47,  3.37it/s]
 98%|█████████▊| 364851/371472 [6:52:13<32:52,  3.36it/s]
 98%|█████████▊| 364852/371472 [6:52:14<33:51,  3.26it/s]
 98%|█████████▊| 364853/371472 [6:52:14<31:51,  3.46it/s]
 98%|█████████▊| 364854/371472 [6:52:14<31:28,  3.50it/s]
 98%|█████████▊| 364855/371472 [6:52:15<30:48,  3.58it/s]
 98%|█████████▊| 364856/371472 [6:52:15<32:29,  3.39it/s]
 98%|█████████▊| 364857/371472 [6:52:15<32:25,  3.40it/s]
 98%|█████████▊| 364858/371472 [6:52:15<32:26,  3.40it/s]
 98%|█████████▊| 364859/371472 [6:52:16<30:21,  3.63it/s]
 98%|█████████▊| 364860/371472 [6:52:16<32:08,  3.43it/s]
                                                         
{'loss': 2.5466, 'learning_rate': 1.160281410933224e-07, 'epoch': 15.72}
+
 98%|█████████▊| 364860/371472 [6:52:16<32:08,  3.43it/s]
 98%|█████████▊| 364861/371472 [6:52:16<33:04,  3.33it/s]
 98%|█████████▊| 364862/371472 [6:52:17<30:59,  3.55it/s]
 98%|█████████▊| 364863/371472 [6:52:17<30:28,  3.61it/s]
 98%|█████████▊| 364864/371472 [6:52:17<35:44,  3.08it/s]
 98%|█████████▊| 364865/371472 [6:52:18<34:13,  3.22it/s]
 98%|█████████▊| 364866/371472 [6:52:18<33:46,  3.26it/s]
 98%|█████████▊| 364867/371472 [6:52:18<32:07,  3.43it/s]
 98%|█████████▊| 364868/371472 [6:52:18<31:49,  3.46it/s]
 98%|█████████▊| 364869/371472 [6:52:19<30:43,  3.58it/s]
 98%|█████████▊| 364870/371472 [6:52:19<30:31,  3.61it/s]
 98%|█████████▊| 364871/371472 [6:52:19<31:24,  3.50it/s]
 98%|█████████▊| 364872/371472 [6:52:20<32:51,  3.35it/s]
 98%|█████████▊| 364873/371472 [6:52:20<32:25,  3.39it/s]
 98%|█████████▊| 364874/371472 [6:52:20<32:28,  3.39it/s]
 98%|█████████▊| 364875/371472 [6:52:20<31:37,  3.48it/s]
 98%|█████████▊| 364876/371472 [6:52:21<30:34,  3.60it/s]
 98%|█████████▊| 364877/371472 [6:52:21<29:46,  3.69it/s]
 98%|█████████▊| 364878/371472 [6:52:21<30:17,  3.63it/s]
 98%|█████████▊| 364879/371472 [6:52:21<30:23,  3.61it/s]
 98%|█████████▊| 364880/371472 [6:52:22<30:55,  3.55it/s]
                                                         
{'loss': 2.5927, 'learning_rate': 1.1597965911784356e-07, 'epoch': 15.72}
+
 98%|█████████▊| 364880/371472 [6:52:22<30:55,  3.55it/s]
 98%|█████████▊| 364881/371472 [6:52:22<31:02,  3.54it/s]
 98%|█████████▊| 364882/371472 [6:52:22<30:19,  3.62it/s]
 98%|█████████▊| 364883/371472 [6:52:23<29:58,  3.66it/s]
 98%|█████████▊| 364884/371472 [6:52:23<30:42,  3.57it/s]
 98%|█████████▊| 364885/371472 [6:52:23<30:13,  3.63it/s]
 98%|█████████▊| 364886/371472 [6:52:23<32:48,  3.35it/s]
 98%|█████████▊| 364887/371472 [6:52:24<31:39,  3.47it/s]
 98%|█████████▊| 364888/371472 [6:52:24<30:23,  3.61it/s]
 98%|█████████▊| 364889/371472 [6:52:24<29:23,  3.73it/s]
 98%|█████████▊| 364890/371472 [6:52:25<31:04,  3.53it/s]
 98%|█████████▊| 364891/371472 [6:52:25<30:47,  3.56it/s]
 98%|█████████▊| 364892/371472 [6:52:25<33:41,  3.25it/s]
 98%|█████████▊| 364893/371472 [6:52:26<32:57,  3.33it/s]
 98%|█████████▊| 364894/371472 [6:52:26<31:08,  3.52it/s]
 98%|█████████▊| 364895/371472 [6:52:26<30:08,  3.64it/s]
 98%|█████████▊| 364896/371472 [6:52:26<30:23,  3.61it/s]
 98%|█████████▊| 364897/371472 [6:52:27<30:55,  3.54it/s]
 98%|█████████▊| 364898/371472 [6:52:27<31:42,  3.45it/s]
 98%|█████████▊| 364899/371472 [6:52:27<31:19,  3.50it/s]
 98%|█████████▊| 364900/371472 [6:52:27<31:57,  3.43it/s]
                                                         
{'loss': 2.628, 'learning_rate': 1.1593117714236463e-07, 'epoch': 15.72}
+
 98%|█████████▊| 364900/371472 [6:52:27<31:57,  3.43it/s]
 98%|█████████▊| 364901/371472 [6:52:28<31:06,  3.52it/s]
 98%|█████████▊| 364902/371472 [6:52:28<30:46,  3.56it/s]
 98%|█████████▊| 364903/371472 [6:52:28<31:24,  3.49it/s]
 98%|█████████▊| 364904/371472 [6:52:29<33:55,  3.23it/s]
 98%|█████████▊| 364905/371472 [6:52:29<32:35,  3.36it/s]
 98%|█████████▊| 364906/371472 [6:52:29<31:01,  3.53it/s]
 98%|█████████▊| 364907/371472 [6:52:29<30:26,  3.59it/s]
 98%|█████████▊| 364908/371472 [6:52:30<32:38,  3.35it/s]
 98%|█████████▊| 364909/371472 [6:52:30<32:08,  3.40it/s]
 98%|█████████▊| 364910/371472 [6:52:30<31:51,  3.43it/s]
 98%|█████████▊| 364911/371472 [6:52:31<32:08,  3.40it/s]
 98%|█████████▊| 364912/371472 [6:52:31<32:41,  3.34it/s]
 98%|█████████▊| 364913/371472 [6:52:31<36:52,  2.96it/s]
 98%|█████████▊| 364914/371472 [6:52:32<33:28,  3.26it/s]
 98%|█████████▊| 364915/371472 [6:52:32<31:56,  3.42it/s]
 98%|█████████▊| 364916/371472 [6:52:32<31:03,  3.52it/s]
 98%|█████████▊| 364917/371472 [6:52:32<31:43,  3.44it/s]
 98%|█████████▊| 364918/371472 [6:52:33<31:28,  3.47it/s]
 98%|█████████▊| 364919/371472 [6:52:33<30:33,  3.57it/s]
 98%|█████████▊| 364920/371472 [6:52:33<29:12,  3.74it/s]
                                                         
{'loss': 2.6255, 'learning_rate': 1.1588269516688567e-07, 'epoch': 15.72}
+
 98%|█████████▊| 364920/371472 [6:52:33<29:12,  3.74it/s]
 98%|█████████▊| 364921/371472 [6:52:34<30:02,  3.63it/s]
 98%|█████████▊| 364922/371472 [6:52:34<30:02,  3.63it/s]
 98%|█████████▊| 364923/371472 [6:52:34<29:32,  3.70it/s]
 98%|█████████▊| 364924/371472 [6:52:34<30:40,  3.56it/s]
 98%|█████████▊| 364925/371472 [6:52:35<29:28,  3.70it/s]
 98%|█████████▊| 364926/371472 [6:52:35<28:55,  3.77it/s]
 98%|█████████▊| 364927/371472 [6:52:35<31:20,  3.48it/s]
 98%|█████████▊| 364928/371472 [6:52:36<30:50,  3.54it/s]
 98%|█████████▊| 364929/371472 [6:52:36<30:33,  3.57it/s]
 98%|█████████▊| 364930/371472 [6:52:36<29:57,  3.64it/s]
 98%|█████████▊| 364931/371472 [6:52:36<30:24,  3.59it/s]
 98%|█████████▊| 364932/371472 [6:52:37<33:58,  3.21it/s]
 98%|█████████▊| 364933/371472 [6:52:37<32:31,  3.35it/s]
 98%|█████████▊| 364934/371472 [6:52:37<31:18,  3.48it/s]
 98%|█████████▊| 364935/371472 [6:52:38<30:36,  3.56it/s]
 98%|█████████▊| 364936/371472 [6:52:38<29:34,  3.68it/s]
 98%|█████████▊| 364937/371472 [6:52:38<31:37,  3.44it/s]
 98%|█████████▊| 364938/371472 [6:52:38<30:44,  3.54it/s]
 98%|█████████▊| 364939/371472 [6:52:39<33:23,  3.26it/s]
 98%|█████████▊| 364940/371472 [6:52:39<31:16,  3.48it/s]
                                                         
{'loss': 2.6177, 'learning_rate': 1.1583421319140684e-07, 'epoch': 15.72}
+
 98%|█████████▊| 364940/371472 [6:52:39<31:16,  3.48it/s]
 98%|█████████▊| 364941/371472 [6:52:39<32:19,  3.37it/s]
 98%|█████████▊| 364942/371472 [6:52:40<31:56,  3.41it/s]
 98%|█████████▊| 364943/371472 [6:52:40<30:52,  3.52it/s]
 98%|█████████▊| 364944/371472 [6:52:40<30:12,  3.60it/s]
 98%|█████████▊| 364945/371472 [6:52:40<29:47,  3.65it/s]
 98%|█████████▊| 364946/371472 [6:52:41<29:56,  3.63it/s]
 98%|█████████▊| 364947/371472 [6:52:41<29:29,  3.69it/s]
 98%|█████████▊| 364948/371472 [6:52:41<29:29,  3.69it/s]
 98%|█████████▊| 364949/371472 [6:52:41<30:01,  3.62it/s]
 98%|█████████▊| 364950/371472 [6:52:42<29:26,  3.69it/s]
 98%|█████████▊| 364951/371472 [6:52:42<29:41,  3.66it/s]
 98%|█████████▊| 364952/371472 [6:52:42<28:24,  3.83it/s]
 98%|█████████▊| 364953/371472 [6:52:43<29:13,  3.72it/s]
 98%|█████████▊| 364954/371472 [6:52:43<28:44,  3.78it/s]
 98%|█████████▊| 364955/371472 [6:52:43<29:27,  3.69it/s]
 98%|█████████▊| 364956/371472 [6:52:43<30:24,  3.57it/s]
 98%|█████████▊| 364957/371472 [6:52:44<33:42,  3.22it/s]
 98%|█████████▊| 364958/371472 [6:52:44<32:52,  3.30it/s]
 98%|█████████▊| 364959/371472 [6:52:44<33:31,  3.24it/s]
 98%|█████████▊| 364960/371472 [6:52:45<31:21,  3.46it/s]
                                                         
{'loss': 2.7174, 'learning_rate': 1.157857312159279e-07, 'epoch': 15.72}
+
 98%|█████████▊| 364960/371472 [6:52:45<31:21,  3.46it/s]
 98%|█████████▊| 364961/371472 [6:52:45<29:49,  3.64it/s]
 98%|█████████▊| 364962/371472 [6:52:45<28:47,  3.77it/s]
 98%|█████████▊| 364963/371472 [6:52:45<28:41,  3.78it/s]
 98%|█████████▊| 364964/371472 [6:52:46<30:35,  3.54it/s]
 98%|█████████▊| 364965/371472 [6:52:46<30:17,  3.58it/s]
 98%|█████████▊| 364966/371472 [6:52:46<31:13,  3.47it/s]
 98%|█████████▊| 364967/371472 [6:52:47<31:14,  3.47it/s]
 98%|█████████▊| 364968/371472 [6:52:47<30:20,  3.57it/s]
 98%|█████████▊| 364969/371472 [6:52:47<29:44,  3.65it/s]
 98%|█████████▊| 364970/371472 [6:52:47<29:16,  3.70it/s]
 98%|█████████▊| 364971/371472 [6:52:48<28:31,  3.80it/s]
 98%|█████████▊| 364972/371472 [6:52:48<30:01,  3.61it/s]
 98%|█████████▊| 364973/371472 [6:52:48<33:52,  3.20it/s]
 98%|█████████▊| 364974/371472 [6:52:49<32:42,  3.31it/s]
 98%|█████████▊| 364975/371472 [6:52:49<31:39,  3.42it/s]
 98%|█████████▊| 364976/371472 [6:52:49<30:36,  3.54it/s]
 98%|█████████▊| 364977/371472 [6:52:49<29:36,  3.66it/s]
 98%|█████████▊| 364978/371472 [6:52:50<28:26,  3.81it/s]
 98%|█████████▊| 364979/371472 [6:52:50<28:11,  3.84it/s]
 98%|█████████▊| 364980/371472 [6:52:50<28:53,  3.74it/s]
                                                         
{'loss': 2.5315, 'learning_rate': 1.1573724924044905e-07, 'epoch': 15.72}
+
 98%|█████████▊| 364980/371472 [6:52:50<28:53,  3.74it/s]
 98%|█████████▊| 364981/371472 [6:52:50<31:52,  3.39it/s]
 98%|█████████▊| 364982/371472 [6:52:51<32:04,  3.37it/s]
 98%|█████████▊| 364983/371472 [6:52:51<32:33,  3.32it/s]
 98%|█████████▊| 364984/371472 [6:52:51<30:59,  3.49it/s]
 98%|█████████▊| 364985/371472 [6:52:52<31:16,  3.46it/s]
 98%|█████████▊| 364986/371472 [6:52:52<30:21,  3.56it/s]
 98%|█████████▊| 364987/371472 [6:52:52<30:07,  3.59it/s]
 98%|█████████▊| 364988/371472 [6:52:52<31:36,  3.42it/s]
 98%|█████████▊| 364989/371472 [6:52:53<32:50,  3.29it/s]
 98%|█████████▊| 364990/371472 [6:52:53<33:45,  3.20it/s]
 98%|█████████▊| 364991/371472 [6:52:54<36:03,  3.00it/s]
 98%|█████████▊| 364992/371472 [6:52:54<34:05,  3.17it/s]
 98%|█████████▊| 364993/371472 [6:52:54<31:37,  3.41it/s]
 98%|█████████▊| 364994/371472 [6:52:54<33:21,  3.24it/s]
 98%|█████████▊| 364995/371472 [6:52:55<32:37,  3.31it/s]
 98%|█████████▊| 364996/371472 [6:52:55<32:10,  3.35it/s]
 98%|█████████▊| 364997/371472 [6:52:55<30:34,  3.53it/s]
 98%|█████████▊| 364998/371472 [6:52:56<31:03,  3.47it/s]
 98%|█████████▊| 364999/371472 [6:52:56<32:19,  3.34it/s]
 98%|█████████▊| 365000/371472 [6:52:56<31:36,  3.41it/s]
                                                         
{'loss': 2.5428, 'learning_rate': 1.1568876726497012e-07, 'epoch': 15.72}
+
 98%|█████████▊| 365000/371472 [6:52:56<31:36,  3.41it/s]
 98%|█████████▊| 365001/371472 [6:52:56<30:14,  3.57it/s]
 98%|█████████▊| 365002/371472 [6:52:57<28:53,  3.73it/s]
 98%|█████████▊| 365003/371472 [6:52:57<31:18,  3.44it/s]
 98%|█████████▊| 365004/371472 [6:52:57<30:31,  3.53it/s]
 98%|█████████▊| 365005/371472 [6:52:57<28:58,  3.72it/s]
 98%|█████████▊| 365006/371472 [6:52:58<29:00,  3.72it/s]
 98%|█████████▊| 365007/371472 [6:52:58<29:37,  3.64it/s]
 98%|█████████▊| 365008/371472 [6:52:58<28:48,  3.74it/s]
 98%|█████████▊| 365009/371472 [6:52:59<30:58,  3.48it/s]
 98%|█████████▊| 365010/371472 [6:52:59<30:20,  3.55it/s]
 98%|█████████▊| 365011/371472 [6:52:59<29:48,  3.61it/s]
 98%|█████████▊| 365012/371472 [6:52:59<30:50,  3.49it/s]
 98%|█████████▊| 365013/371472 [6:53:00<29:53,  3.60it/s]
 98%|█████████▊| 365014/371472 [6:53:00<29:49,  3.61it/s]
 98%|█████████▊| 365015/371472 [6:53:00<28:49,  3.73it/s]
 98%|█████████▊| 365016/371472 [6:53:00<28:38,  3.76it/s]
 98%|█████████▊| 365017/371472 [6:53:01<28:27,  3.78it/s]
 98%|█████████▊| 365018/371472 [6:53:01<27:59,  3.84it/s]
 98%|█████████▊| 365019/371472 [6:53:01<27:38,  3.89it/s]
 98%|█████████▊| 365020/371472 [6:53:02<28:19,  3.80it/s]
                                                         
{'loss': 2.5803, 'learning_rate': 1.1564028528949127e-07, 'epoch': 15.72}
+
 98%|█████████▊| 365020/371472 [6:53:02<28:19,  3.80it/s]
 98%|█████████▊| 365021/371472 [6:53:02<29:07,  3.69it/s]
 98%|█████████▊| 365022/371472 [6:53:02<31:18,  3.43it/s]
 98%|█████████▊| 365023/371472 [6:53:02<30:57,  3.47it/s]
 98%|█████████▊| 365024/371472 [6:53:03<31:29,  3.41it/s]
 98%|█████████▊| 365025/371472 [6:53:03<31:17,  3.43it/s]
 98%|█████████▊| 365026/371472 [6:53:03<32:53,  3.27it/s]
 98%|█████████▊| 365027/371472 [6:53:04<33:02,  3.25it/s]
 98%|█████████▊| 365028/371472 [6:53:04<32:30,  3.30it/s]
 98%|█████████▊| 365029/371472 [6:53:04<30:33,  3.51it/s]
 98%|█████████▊| 365030/371472 [6:53:04<31:10,  3.44it/s]
 98%|█████████▊| 365031/371472 [6:53:05<32:09,  3.34it/s]
 98%|█████████▊| 365032/371472 [6:53:05<33:05,  3.24it/s]
 98%|█████████▊| 365033/371472 [6:53:05<32:31,  3.30it/s]
 98%|█████████▊| 365034/371472 [6:53:06<32:36,  3.29it/s]
 98%|█████████▊| 365035/371472 [6:53:06<31:42,  3.38it/s]
 98%|█████████▊| 365036/371472 [6:53:06<30:41,  3.50it/s]
 98%|█████████▊| 365037/371472 [6:53:07<31:03,  3.45it/s]
 98%|█████████▊| 365038/371472 [6:53:07<31:03,  3.45it/s]
 98%|█████████▊| 365039/371472 [6:53:07<35:07,  3.05it/s]
 98%|█████████▊| 365040/371472 [6:53:08<34:05,  3.15it/s]
                                                         
{'loss': 2.5011, 'learning_rate': 1.1559180331401233e-07, 'epoch': 15.72}
+
 98%|█████████▊| 365040/371472 [6:53:08<34:05,  3.15it/s]
 98%|█████████▊| 365041/371472 [6:53:08<32:35,  3.29it/s]
 98%|█████████▊| 365042/371472 [6:53:08<31:52,  3.36it/s]
 98%|█████████▊| 365043/371472 [6:53:08<33:39,  3.18it/s]
 98%|█████████▊| 365044/371472 [6:53:09<33:13,  3.23it/s]
 98%|█████████▊| 365045/371472 [6:53:09<32:02,  3.34it/s]
 98%|█████████▊| 365046/371472 [6:53:09<33:05,  3.24it/s]
 98%|█████████▊| 365047/371472 [6:53:10<32:46,  3.27it/s]
 98%|█████████▊| 365048/371472 [6:53:10<31:57,  3.35it/s]
 98%|█████████▊| 365049/371472 [6:53:10<29:52,  3.58it/s]
 98%|█████████▊| 365050/371472 [6:53:10<29:59,  3.57it/s]
 98%|█████████▊| 365051/371472 [6:53:11<28:56,  3.70it/s]
 98%|█████████▊| 365052/371472 [6:53:11<29:17,  3.65it/s]
 98%|█████████▊| 365053/371472 [6:53:11<30:54,  3.46it/s]
 98%|█████████▊| 365054/371472 [6:53:12<31:10,  3.43it/s]
 98%|█████████▊| 365055/371472 [6:53:12<31:18,  3.42it/s]
 98%|█████████▊| 365056/371472 [6:53:12<30:34,  3.50it/s]
 98%|█████████▊| 365057/371472 [6:53:13<33:04,  3.23it/s]
 98%|█████████▊| 365058/371472 [6:53:13<31:40,  3.37it/s]
 98%|█████████▊| 365059/371472 [6:53:13<31:04,  3.44it/s]
 98%|█████████▊| 365060/371472 [6:53:13<30:55,  3.46it/s]
                                                         
{'loss': 2.5331, 'learning_rate': 1.1554332133853349e-07, 'epoch': 15.72}
+
 98%|█████████▊| 365060/371472 [6:53:13<30:55,  3.46it/s]
 98%|█████████▊| 365061/371472 [6:53:14<29:59,  3.56it/s]
 98%|█████████▊| 365062/371472 [6:53:14<31:09,  3.43it/s]
 98%|█████████▊| 365063/371472 [6:53:14<29:37,  3.61it/s]
 98%|█████████▊| 365064/371472 [6:53:14<28:46,  3.71it/s]
 98%|█████████▊| 365065/371472 [6:53:15<28:32,  3.74it/s]
 98%|█████████▊| 365066/371472 [6:53:15<28:47,  3.71it/s]
 98%|█████████▊| 365067/371472 [6:53:15<28:34,  3.74it/s]
 98%|█████████▊| 365068/371472 [6:53:16<28:19,  3.77it/s]
 98%|█████████▊| 365069/371472 [6:53:16<28:18,  3.77it/s]
 98%|█████████▊| 365070/371472 [6:53:16<29:14,  3.65it/s]
 98%|█████████▊| 365071/371472 [6:53:16<28:46,  3.71it/s]
 98%|█████████▊| 365072/371472 [6:53:17<28:19,  3.77it/s]
 98%|█████████▊| 365073/371472 [6:53:17<27:40,  3.85it/s]
 98%|█████████▊| 365074/371472 [6:53:17<27:36,  3.86it/s]
 98%|█████████▊| 365075/371472 [6:53:17<27:41,  3.85it/s]
 98%|█████████▊| 365076/371472 [6:53:18<29:04,  3.67it/s]
 98%|█████████▊| 365077/371472 [6:53:18<30:35,  3.48it/s]
 98%|█████████▊| 365078/371472 [6:53:18<30:12,  3.53it/s]
 98%|█████████▊| 365079/371472 [6:53:19<29:29,  3.61it/s]
 98%|█████████▊| 365080/371472 [6:53:19<32:43,  3.26it/s]
                                                         
{'loss': 2.5868, 'learning_rate': 1.1549483936305454e-07, 'epoch': 15.72}
+
 98%|█████████▊| 365080/371472 [6:53:19<32:43,  3.26it/s]
 98%|█████████▊| 365081/371472 [6:53:19<31:42,  3.36it/s]
 98%|█████████▊| 365082/371472 [6:53:19<29:50,  3.57it/s]
 98%|█████████▊| 365083/371472 [6:53:20<30:33,  3.48it/s]
 98%|█████████▊| 365084/371472 [6:53:20<30:10,  3.53it/s]
 98%|█████████▊| 365085/371472 [6:53:20<29:03,  3.66it/s]
 98%|█████████▊| 365086/371472 [6:53:21<32:02,  3.32it/s]
 98%|█████████▊| 365087/371472 [6:53:21<30:30,  3.49it/s]
 98%|█████████▊| 365088/371472 [6:53:21<33:14,  3.20it/s]
 98%|█████████▊| 365089/371472 [6:53:22<32:09,  3.31it/s]
 98%|█████████▊| 365090/371472 [6:53:22<30:54,  3.44it/s]
 98%|█████████▊| 365091/371472 [6:53:22<30:49,  3.45it/s]
 98%|█████████▊| 365092/371472 [6:53:22<30:25,  3.49it/s]
 98%|█████████▊| 365093/371472 [6:53:23<29:57,  3.55it/s]
 98%|█████████▊| 365094/371472 [6:53:23<29:38,  3.59it/s]
 98%|█████████▊| 365095/371472 [6:53:23<29:26,  3.61it/s]
 98%|█████████▊| 365096/371472 [6:53:23<30:46,  3.45it/s]
 98%|█████████▊| 365097/371472 [6:53:24<30:25,  3.49it/s]
 98%|█████████▊| 365098/371472 [6:53:24<32:08,  3.30it/s]
 98%|█████████▊| 365099/371472 [6:53:24<31:03,  3.42it/s]
 98%|█████████▊| 365100/371472 [6:53:25<30:47,  3.45it/s]
                                                         
{'loss': 2.7841, 'learning_rate': 1.154463573875757e-07, 'epoch': 15.73}
+
 98%|█████████▊| 365100/371472 [6:53:25<30:47,  3.45it/s]
 98%|█████████▊| 365101/371472 [6:53:25<32:06,  3.31it/s]
 98%|█████████▊| 365102/371472 [6:53:25<32:47,  3.24it/s]
 98%|█████████▊| 365103/371472 [6:53:26<33:02,  3.21it/s]
 98%|█████████▊| 365104/371472 [6:53:26<31:41,  3.35it/s]
 98%|█████████▊| 365105/371472 [6:53:26<31:11,  3.40it/s]
 98%|█████████▊| 365106/371472 [6:53:26<30:26,  3.49it/s]
 98%|█████████▊| 365107/371472 [6:53:27<32:23,  3.27it/s]
 98%|█████████▊| 365108/371472 [6:53:27<31:05,  3.41it/s]
 98%|█████████▊| 365109/371472 [6:53:27<30:52,  3.43it/s]
 98%|█████████▊| 365110/371472 [6:53:28<30:44,  3.45it/s]
 98%|█████████▊| 365111/371472 [6:53:28<31:20,  3.38it/s]
 98%|█████████▊| 365112/371472 [6:53:28<30:13,  3.51it/s]
 98%|█████████▊| 365113/371472 [6:53:29<32:03,  3.31it/s]
 98%|█████████▊| 365114/371472 [6:53:29<30:21,  3.49it/s]
 98%|█████████▊| 365115/371472 [6:53:29<31:32,  3.36it/s]
 98%|█████████▊| 365116/371472 [6:53:29<31:28,  3.37it/s]
 98%|█████████▊| 365117/371472 [6:53:30<31:25,  3.37it/s]
 98%|█████████▊| 365118/371472 [6:53:30<29:49,  3.55it/s]
 98%|█████████▊| 365119/371472 [6:53:30<31:22,  3.37it/s]
 98%|█████████▊| 365120/371472 [6:53:31<30:05,  3.52it/s]
                                                         
{'loss': 2.6285, 'learning_rate': 1.1539787541209676e-07, 'epoch': 15.73}
+
 98%|█████████▊| 365120/371472 [6:53:31<30:05,  3.52it/s]
 98%|█████████▊| 365121/371472 [6:53:31<29:31,  3.58it/s]
 98%|█████████▊| 365122/371472 [6:53:31<28:34,  3.70it/s]
 98%|█████████▊| 365123/371472 [6:53:31<28:30,  3.71it/s]
 98%|█████████▊| 365124/371472 [6:53:32<28:09,  3.76it/s]
 98%|█████████▊| 365125/371472 [6:53:32<27:58,  3.78it/s]
 98%|█████████▊| 365126/371472 [6:53:32<27:18,  3.87it/s]
 98%|█████████▊| 365127/371472 [6:53:32<28:36,  3.70it/s]
 98%|█████████▊| 365128/371472 [6:53:33<29:58,  3.53it/s]
 98%|█████████▊| 365129/371472 [6:53:33<29:15,  3.61it/s]
 98%|█████████▊| 365130/371472 [6:53:33<29:43,  3.55it/s]
 98%|█████████▊| 365131/371472 [6:53:34<35:45,  2.96it/s]
 98%|█████████▊| 365132/371472 [6:53:34<33:55,  3.11it/s]
 98%|█████████▊| 365133/371472 [6:53:34<32:53,  3.21it/s]
 98%|█████████▊| 365134/371472 [6:53:35<31:08,  3.39it/s]
 98%|█████████▊| 365135/371472 [6:53:35<31:28,  3.36it/s]
 98%|█████████▊| 365136/371472 [6:53:35<30:30,  3.46it/s]
 98%|█████████▊| 365137/371472 [6:53:35<29:47,  3.54it/s]
 98%|█████████▊| 365138/371472 [6:53:36<28:50,  3.66it/s]
 98%|█████████▊| 365139/371472 [6:53:36<28:34,  3.69it/s]
 98%|█████████▊| 365140/371472 [6:53:36<29:33,  3.57it/s]
                                                         
{'loss': 2.5816, 'learning_rate': 1.1534939343661791e-07, 'epoch': 15.73}
+
 98%|█████████▊| 365140/371472 [6:53:36<29:33,  3.57it/s]
 98%|█████████▊| 365141/371472 [6:53:37<38:21,  2.75it/s]
 98%|█████████▊| 365142/371472 [6:53:37<35:06,  3.00it/s]
 98%|█████████▊| 365143/371472 [6:53:37<33:20,  3.16it/s]
 98%|█████████▊| 365144/371472 [6:53:38<31:53,  3.31it/s]
 98%|█████████▊| 365145/371472 [6:53:38<31:45,  3.32it/s]
 98%|█████████▊| 365146/371472 [6:53:38<32:28,  3.25it/s]
 98%|█████████▊| 365147/371472 [6:53:38<31:48,  3.31it/s]
 98%|█████████▊| 365148/371472 [6:53:39<31:26,  3.35it/s]
 98%|█████████▊| 365149/371472 [6:53:39<31:22,  3.36it/s]
 98%|█████████▊| 365150/371472 [6:53:39<31:49,  3.31it/s]
 98%|█████████▊| 365151/371472 [6:53:40<30:53,  3.41it/s]
 98%|█████████▊| 365152/371472 [6:53:40<31:18,  3.36it/s]
 98%|█████████▊| 365153/371472 [6:53:40<30:32,  3.45it/s]
 98%|█████████▊| 365154/371472 [6:53:41<29:54,  3.52it/s]
 98%|█████████▊| 365155/371472 [6:53:41<30:17,  3.48it/s]
 98%|█████████▊| 365156/371472 [6:53:41<29:58,  3.51it/s]
 98%|█████████▊| 365157/371472 [6:53:41<30:50,  3.41it/s]
 98%|█████████▊| 365158/371472 [6:53:42<30:10,  3.49it/s]
 98%|█████████▊| 365159/371472 [6:53:42<31:37,  3.33it/s]
 98%|█████████▊| 365160/371472 [6:53:42<30:25,  3.46it/s]
                                                         
{'loss': 2.4902, 'learning_rate': 1.1530091146113898e-07, 'epoch': 15.73}
+
 98%|█████████▊| 365160/371472 [6:53:42<30:25,  3.46it/s]
 98%|█████████▊| 365161/371472 [6:53:43<33:08,  3.17it/s]
 98%|█████████▊| 365162/371472 [6:53:43<32:03,  3.28it/s]
 98%|█████████▊| 365163/371472 [6:53:43<31:02,  3.39it/s]
 98%|█████████▊| 365164/371472 [6:53:43<30:00,  3.50it/s]
 98%|█████████▊| 365165/371472 [6:53:44<28:34,  3.68it/s]
 98%|█████████▊| 365166/371472 [6:53:44<29:01,  3.62it/s]
 98%|█████████▊| 365167/371472 [6:53:44<28:24,  3.70it/s]
 98%|█████████▊| 365168/371472 [6:53:45<29:15,  3.59it/s]
 98%|█████████▊| 365169/371472 [6:53:45<31:00,  3.39it/s]
 98%|█████████▊| 365170/371472 [6:53:45<29:17,  3.59it/s]
 98%|█████████▊| 365171/371472 [6:53:45<29:44,  3.53it/s]
 98%|█████████▊| 365172/371472 [6:53:46<29:10,  3.60it/s]
 98%|█████████▊| 365173/371472 [6:53:46<29:00,  3.62it/s]
 98%|█████████▊| 365174/371472 [6:53:46<29:15,  3.59it/s]
 98%|█████████▊| 365175/371472 [6:53:46<28:35,  3.67it/s]
 98%|█████████▊| 365176/371472 [6:53:47<27:58,  3.75it/s]
 98%|█████████▊| 365177/371472 [6:53:47<27:51,  3.77it/s]
 98%|█████████▊| 365178/371472 [6:53:47<28:37,  3.66it/s]
 98%|█████████▊| 365179/371472 [6:53:48<29:55,  3.51it/s]
 98%|█████████▊| 365180/371472 [6:53:48<30:30,  3.44it/s]
                                                         
{'loss': 2.5158, 'learning_rate': 1.1525242948566013e-07, 'epoch': 15.73}
+
 98%|█████████▊| 365180/371472 [6:53:48<30:30,  3.44it/s]
 98%|█████████▊| 365181/371472 [6:53:48<30:52,  3.40it/s]
 98%|█████████▊| 365182/371472 [6:53:49<31:34,  3.32it/s]
 98%|█████████▊| 365183/371472 [6:53:49<32:36,  3.21it/s]
 98%|█████████▊| 365184/371472 [6:53:49<30:46,  3.41it/s]
 98%|█████████▊| 365185/371472 [6:53:49<32:24,  3.23it/s]
 98%|█████████▊| 365186/371472 [6:53:50<31:05,  3.37it/s]
 98%|█████████▊| 365187/371472 [6:53:50<31:28,  3.33it/s]
 98%|█████████▊| 365188/371472 [6:53:50<31:27,  3.33it/s]
 98%|█████████▊| 365189/371472 [6:53:51<31:15,  3.35it/s]
 98%|█████████▊| 365190/371472 [6:53:51<30:41,  3.41it/s]
 98%|█████████▊| 365191/371472 [6:53:51<31:36,  3.31it/s]
 98%|█████████▊| 365192/371472 [6:53:52<31:53,  3.28it/s]
 98%|█████████▊| 365193/371472 [6:53:52<30:03,  3.48it/s]
 98%|█████████▊| 365194/371472 [6:53:52<29:30,  3.55it/s]
 98%|█████████▊| 365195/371472 [6:53:52<28:28,  3.67it/s]
 98%|█████████▊| 365196/371472 [6:53:53<29:01,  3.60it/s]
 98%|█████████▊| 365197/371472 [6:53:53<29:09,  3.59it/s]
 98%|█████████▊| 365198/371472 [6:53:53<29:15,  3.57it/s]
 98%|█████████▊| 365199/371472 [6:53:53<30:36,  3.42it/s]
 98%|█████████▊| 365200/371472 [6:53:54<29:54,  3.50it/s]
                                                         
{'loss': 2.5828, 'learning_rate': 1.1520394751018119e-07, 'epoch': 15.73}
+
 98%|█████████▊| 365200/371472 [6:53:54<29:54,  3.50it/s]
 98%|█████████▊| 365201/371472 [6:53:54<29:25,  3.55it/s]
 98%|█████████▊| 365202/371472 [6:53:54<29:15,  3.57it/s]
 98%|█████████▊| 365203/371472 [6:53:55<28:21,  3.68it/s]
 98%|█████████▊| 365204/371472 [6:53:55<32:02,  3.26it/s]
 98%|█████████▊| 365205/371472 [6:53:55<32:37,  3.20it/s]
 98%|█████████▊| 365206/371472 [6:53:56<30:54,  3.38it/s]
 98%|█████████▊| 365207/371472 [6:53:56<30:29,  3.42it/s]
 98%|█████████▊| 365208/371472 [6:53:56<29:33,  3.53it/s]
 98%|█████████▊| 365209/371472 [6:53:56<29:11,  3.58it/s]
 98%|█████████▊| 365210/371472 [6:53:57<29:09,  3.58it/s]
 98%|█████████▊| 365211/371472 [6:53:57<29:00,  3.60it/s]
 98%|█████████▊| 365212/371472 [6:53:57<30:41,  3.40it/s]
 98%|█████████▊| 365213/371472 [6:53:58<30:14,  3.45it/s]
 98%|█████████▊| 365214/371472 [6:53:58<32:00,  3.26it/s]
 98%|█████████▊| 365215/371472 [6:53:58<32:12,  3.24it/s]
 98%|█████████▊| 365216/371472 [6:53:58<30:42,  3.40it/s]
 98%|█████████▊| 365217/371472 [6:53:59<29:25,  3.54it/s]
 98%|█████████▊| 365218/371472 [6:53:59<28:17,  3.68it/s]
 98%|█████████▊| 365219/371472 [6:53:59<30:28,  3.42it/s]
 98%|█████████▊| 365220/371472 [6:54:00<30:45,  3.39it/s]
                                                         
{'loss': 2.6135, 'learning_rate': 1.1515546553470235e-07, 'epoch': 15.73}
+
 98%|█████████▊| 365220/371472 [6:54:00<30:45,  3.39it/s]
 98%|█████████▊| 365221/371472 [6:54:00<30:32,  3.41it/s]
 98%|█████████▊| 365222/371472 [6:54:00<29:16,  3.56it/s]
 98%|█████████▊| 365223/371472 [6:54:00<29:07,  3.58it/s]
 98%|█████████▊| 365224/371472 [6:54:01<30:38,  3.40it/s]
 98%|█████████▊| 365225/371472 [6:54:01<30:38,  3.40it/s]
 98%|█████████▊| 365226/371472 [6:54:01<30:26,  3.42it/s]
 98%|█████████▊| 365227/371472 [6:54:02<30:11,  3.45it/s]
 98%|█████████▊| 365228/371472 [6:54:02<28:41,  3.63it/s]
 98%|█████████▊| 365229/371472 [6:54:02<28:45,  3.62it/s]
 98%|█████████▊| 365230/371472 [6:54:02<29:59,  3.47it/s]
 98%|█████████▊| 365231/371472 [6:54:03<29:11,  3.56it/s]
 98%|█████████▊| 365232/371472 [6:54:03<29:00,  3.59it/s]
 98%|█████████▊| 365233/371472 [6:54:03<29:32,  3.52it/s]
 98%|█████████▊| 365234/371472 [6:54:04<29:14,  3.56it/s]
 98%|█████████▊| 365235/371472 [6:54:04<30:51,  3.37it/s]
 98%|█████████▊| 365236/371472 [6:54:04<29:24,  3.53it/s]
 98%|█████████▊| 365237/371472 [6:54:04<28:42,  3.62it/s]
 98%|█████████▊| 365238/371472 [6:54:05<28:10,  3.69it/s]
 98%|█████████▊| 365239/371472 [6:54:05<27:59,  3.71it/s]
 98%|█████████▊| 365240/371472 [6:54:05<29:05,  3.57it/s]
                                                         
{'loss': 2.6731, 'learning_rate': 1.151069835592234e-07, 'epoch': 15.73}
+
 98%|█████████▊| 365240/371472 [6:54:05<29:05,  3.57it/s]
 98%|█████████▊| 365241/371472 [6:54:05<28:00,  3.71it/s]
 98%|█████████▊| 365242/371472 [6:54:06<27:23,  3.79it/s]
 98%|█████████▊| 365243/371472 [6:54:06<26:35,  3.91it/s]
 98%|█████████▊| 365244/371472 [6:54:06<27:44,  3.74it/s]
 98%|█████████▊| 365245/371472 [6:54:06<27:14,  3.81it/s]
 98%|█████████▊| 365246/371472 [6:54:07<27:33,  3.76it/s]
 98%|█████████▊| 365247/371472 [6:54:07<29:41,  3.49it/s]
 98%|█████████▊| 365248/371472 [6:54:07<29:05,  3.57it/s]
 98%|█████████▊| 365249/371472 [6:54:08<29:33,  3.51it/s]
 98%|█████████▊| 365250/371472 [6:54:08<30:08,  3.44it/s]
 98%|█████████▊| 365251/371472 [6:54:08<29:22,  3.53it/s]
 98%|█████████▊| 365252/371472 [6:54:08<28:18,  3.66it/s]
 98%|█████████▊| 365253/371472 [6:54:09<28:27,  3.64it/s]
 98%|█████████▊| 365254/371472 [6:54:09<29:17,  3.54it/s]
 98%|█████████▊| 365255/371472 [6:54:09<28:51,  3.59it/s]
 98%|█████████▊| 365256/371472 [6:54:10<29:55,  3.46it/s]
 98%|█████████▊| 365257/371472 [6:54:10<29:38,  3.49it/s]
 98%|█████████▊| 365258/371472 [6:54:10<28:37,  3.62it/s]
 98%|█████████▊| 365259/371472 [6:54:10<29:05,  3.56it/s]
 98%|█████████▊| 365260/371472 [6:54:11<28:49,  3.59it/s]
                                                         
{'loss': 2.5777, 'learning_rate': 1.1505850158374456e-07, 'epoch': 15.73}
+
 98%|█████████▊| 365260/371472 [6:54:11<28:49,  3.59it/s]
 98%|█████████▊| 365261/371472 [6:54:11<28:24,  3.64it/s]
 98%|█████████▊| 365262/371472 [6:54:11<27:41,  3.74it/s]
 98%|█████████▊| 365263/371472 [6:54:12<27:47,  3.72it/s]
 98%|█████████▊| 365264/371472 [6:54:12<29:01,  3.57it/s]
 98%|█████████▊| 365265/371472 [6:54:12<28:35,  3.62it/s]
 98%|█████████▊| 365266/371472 [6:54:12<27:26,  3.77it/s]
 98%|█████████▊| 365267/371472 [6:54:13<29:11,  3.54it/s]
 98%|█████████▊| 365268/371472 [6:54:13<28:23,  3.64it/s]
 98%|█████████▊| 365269/371472 [6:54:13<28:32,  3.62it/s]
 98%|█████████▊| 365270/371472 [6:54:13<28:58,  3.57it/s]
 98%|█████████▊| 365271/371472 [6:54:14<28:25,  3.64it/s]
 98%|█████████▊| 365272/371472 [6:54:14<28:18,  3.65it/s]
 98%|█████████▊| 365273/371472 [6:54:14<28:03,  3.68it/s]
 98%|█████████▊| 365274/371472 [6:54:15<28:08,  3.67it/s]
 98%|█████████▊| 365275/371472 [6:54:15<28:25,  3.63it/s]
 98%|█████████▊| 365276/371472 [6:54:15<28:10,  3.66it/s]
 98%|█████████▊| 365277/371472 [6:54:15<27:52,  3.70it/s]
 98%|█████████▊| 365278/371472 [6:54:16<27:21,  3.77it/s]
 98%|█████████▊| 365279/371472 [6:54:16<31:07,  3.32it/s]
 98%|██���██████▊| 365280/371472 [6:54:16<30:35,  3.37it/s]
                                                         
{'loss': 2.5594, 'learning_rate': 1.1501001960826562e-07, 'epoch': 15.73}
+
 98%|█████████▊| 365280/371472 [6:54:16<30:35,  3.37it/s]
 98%|█████████▊| 365281/371472 [6:54:17<30:46,  3.35it/s]
 98%|█████████▊| 365282/371472 [6:54:17<28:51,  3.57it/s]
 98%|█████████▊| 365283/371472 [6:54:17<27:37,  3.73it/s]
 98%|█████████▊| 365284/371472 [6:54:17<27:02,  3.81it/s]
 98%|█████████▊| 365285/371472 [6:54:18<26:33,  3.88it/s]
 98%|█████████▊| 365286/371472 [6:54:18<27:42,  3.72it/s]
 98%|█████████▊| 365287/371472 [6:54:18<27:53,  3.70it/s]
 98%|█████████▊| 365288/371472 [6:54:18<28:44,  3.59it/s]
 98%|█████████▊| 365289/371472 [6:54:19<27:57,  3.69it/s]
 98%|█████████▊| 365290/371472 [6:54:19<27:23,  3.76it/s]
 98%|█████████▊| 365291/371472 [6:54:19<29:34,  3.48it/s]
 98%|█████████▊| 365292/371472 [6:54:20<29:34,  3.48it/s]
 98%|█████████▊| 365293/371472 [6:54:20<30:09,  3.41it/s]
 98%|█████████▊| 365294/371472 [6:54:20<28:14,  3.65it/s]
 98%|█████████▊| 365295/371472 [6:54:20<28:24,  3.62it/s]
 98%|█████████▊| 365296/371472 [6:54:21<28:07,  3.66it/s]
 98%|█████████▊| 365297/371472 [6:54:21<27:13,  3.78it/s]
 98%|█████████▊| 365298/371472 [6:54:21<27:27,  3.75it/s]
 98%|█████████▊| 365299/371472 [6:54:21<27:10,  3.79it/s]
 98%|█████████▊| 365300/371472 [6:54:22<28:28,  3.61it/s]
                                                         
{'loss': 2.6255, 'learning_rate': 1.1496153763278676e-07, 'epoch': 15.73}
+
 98%|█████████▊| 365300/371472 [6:54:22<28:28,  3.61it/s]
 98%|█████████▊| 365301/371472 [6:54:22<27:33,  3.73it/s]
 98%|█████████▊| 365302/371472 [6:54:22<27:40,  3.72it/s]
 98%|█████████▊| 365303/371472 [6:54:23<27:23,  3.75it/s]
 98%|█████████▊| 365304/371472 [6:54:23<30:06,  3.42it/s]
 98%|█████████▊| 365305/371472 [6:54:23<28:37,  3.59it/s]
 98%|█████████▊| 365306/371472 [6:54:23<30:58,  3.32it/s]
 98%|█████████▊| 365307/371472 [6:54:24<29:11,  3.52it/s]
 98%|█████████▊| 365308/371472 [6:54:24<28:11,  3.64it/s]
 98%|█████████▊| 365309/371472 [6:54:24<29:01,  3.54it/s]
 98%|█████████▊| 365310/371472 [6:54:25<28:22,  3.62it/s]
 98%|█████████▊| 365311/371472 [6:54:25<28:54,  3.55it/s]
 98%|█████████▊| 365312/371472 [6:54:25<28:29,  3.60it/s]
 98%|█████████▊| 365313/371472 [6:54:25<28:52,  3.55it/s]
 98%|█████████▊| 365314/371472 [6:54:26<29:13,  3.51it/s]
 98%|█████████▊| 365315/371472 [6:54:26<28:28,  3.60it/s]
 98%|█████████▊| 365316/371472 [6:54:26<29:10,  3.52it/s]
 98%|█████████▊| 365317/371472 [6:54:27<29:03,  3.53it/s]
 98%|█████████▊| 365318/371472 [6:54:27<28:40,  3.58it/s]
 98%|█████████▊| 365319/371472 [6:54:27<27:57,  3.67it/s]
 98%|█████████▊| 365320/371472 [6:54:27<28:34,  3.59it/s]
                                                         
{'loss': 2.5444, 'learning_rate': 1.1491305565730783e-07, 'epoch': 15.74}
+
 98%|█████████▊| 365320/371472 [6:54:27<28:34,  3.59it/s]
 98%|█████████▊| 365321/371472 [6:54:28<31:00,  3.31it/s]
 98%|█████████▊| 365322/371472 [6:54:28<30:20,  3.38it/s]
 98%|█████████▊| 365323/371472 [6:54:28<29:34,  3.47it/s]
 98%|█████████▊| 365324/371472 [6:54:29<28:54,  3.54it/s]
 98%|█████████▊| 365325/371472 [6:54:29<28:57,  3.54it/s]
 98%|█████████▊| 365326/371472 [6:54:29<29:02,  3.53it/s]
 98%|█████████▊| 365327/371472 [6:54:29<29:42,  3.45it/s]
 98%|█████████▊| 365328/371472 [6:54:30<31:18,  3.27it/s]
 98%|█████████▊| 365329/371472 [6:54:30<29:32,  3.47it/s]
 98%|█████████▊| 365330/371472 [6:54:30<28:54,  3.54it/s]
 98%|█████████▊| 365331/371472 [6:54:31<28:46,  3.56it/s]
 98%|█████████▊| 365332/371472 [6:54:31<28:27,  3.60it/s]
 98%|█████████▊| 365333/371472 [6:54:31<28:30,  3.59it/s]
 98%|█████████▊| 365334/371472 [6:54:31<27:56,  3.66it/s]
 98%|█████████▊| 365335/371472 [6:54:32<27:17,  3.75it/s]
 98%|█████████▊| 365336/371472 [6:54:32<27:24,  3.73it/s]
 98%|█████████▊| 365337/371472 [6:54:32<26:58,  3.79it/s]
 98%|█████████▊| 365338/371472 [6:54:32<26:39,  3.84it/s]
 98%|█████████▊| 365339/371472 [6:54:33<31:04,  3.29it/s]
 98%|█████████▊| 365340/371472 [6:54:33<30:39,  3.33it/s]
                                                         
{'loss': 2.4617, 'learning_rate': 1.14864573681829e-07, 'epoch': 15.74}
+
 98%|█████████▊| 365340/371472 [6:54:33<30:39,  3.33it/s]
 98%|█████████▊| 365341/371472 [6:54:33<29:27,  3.47it/s]
 98%|█████████▊| 365342/371472 [6:54:34<28:31,  3.58it/s]
 98%|█████████▊| 365343/371472 [6:54:34<27:54,  3.66it/s]
 98%|█████████▊| 365344/371472 [6:54:34<27:21,  3.73it/s]
 98%|█████████▊| 365345/371472 [6:54:34<27:30,  3.71it/s]
 98%|█████████▊| 365346/371472 [6:54:35<28:31,  3.58it/s]
 98%|█████████▊| 365347/371472 [6:54:35<27:34,  3.70it/s]
 98%|█████████▊| 365348/371472 [6:54:35<28:11,  3.62it/s]
 98%|█████████▊| 365349/371472 [6:54:35<27:21,  3.73it/s]
 98%|█████████▊| 365350/371472 [6:54:36<27:44,  3.68it/s]
 98%|█████████▊| 365351/371472 [6:54:36<28:26,  3.59it/s]
 98%|█████████▊| 365352/371472 [6:54:36<29:24,  3.47it/s]
 98%|█████████▊| 365353/371472 [6:54:37<31:16,  3.26it/s]
 98%|█████████▊| 365354/371472 [6:54:37<29:42,  3.43it/s]
 98%|█████████▊| 365355/371472 [6:54:37<28:33,  3.57it/s]
 98%|█████████▊| 365356/371472 [6:54:37<27:49,  3.66it/s]
 98%|█████████▊| 365357/371472 [6:54:38<27:42,  3.68it/s]
 98%|█████████▊| 365358/371472 [6:54:38<27:36,  3.69it/s]
 98%|█████████▊| 365359/371472 [6:54:38<30:30,  3.34it/s]
 98%|█████████▊| 365360/371472 [6:54:39<28:42,  3.55it/s]
                                                         
{'loss': 2.6617, 'learning_rate': 1.1481609170635005e-07, 'epoch': 15.74}
+
 98%|█████████▊| 365360/371472 [6:54:39<28:42,  3.55it/s]
 98%|█████████▊| 365361/371472 [6:54:39<29:34,  3.44it/s]
 98%|█████████▊| 365362/371472 [6:54:39<28:20,  3.59it/s]
 98%|█████████▊| 365363/371472 [6:54:39<27:44,  3.67it/s]
 98%|█████████▊| 365364/371472 [6:54:40<27:28,  3.71it/s]
 98%|█████████▊| 365365/371472 [6:54:40<27:37,  3.68it/s]
 98%|█████████▊| 365366/371472 [6:54:40<28:39,  3.55it/s]
 98%|█████████▊| 365367/371472 [6:54:41<27:23,  3.71it/s]
 98%|█████████▊| 365368/371472 [6:54:41<26:40,  3.81it/s]
 98%|█████████▊| 365369/371472 [6:54:41<26:13,  3.88it/s]
 98%|█████████▊| 365370/371472 [6:54:41<26:04,  3.90it/s]
 98%|█████████▊| 365371/371472 [6:54:42<28:31,  3.57it/s]
 98%|█████████▊| 365372/371472 [6:54:42<28:56,  3.51it/s]
 98%|█████████▊| 365373/371472 [6:54:42<29:17,  3.47it/s]
 98%|█████████▊| 365374/371472 [6:54:42<28:13,  3.60it/s]
 98%|█████████▊| 365375/371472 [6:54:43<28:00,  3.63it/s]
 98%|█████████▊| 365376/371472 [6:54:43<27:09,  3.74it/s]
 98%|█████████▊| 365377/371472 [6:54:43<27:25,  3.70it/s]
 98%|█████████▊| 365378/371472 [6:54:44<27:37,  3.68it/s]
 98%|█████████▊| 365379/371472 [6:54:44<27:47,  3.65it/s]
 98%|█████████▊| 365380/371472 [6:54:44<27:07,  3.74it/s]
                                                         
{'loss': 2.5663, 'learning_rate': 1.147676097308712e-07, 'epoch': 15.74}
+
 98%|█████████▊| 365380/371472 [6:54:44<27:07,  3.74it/s]
 98%|█████████▊| 365381/371472 [6:54:44<26:34,  3.82it/s]
 98%|█████████▊| 365382/371472 [6:54:45<27:18,  3.72it/s]
 98%|█████████▊| 365383/371472 [6:54:45<29:21,  3.46it/s]
 98%|█████████▊| 365384/371472 [6:54:45<28:03,  3.62it/s]
 98%|█████████▊| 365385/371472 [6:54:45<28:23,  3.57it/s]
 98%|█████████▊| 365386/371472 [6:54:46<27:33,  3.68it/s]
 98%|█████████▊| 365387/371472 [6:54:46<27:56,  3.63it/s]
 98%|█████████▊| 365388/371472 [6:54:46<30:49,  3.29it/s]
 98%|█████████▊| 365389/371472 [6:54:47<30:02,  3.38it/s]
 98%|█████████▊| 365390/371472 [6:54:47<29:30,  3.44it/s]
 98%|█████████▊| 365391/371472 [6:54:47<28:36,  3.54it/s]
 98%|█████████▊| 365392/371472 [6:54:47<27:30,  3.68it/s]
 98%|█████████▊| 365393/371472 [6:54:48<26:51,  3.77it/s]
 98%|█████████▊| 365394/371472 [6:54:48<26:52,  3.77it/s]
 98%|██��██████▊| 365395/371472 [6:54:48<26:44,  3.79it/s]
 98%|█████████▊| 365396/371472 [6:54:48<26:49,  3.78it/s]
 98%|█████████▊| 365397/371472 [6:54:49<28:29,  3.55it/s]
 98%|█████████▊| 365398/371472 [6:54:49<29:41,  3.41it/s]
 98%|█████████▊| 365399/371472 [6:54:50<35:10,  2.88it/s]
 98%|█████████▊| 365400/371472 [6:54:50<35:42,  2.83it/s]
                                                         
{'loss': 2.527, 'learning_rate': 1.1471912775539226e-07, 'epoch': 15.74}
+
 98%|█████████▊| 365400/371472 [6:54:50<35:42,  2.83it/s]
 98%|█████████▊| 365401/371472 [6:54:50<33:02,  3.06it/s]
 98%|█████████▊| 365402/371472 [6:54:50<30:47,  3.29it/s]
 98%|█████████▊| 365403/371472 [6:54:51<29:15,  3.46it/s]
 98%|█████████▊| 365404/371472 [6:54:51<28:40,  3.53it/s]
 98%|█████████▊| 365405/371472 [6:54:51<27:35,  3.66it/s]
 98%|█████████▊| 365406/371472 [6:54:52<27:48,  3.64it/s]
 98%|█████████▊| 365407/371472 [6:54:52<27:46,  3.64it/s]
 98%|█████████▊| 365408/371472 [6:54:52<29:17,  3.45it/s]
 98%|█████████▊| 365409/371472 [6:54:52<31:23,  3.22it/s]
 98%|█████████▊| 365410/371472 [6:54:53<28:58,  3.49it/s]
 98%|█████████▊| 365411/371472 [6:54:53<28:20,  3.56it/s]
 98%|█████████▊| 365412/371472 [6:54:53<26:42,  3.78it/s]
 98%|█████████▊| 365413/371472 [6:54:54<28:54,  3.49it/s]
 98%|█████████▊| 365414/371472 [6:54:54<27:17,  3.70it/s]
 98%|█████████▊| 365415/371472 [6:54:54<28:00,  3.60it/s]
 98%|█████████▊| 365416/371472 [6:54:54<30:57,  3.26it/s]
 98%|█████████▊| 365417/371472 [6:54:55<29:38,  3.40it/s]
 98%|█████████▊| 365418/371472 [6:54:55<30:10,  3.34it/s]
 98%|█████████▊| 365419/371472 [6:54:55<28:54,  3.49it/s]
 98%|█████████▊| 365420/371472 [6:54:56<27:35,  3.66it/s]
                                                         
{'loss': 2.7046, 'learning_rate': 1.1467064577991343e-07, 'epoch': 15.74}
+
 98%|█████████▊| 365420/371472 [6:54:56<27:35,  3.66it/s]
 98%|█████████▊| 365421/371472 [6:54:56<28:43,  3.51it/s]
 98%|█████████▊| 365422/371472 [6:54:56<29:33,  3.41it/s]
 98%|█████████▊| 365423/371472 [6:54:56<29:26,  3.42it/s]
 98%|█████████▊| 365424/371472 [6:54:57<28:44,  3.51it/s]
 98%|█████████▊| 365425/371472 [6:54:57<29:38,  3.40it/s]
 98%|█████████▊| 365426/371472 [6:54:57<29:33,  3.41it/s]
 98%|█████████▊| 365427/371472 [6:54:58<29:46,  3.38it/s]
 98%|█████████▊| 365428/371472 [6:54:58<28:03,  3.59it/s]
 98%|█████████▊| 365429/371472 [6:54:58<27:56,  3.60it/s]
 98%|█████████▊| 365430/371472 [6:54:58<26:42,  3.77it/s]
 98%|█████████▊| 365431/371472 [6:54:59<26:36,  3.78it/s]
 98%|█████████▊| 365432/371472 [6:54:59<26:52,  3.75it/s]
 98%|█████████▊| 365433/371472 [6:54:59<26:23,  3.81it/s]
 98%|█████████▊| 365434/371472 [6:54:59<28:09,  3.57it/s]
 98%|█████████▊| 365435/371472 [6:55:00<28:04,  3.58it/s]
 98%|█████████▊| 365436/371472 [6:55:00<28:02,  3.59it/s]
 98%|█████████▊| 365437/371472 [6:55:00<28:59,  3.47it/s]
 98%|█████████▊| 365438/371472 [6:55:01<30:02,  3.35it/s]
 98%|█████████▊| 365439/371472 [6:55:01<29:31,  3.41it/s]
 98%|█████████▊| 365440/371472 [6:55:01<29:53,  3.36it/s]
                                                         
{'loss': 2.535, 'learning_rate': 1.1462216380443448e-07, 'epoch': 15.74}
+
 98%|█████████▊| 365440/371472 [6:55:01<29:53,  3.36it/s]
 98%|█████████▊| 365441/371472 [6:55:02<30:37,  3.28it/s]
 98%|█████████▊| 365442/371472 [6:55:02<31:08,  3.23it/s]
 98%|█████████▊| 365443/371472 [6:55:02<29:32,  3.40it/s]
 98%|█████████▊| 365444/371472 [6:55:02<29:32,  3.40it/s]
 98%|█████████▊| 365445/371472 [6:55:03<29:12,  3.44it/s]
 98%|█████████▊| 365446/371472 [6:55:03<27:56,  3.59it/s]
 98%|█████████▊| 365447/371472 [6:55:03<29:14,  3.43it/s]
 98%|█████████▊| 365448/371472 [6:55:04<29:43,  3.38it/s]
 98%|█████████▊| 365449/371472 [6:55:04<33:03,  3.04it/s]
 98%|█████████▊| 365450/371472 [6:55:04<30:34,  3.28it/s]
 98%|█████████▊| 365451/371472 [6:55:04<28:34,  3.51it/s]
 98%|█████████▊| 365452/371472 [6:55:05<27:36,  3.63it/s]
 98%|█████████▊| 365453/371472 [6:55:05<28:51,  3.48it/s]
 98%|█████████▊| 365454/371472 [6:55:05<29:17,  3.42it/s]
 98%|█████████▊| 365455/371472 [6:55:06<28:08,  3.56it/s]
 98%|█████████▊| 365456/371472 [6:55:06<27:53,  3.59it/s]
 98%|█████████▊| 365457/371472 [6:55:06<28:53,  3.47it/s]
 98%|█████████▊| 365458/371472 [6:55:06<27:58,  3.58it/s]
 98%|█████████▊| 365459/371472 [6:55:07<28:23,  3.53it/s]
 98%|█████████▊| 365460/371472 [6:55:07<28:56,  3.46it/s]
                                                         
{'loss': 2.5005, 'learning_rate': 1.1457368182895554e-07, 'epoch': 15.74}
+
 98%|█████████▊| 365460/371472 [6:55:07<28:56,  3.46it/s]
 98%|█████████▊| 365461/371472 [6:55:07<28:51,  3.47it/s]
 98%|█████████▊| 365462/371472 [6:55:08<28:04,  3.57it/s]
 98%|█████████▊| 365463/371472 [6:55:08<27:00,  3.71it/s]
 98%|█████████▊| 365464/371472 [6:55:08<26:59,  3.71it/s]
 98%|█████████▊| 365465/371472 [6:55:08<26:34,  3.77it/s]
 98%|█████████▊| 365466/371472 [6:55:09<25:40,  3.90it/s]
 98%|█████████▊| 365467/371472 [6:55:09<25:25,  3.94it/s]
 98%|█████████▊| 365468/371472 [6:55:09<25:34,  3.91it/s]
 98%|█████████▊| 365469/371472 [6:55:10<30:08,  3.32it/s]
 98%|█████████▊| 365470/371472 [6:55:10<28:39,  3.49it/s]
 98%|█████████▊| 365471/371472 [6:55:10<27:38,  3.62it/s]
 98%|█████████▊| 365472/371472 [6:55:10<30:24,  3.29it/s]
 98%|█████████▊| 365473/371472 [6:55:11<28:58,  3.45it/s]
 98%|█████████▊| 365474/371472 [6:55:11<29:19,  3.41it/s]
 98%|█████████▊| 365475/371472 [6:55:11<28:15,  3.54it/s]
 98%|█████████▊| 365476/371472 [6:55:12<29:09,  3.43it/s]
 98%|█████████▊| 365477/371472 [6:55:12<30:24,  3.29it/s]
 98%|█████████▊| 365478/371472 [6:55:12<30:16,  3.30it/s]
 98%|█████████▊| 365479/371472 [6:55:12<28:59,  3.45it/s]
 98%|█████████▊| 365480/371472 [6:55:13<29:28,  3.39it/s]
                                                         
{'loss': 2.4425, 'learning_rate': 1.1452519985347669e-07, 'epoch': 15.74}
+
 98%|█████████▊| 365480/371472 [6:55:13<29:28,  3.39it/s]
 98%|█████████▊| 365481/371472 [6:55:13<28:05,  3.55it/s]
 98%|█████████▊| 365482/371472 [6:55:13<26:52,  3.72it/s]
 98%|█████████▊| 365483/371472 [6:55:14<27:34,  3.62it/s]
 98%|█████████▊| 365484/371472 [6:55:14<27:41,  3.60it/s]
 98%|█████████▊| 365485/371472 [6:55:14<26:44,  3.73it/s]
 98%|█████████▊| 365486/371472 [6:55:14<27:00,  3.69it/s]
 98%|█████████▊| 365487/371472 [6:55:15<30:13,  3.30it/s]
 98%|█████████▊| 365488/371472 [6:55:15<28:58,  3.44it/s]
 98%|█████████▊| 365489/371472 [6:55:15<28:20,  3.52it/s]
 98%|█████████▊| 365490/371472 [6:55:15<27:29,  3.63it/s]
 98%|█████████▊| 365491/371472 [6:55:16<28:00,  3.56it/s]
 98%|█████████▊| 365492/371472 [6:55:16<29:45,  3.35it/s]
 98%|█████████▊| 365493/371472 [6:55:16<28:35,  3.48it/s]
 98%|█████████▊| 365494/371472 [6:55:17<28:39,  3.48it/s]
 98%|█████████▊| 365495/371472 [6:55:17<26:59,  3.69it/s]
 98%|█████████▊| 365496/371472 [6:55:17<26:59,  3.69it/s]
 98%|█████████▊| 365497/371472 [6:55:17<25:58,  3.83it/s]
 98%|█████████▊| 365498/371472 [6:55:18<26:19,  3.78it/s]
 98%|█████████▊| 365499/371472 [6:55:18<26:15,  3.79it/s]
 98%|█████████▊| 365500/371472 [6:55:18<27:09,  3.66it/s]
                                                         
{'loss': 2.5398, 'learning_rate': 1.1447671787799774e-07, 'epoch': 15.74}
+
 98%|█████████▊| 365500/371472 [6:55:18<27:09,  3.66it/s]
 98%|█████████▊| 365501/371472 [6:55:18<27:01,  3.68it/s]
 98%|█████████▊| 365502/371472 [6:55:19<26:05,  3.81it/s]
 98%|█████████▊| 365503/371472 [6:55:19<25:45,  3.86it/s]
 98%|█████████▊| 365504/371472 [6:55:19<28:30,  3.49it/s]
 98%|█████████▊| 365505/371472 [6:55:20<27:37,  3.60it/s]
 98%|█████████▊| 365506/371472 [6:55:20<27:21,  3.63it/s]
 98%|█████████▊| 365507/371472 [6:55:20<30:09,  3.30it/s]
 98%|█████████▊| 365508/371472 [6:55:21<32:26,  3.06it/s]
 98%|█████████▊| 365509/371472 [6:55:21<31:08,  3.19it/s]
 98%|██��██████▊| 365510/371472 [6:55:21<30:05,  3.30it/s]
 98%|█████████▊| 365511/371472 [6:55:21<28:43,  3.46it/s]
 98%|█████████▊| 365512/371472 [6:55:22<29:00,  3.42it/s]
 98%|█████████▊| 365513/371472 [6:55:22<27:57,  3.55it/s]
 98%|█████████▊| 365514/371472 [6:55:22<27:17,  3.64it/s]
 98%|█████████▊| 365515/371472 [6:55:23<27:27,  3.62it/s]
 98%|█████████▊| 365516/371472 [6:55:23<27:38,  3.59it/s]
 98%|█████████▊| 365517/371472 [6:55:23<26:51,  3.69it/s]
 98%|█████████▊| 365518/371472 [6:55:23<27:19,  3.63it/s]
 98%|█████████▊| 365519/371472 [6:55:24<28:44,  3.45it/s]
 98%|█████████▊| 365520/371472 [6:55:24<27:44,  3.58it/s]
                                                         
{'loss': 2.5763, 'learning_rate': 1.1442823590251892e-07, 'epoch': 15.74}
+
 98%|█████████▊| 365520/371472 [6:55:24<27:44,  3.58it/s]
 98%|█████████▊| 365521/371472 [6:55:24<26:47,  3.70it/s]
 98%|█████████▊| 365522/371472 [6:55:24<26:34,  3.73it/s]
 98%|█████████▊| 365523/371472 [6:55:25<25:49,  3.84it/s]
 98%|█████████▊| 365524/371472 [6:55:25<25:49,  3.84it/s]
 98%|█████████▊| 365525/371472 [6:55:25<26:17,  3.77it/s]
 98%|█████████▊| 365526/371472 [6:55:25<25:59,  3.81it/s]
 98%|█████████▊| 365527/371472 [6:55:26<30:00,  3.30it/s]
 98%|█████████▊| 365528/371472 [6:55:26<31:28,  3.15it/s]
 98%|█████████▊| 365529/371472 [6:55:27<31:36,  3.13it/s]
 98%|█████████▊| 365530/371472 [6:55:27<29:24,  3.37it/s]
 98%|█████████▊| 365531/371472 [6:55:27<28:27,  3.48it/s]
 98%|█████████▊| 365532/371472 [6:55:27<27:32,  3.59it/s]
 98%|█████████▊| 365533/371472 [6:55:28<27:35,  3.59it/s]
 98%|█████████▊| 365534/371472 [6:55:28<27:10,  3.64it/s]
 98%|█████████▊| 365535/371472 [6:55:28<27:30,  3.60it/s]
 98%|█████████▊| 365536/371472 [6:55:28<26:43,  3.70it/s]
 98%|█████████▊| 365537/371472 [6:55:29<26:34,  3.72it/s]
 98%|█████████▊| 365538/371472 [6:55:29<26:00,  3.80it/s]
 98%|█████████▊| 365539/371472 [6:55:29<26:23,  3.75it/s]
 98%|█████████▊| 365540/371472 [6:55:29<25:43,  3.84it/s]
                                                         
{'loss': 2.4945, 'learning_rate': 1.1437975392703997e-07, 'epoch': 15.74}
+
 98%|█████████▊| 365540/371472 [6:55:29<25:43,  3.84it/s]
 98%|█████████▊| 365541/371472 [6:55:30<26:45,  3.69it/s]
 98%|█████████▊| 365542/371472 [6:55:30<27:03,  3.65it/s]
 98%|█████████▊| 365543/371472 [6:55:30<27:34,  3.58it/s]
 98%|█████████▊| 365544/371472 [6:55:31<28:30,  3.46it/s]
 98%|█████████▊| 365545/371472 [6:55:31<27:13,  3.63it/s]
 98%|█████████▊| 365546/371472 [6:55:31<26:41,  3.70it/s]
 98%|█████████▊| 365547/371472 [6:55:31<27:51,  3.54it/s]
 98%|█████████▊| 365548/371472 [6:55:32<29:37,  3.33it/s]
 98%|█████████▊| 365549/371472 [6:55:32<30:23,  3.25it/s]
 98%|█████████▊| 365550/371472 [6:55:32<28:29,  3.46it/s]
 98%|█████████▊| 365551/371472 [6:55:33<28:45,  3.43it/s]
 98%|█████████▊| 365552/371472 [6:55:33<29:19,  3.37it/s]
 98%|█████████▊| 365553/371472 [6:55:33<29:59,  3.29it/s]
 98%|█████████▊| 365554/371472 [6:55:34<29:04,  3.39it/s]
 98%|█████████▊| 365555/371472 [6:55:34<28:37,  3.45it/s]
 98%|█████████▊| 365556/371472 [6:55:34<28:01,  3.52it/s]
 98%|█████████▊| 365557/371472 [6:55:34<28:02,  3.51it/s]
 98%|█████████▊| 365558/371472 [6:55:35<27:48,  3.54it/s]
 98%|█████████▊| 365559/371472 [6:55:35<30:14,  3.26it/s]
 98%|█████████▊| 365560/371472 [6:55:35<30:05,  3.27it/s]
                                                         
{'loss': 2.5516, 'learning_rate': 1.1433127195156112e-07, 'epoch': 15.75}
+
 98%|█████████▊| 365560/371472 [6:55:35<30:05,  3.27it/s]
 98%|█████████▊| 365561/371472 [6:55:36<30:00,  3.28it/s]
 98%|█████████▊| 365562/371472 [6:55:36<28:36,  3.44it/s]
 98%|█████████▊| 365563/371472 [6:55:36<27:57,  3.52it/s]
 98%|█████████▊| 365564/371472 [6:55:36<27:45,  3.55it/s]
 98%|█████████▊| 365565/371472 [6:55:37<29:04,  3.39it/s]
 98%|█████████▊| 365566/371472 [6:55:37<29:14,  3.37it/s]
 98%|█████████▊| 365567/371472 [6:55:37<27:35,  3.57it/s]
 98%|█████████▊| 365568/371472 [6:55:38<26:50,  3.67it/s]
 98%|█████████▊| 365569/371472 [6:55:38<27:22,  3.59it/s]
 98%|█████████▊| 365570/371472 [6:55:38<27:01,  3.64it/s]
 98%|█████████▊| 365571/371472 [6:55:38<27:53,  3.53it/s]
 98%|█████████▊| 365572/371472 [6:55:39<27:52,  3.53it/s]
 98%|█████████▊| 365573/371472 [6:55:39<27:34,  3.57it/s]
 98%|█████████▊| 365574/371472 [6:55:39<28:24,  3.46it/s]
 98%|█████████▊| 365575/371472 [6:55:40<27:31,  3.57it/s]
 98%|█████████▊| 365576/371472 [6:55:40<26:41,  3.68it/s]
 98%|█████████▊| 365577/371472 [6:55:40<28:43,  3.42it/s]
 98%|█████████▊| 365578/371472 [6:55:40<27:18,  3.60it/s]
 98%|█████████▊| 365579/371472 [6:55:41<29:12,  3.36it/s]
 98%|█████████▊| 365580/371472 [6:55:41<28:03,  3.50it/s]
                                                         
{'loss': 2.5732, 'learning_rate': 1.1428278997608218e-07, 'epoch': 15.75}
+
 98%|█████████▊| 365580/371472 [6:55:41<28:03,  3.50it/s]
 98%|█████████▊| 365581/371472 [6:55:41<26:46,  3.67it/s]
 98%|█████████▊| 365582/371472 [6:55:41<25:47,  3.81it/s]
 98%|█████████▊| 365583/371472 [6:55:42<26:26,  3.71it/s]
 98%|█████████▊| 365584/371472 [6:55:42<26:59,  3.63it/s]
 98%|█████████▊| 365585/371472 [6:55:42<26:41,  3.67it/s]
 98%|█████████▊| 365586/371472 [6:55:43<27:30,  3.57it/s]
 98%|█████████▊| 365587/371472 [6:55:43<27:50,  3.52it/s]
 98%|█████████▊| 365588/371472 [6:55:43<29:18,  3.35it/s]
 98%|█████████▊| 365589/371472 [6:55:43<28:23,  3.45it/s]
 98%|█████████▊| 365590/371472 [6:55:44<28:09,  3.48it/s]
 98%|█████████▊| 365591/371472 [6:55:44<27:10,  3.61it/s]
 98%|█████████▊| 365592/371472 [6:55:44<26:35,  3.69it/s]
 98%|█████████▊| 365593/371472 [6:55:45<26:27,  3.70it/s]
 98%|█████████▊| 365594/371472 [6:55:45<26:11,  3.74it/s]
 98%|█████████▊| 365595/371472 [6:55:45<27:37,  3.54it/s]
 98%|█████████▊| 365596/371472 [6:55:45<28:16,  3.46it/s]
 98%|█████████▊| 365597/371472 [6:55:46<27:52,  3.51it/s]
 98%|█████████▊| 365598/371472 [6:55:46<27:10,  3.60it/s]
 98%|█████████▊| 365599/371472 [6:55:46<27:51,  3.51it/s]
 98%|█████████▊| 365600/371472 [6:55:47<26:44,  3.66it/s]
                                                         
{'loss': 2.6329, 'learning_rate': 1.1423430800060335e-07, 'epoch': 15.75}
+
 98%|█████████▊| 365600/371472 [6:55:47<26:44,  3.66it/s]
 98%|█████████▊| 365601/371472 [6:55:47<26:27,  3.70it/s]
 98%|█████████▊| 365602/371472 [6:55:47<26:36,  3.68it/s]
 98%|█████████▊| 365603/371472 [6:55:47<28:13,  3.47it/s]
 98%|█████████▊| 365604/371472 [6:55:48<29:48,  3.28it/s]
 98%|█████████▊| 365605/371472 [6:55:48<29:09,  3.35it/s]
 98%|█████████▊| 365606/371472 [6:55:48<27:56,  3.50it/s]
 98%|█████████▊| 365607/371472 [6:55:49<27:35,  3.54it/s]
 98%|█████████▊| 365608/371472 [6:55:49<27:20,  3.57it/s]
 98%|█████████▊| 365609/371472 [6:55:49<28:20,  3.45it/s]
 98%|█████████▊| 365610/371472 [6:55:49<27:01,  3.61it/s]
 98%|█████████▊| 365611/371472 [6:55:50<27:41,  3.53it/s]
 98%|█████████▊| 365612/371472 [6:55:50<27:30,  3.55it/s]
 98%|█████████▊| 365613/371472 [6:55:50<26:55,  3.63it/s]
 98%|█████████▊| 365614/371472 [6:55:50<26:45,  3.65it/s]
 98%|█████████▊| 365615/371472 [6:55:51<28:23,  3.44it/s]
 98%|█████████▊| 365616/371472 [6:55:51<26:55,  3.63it/s]
 98%|█████████▊| 365617/371472 [6:55:51<27:04,  3.60it/s]
 98%|█████████▊| 365618/371472 [6:55:52<25:58,  3.76it/s]
 98%|█████████▊| 365619/371472 [6:55:52<25:22,  3.84it/s]
 98%|█████████▊| 365620/371472 [6:55:52<25:04,  3.89it/s]
                                                         
{'loss': 2.6287, 'learning_rate': 1.1418582602512439e-07, 'epoch': 15.75}
+
 98%|█████████▊| 365620/371472 [6:55:52<25:04,  3.89it/s]
 98%|█████████▊| 365621/371472 [6:55:52<25:47,  3.78it/s]
 98%|█████████▊| 365622/371472 [6:55:53<25:07,  3.88it/s]
 98%|█████████▊| 365623/371472 [6:55:53<25:54,  3.76it/s]
 98%|█████████▊| 365624/371472 [6:55:53<26:35,  3.66it/s]
 98%|█████████▊| 365625/371472 [6:55:53<27:41,  3.52it/s]
 98%|█████████▊| 365626/371472 [6:55:54<27:10,  3.59it/s]
 98%|█████████▊| 365627/371472 [6:55:54<27:23,  3.56it/s]
 98%|█████████▊| 365628/371472 [6:55:54<27:01,  3.60it/s]
 98%|█████████▊| 365629/371472 [6:55:55<27:32,  3.54it/s]
 98%|█████████▊| 365630/371472 [6:55:55<27:16,  3.57it/s]
 98%|█████████▊| 365631/371472 [6:55:55<27:16,  3.57it/s]
 98%|█████████▊| 365632/371472 [6:55:55<27:22,  3.56it/s]
 98%|█████████▊| 365633/371472 [6:55:56<26:27,  3.68it/s]
 98%|█████████▊| 365634/371472 [6:55:56<26:01,  3.74it/s]
 98%|█████████▊| 365635/371472 [6:55:56<27:25,  3.55it/s]
 98%|█████████▊| 365636/371472 [6:55:56<26:14,  3.71it/s]
 98%|█████████▊| 365637/371472 [6:55:57<26:11,  3.71it/s]
 98%|█████████▊| 365638/371472 [6:55:57<25:52,  3.76it/s]
 98%|█████████▊| 365639/371472 [6:55:57<27:39,  3.51it/s]
 98%|█████████▊| 365640/371472 [6:55:58<30:02,  3.24it/s]
                                                         
{'loss': 2.4966, 'learning_rate': 1.1413734404964556e-07, 'epoch': 15.75}
+
 98%|█████████▊| 365640/371472 [6:55:58<30:02,  3.24it/s]
 98%|█████████▊| 365641/371472 [6:55:58<30:09,  3.22it/s]
 98%|█████████▊| 365642/371472 [6:55:58<28:20,  3.43it/s]
 98%|█████████▊| 365643/371472 [6:55:59<27:02,  3.59it/s]
 98%|█████████▊| 365644/371472 [6:55:59<28:42,  3.38it/s]
 98%|█████████▊| 365645/371472 [6:55:59<28:29,  3.41it/s]
 98%|█████████▊| 365646/371472 [6:55:59<28:25,  3.42it/s]
 98%|█████████▊| 365647/371472 [6:56:00<28:04,  3.46it/s]
 98%|█████████▊| 365648/371472 [6:56:00<28:33,  3.40it/s]
 98%|█████████▊| 365649/371472 [6:56:00<28:55,  3.36it/s]
 98%|█████████▊| 365650/371472 [6:56:01<28:18,  3.43it/s]
 98%|█████████▊| 365651/371472 [6:56:01<27:15,  3.56it/s]
 98%|█████████▊| 365652/371472 [6:56:01<26:22,  3.68it/s]
 98%|█████████▊| 365653/371472 [6:56:01<25:27,  3.81it/s]
 98%|█████████▊| 365654/371472 [6:56:02<26:16,  3.69it/s]
 98%|█████████▊| 365655/371472 [6:56:02<28:56,  3.35it/s]
 98%|█████████▊| 365656/371472 [6:56:02<28:20,  3.42it/s]
 98%|█████████▊| 365657/371472 [6:56:03<27:53,  3.47it/s]
 98%|█████████▊| 365658/371472 [6:56:03<28:28,  3.40it/s]
 98%|█████████▊| 365659/371472 [6:56:03<27:35,  3.51it/s]
 98%|█████████▊| 365660/371472 [6:56:03<28:07,  3.44it/s]
                                                         
{'loss': 2.7287, 'learning_rate': 1.1408886207416661e-07, 'epoch': 15.75}
+
 98%|█████████▊| 365660/371472 [6:56:03<28:07,  3.44it/s]
 98%|█████████▊| 365661/371472 [6:56:04<28:51,  3.36it/s]
 98%|█████████▊| 365662/371472 [6:56:04<28:31,  3.39it/s]
 98%|█████████▊| 365663/371472 [6:56:04<26:44,  3.62it/s]
 98%|█████████▊| 365664/371472 [6:56:05<26:53,  3.60it/s]
 98%|█████████▊| 365665/371472 [6:56:05<26:55,  3.60it/s]
 98%|█████████▊| 365666/371472 [6:56:05<26:43,  3.62it/s]
 98%|█████████▊| 365667/371472 [6:56:05<26:16,  3.68it/s]
 98%|█████████▊| 365668/371472 [6:56:06<27:46,  3.48it/s]
 98%|█████████▊| 365669/371472 [6:56:06<27:45,  3.49it/s]
 98%|█████████▊| 365670/371472 [6:56:06<27:50,  3.47it/s]
 98%|█████████▊| 365671/371472 [6:56:07<27:53,  3.47it/s]
 98%|█████████▊| 365672/371472 [6:56:07<26:54,  3.59it/s]
 98%|█████████▊| 365673/371472 [6:56:07<26:25,  3.66it/s]
 98%|█████████▊| 365674/371472 [6:56:07<25:26,  3.80it/s]
 98%|█████████▊| 365675/371472 [6:56:08<26:45,  3.61it/s]
 98%|█████████▊| 365676/371472 [6:56:08<25:56,  3.72it/s]
 98%|█████████▊| 365677/371472 [6:56:08<25:41,  3.76it/s]
 98%|█████████▊| 365678/371472 [6:56:08<28:36,  3.38it/s]
 98%|█████████▊| 365679/371472 [6:56:09<28:00,  3.45it/s]
 98%|█████████▊| 365680/371472 [6:56:09<27:29,  3.51it/s]
                                                         
{'loss': 2.565, 'learning_rate': 1.1404038009868778e-07, 'epoch': 15.75}
+
 98%|█████████▊| 365680/371472 [6:56:09<27:29,  3.51it/s]
 98%|█████████▊| 365681/371472 [6:56:09<27:07,  3.56it/s]
 98%|█████████▊| 365682/371472 [6:56:10<26:08,  3.69it/s]
 98%|█████████▊| 365683/371472 [6:56:10<28:20,  3.40it/s]
 98%|█████████▊| 365684/371472 [6:56:10<27:46,  3.47it/s]
 98%|█████████▊| 365685/371472 [6:56:11<31:17,  3.08it/s]
 98%|█████████▊| 365686/371472 [6:56:11<29:55,  3.22it/s]
 98%|█████████▊| 365687/371472 [6:56:11<28:37,  3.37it/s]
 98%|█████████▊| 365688/371472 [6:56:11<29:44,  3.24it/s]
 98%|█████████▊| 365689/371472 [6:56:12<30:08,  3.20it/s]
 98%|█████████▊| 365690/371472 [6:56:12<27:54,  3.45it/s]
 98%|█████████▊| 365691/371472 [6:56:12<28:46,  3.35it/s]
 98%|█████████▊| 365692/371472 [6:56:13<28:05,  3.43it/s]
 98%|█████████▊| 365693/371472 [6:56:13<27:21,  3.52it/s]
 98%|█████████▊| 365694/371472 [6:56:13<26:41,  3.61it/s]
 98%|█████████▊| 365695/371472 [6:56:13<27:32,  3.50it/s]
 98%|█████████▊| 365696/371472 [6:56:14<26:19,  3.66it/s]
 98%|█████████▊| 365697/371472 [6:56:14<26:38,  3.61it/s]
 98%|█████████▊| 365698/371472 [6:56:14<26:56,  3.57it/s]
 98%|█████████▊| 365699/371472 [6:56:15<26:09,  3.68it/s]
 98%|█████████▊| 365700/371472 [6:56:15<27:09,  3.54it/s]
                                                         
{'loss': 2.6777, 'learning_rate': 1.1399189812320884e-07, 'epoch': 15.75}
+
 98%|█████████▊| 365700/371472 [6:56:15<27:09,  3.54it/s]
 98%|█████████▊| 365701/371472 [6:56:15<26:33,  3.62it/s]
 98%|█████████▊| 365702/371472 [6:56:15<27:24,  3.51it/s]
 98%|█████████▊| 365703/371472 [6:56:16<27:16,  3.52it/s]
 98%|█████████▊| 365704/371472 [6:56:16<27:57,  3.44it/s]
 98%|█████████▊| 365705/371472 [6:56:16<27:26,  3.50it/s]
 98%|█████████▊| 365706/371472 [6:56:17<26:57,  3.56it/s]
 98%|█████████▊| 365707/371472 [6:56:17<27:23,  3.51it/s]
 98%|█████████▊| 365708/371472 [6:56:17<27:58,  3.43it/s]
 98%|█████████▊| 365709/371472 [6:56:17<28:08,  3.41it/s]
 98%|█████████▊| 365710/371472 [6:56:18<27:20,  3.51it/s]
 98%|█████████▊| 365711/371472 [6:56:18<26:26,  3.63it/s]
 98%|█████████▊| 365712/371472 [6:56:18<26:44,  3.59it/s]
 98%|█████████▊| 365713/371472 [6:56:19<26:32,  3.62it/s]
 98%|█████████▊| 365714/371472 [6:56:19<26:35,  3.61it/s]
 98%|█████████▊| 365715/371472 [6:56:19<26:35,  3.61it/s]
 98%|█████████▊| 365716/371472 [6:56:19<26:12,  3.66it/s]
 98%|█████████▊| 365717/371472 [6:56:20<27:17,  3.52it/s]
 98%|█████████▊| 365718/371472 [6:56:20<26:12,  3.66it/s]
 98%|█████████▊| 365719/371472 [6:56:20<26:00,  3.69it/s]
 98%|█████████▊| 365720/371472 [6:56:20<25:42,  3.73it/s]
                                                         
{'loss': 2.6527, 'learning_rate': 1.1394341614772999e-07, 'epoch': 15.75}
+
 98%|█████████▊| 365720/371472 [6:56:20<25:42,  3.73it/s]
 98%|█████████▊| 365721/371472 [6:56:21<27:33,  3.48it/s]
 98%|█████████▊| 365722/371472 [6:56:21<26:14,  3.65it/s]
 98%|█████████▊| 365723/371472 [6:56:21<26:06,  3.67it/s]
 98%|█████████▊| 365724/371472 [6:56:22<26:49,  3.57it/s]
 98%|█████████▊| 365725/371472 [6:56:22<28:35,  3.35it/s]
 98%|█████████▊| 365726/371472 [6:56:22<27:12,  3.52it/s]
 98%|█████████▊| 365727/371472 [6:56:22<27:30,  3.48it/s]
 98%|█████████▊| 365728/371472 [6:56:23<28:09,  3.40it/s]
 98%|█████████▊| 365729/371472 [6:56:23<27:48,  3.44it/s]
 98%|█████████▊| 365730/371472 [6:56:23<27:26,  3.49it/s]
 98%|█████████▊| 365731/371472 [6:56:24<26:04,  3.67it/s]
 98%|█████████▊| 365732/371472 [6:56:24<25:28,  3.76it/s]
 98%|█████████▊| 365733/371472 [6:56:24<26:25,  3.62it/s]
 98%|█████████▊| 365734/371472 [6:56:24<27:28,  3.48it/s]
 98%|█████████▊| 365735/371472 [6:56:25<27:32,  3.47it/s]
 98%|█████████▊| 365736/371472 [6:56:25<27:22,  3.49it/s]
 98%|█████████▊| 365737/371472 [6:56:25<26:29,  3.61it/s]
 98%|█████████▊| 365738/371472 [6:56:26<26:09,  3.65it/s]
 98%|█████████▊| 365739/371472 [6:56:26<25:35,  3.73it/s]
 98%|█████████▊| 365740/371472 [6:56:26<25:59,  3.68it/s]
                                                         
{'loss': 2.6014, 'learning_rate': 1.1389493417225105e-07, 'epoch': 15.75}
+
 98%|█��███████▊| 365740/371472 [6:56:26<25:59,  3.68it/s]
 98%|█████████▊| 365741/371472 [6:56:26<25:55,  3.68it/s]
 98%|█████████▊| 365742/371472 [6:56:27<25:29,  3.75it/s]
 98%|█████████▊| 365743/371472 [6:56:27<25:40,  3.72it/s]
 98%|█████████▊| 365744/371472 [6:56:27<25:03,  3.81it/s]
 98%|█████████▊| 365745/371472 [6:56:27<25:23,  3.76it/s]
 98%|█████████▊| 365746/371472 [6:56:28<25:15,  3.78it/s]
 98%|█████████▊| 365747/371472 [6:56:28<25:38,  3.72it/s]
 98%|█████████▊| 365748/371472 [6:56:28<26:41,  3.57it/s]
 98%|█████████▊| 365749/371472 [6:56:29<26:58,  3.54it/s]
 98%|█████████▊| 365750/371472 [6:56:29<26:29,  3.60it/s]
 98%|█████████▊| 365751/371472 [6:56:29<27:17,  3.49it/s]
 98%|█████████▊| 365752/371472 [6:56:29<27:20,  3.49it/s]
 98%|█████████▊| 365753/371472 [6:56:30<27:30,  3.47it/s]
 98%|█████████▊| 365754/371472 [6:56:30<26:01,  3.66it/s]
 98%|█████████▊| 365755/371472 [6:56:30<26:21,  3.61it/s]
 98%|█████████▊| 365756/371472 [6:56:30<27:36,  3.45it/s]
 98%|█████████▊| 365757/371472 [6:56:31<27:59,  3.40it/s]
 98%|█████████▊| 365758/371472 [6:56:31<27:33,  3.46it/s]
 98%|█████████▊| 365759/371472 [6:56:31<27:42,  3.44it/s]
 98%|█████████▊| 365760/371472 [6:56:32<26:33,  3.58it/s]
                                                         
{'loss': 2.5452, 'learning_rate': 1.138464521967722e-07, 'epoch': 15.75}
+
 98%|█████████▊| 365760/371472 [6:56:32<26:33,  3.58it/s]
 98%|█████████▊| 365761/371472 [6:56:32<26:07,  3.64it/s]
 98%|█████████▊| 365762/371472 [6:56:32<27:12,  3.50it/s]
 98%|█████████▊| 365763/371472 [6:56:32<26:48,  3.55it/s]
 98%|█████████▊| 365764/371472 [6:56:33<27:05,  3.51it/s]
 98%|█████████▊| 365765/371472 [6:56:33<26:03,  3.65it/s]
 98%|█████████▊| 365766/371472 [6:56:33<27:25,  3.47it/s]
 98%|█████████▊| 365767/371472 [6:56:34<26:43,  3.56it/s]
 98%|█████████▊| 365768/371472 [6:56:34<26:03,  3.65it/s]
 98%|█████████▊| 365769/371472 [6:56:34<25:38,  3.71it/s]
 98%|█████████▊| 365770/371472 [6:56:34<25:30,  3.73it/s]
 98%|█████████▊| 365771/371472 [6:56:35<26:29,  3.59it/s]
 98%|█████████▊| 365772/371472 [6:56:35<25:25,  3.74it/s]
 98%|█████████▊| 365773/371472 [6:56:35<26:18,  3.61it/s]
 98%|█████████▊| 365774/371472 [6:56:35<25:04,  3.79it/s]
 98%|█████████▊| 365775/371472 [6:56:36<24:27,  3.88it/s]
 98%|█████████▊| 365776/371472 [6:56:36<25:33,  3.71it/s]
 98%|█████████▊| 365777/371472 [6:56:36<28:47,  3.30it/s]
 98%|█████████▊| 365778/371472 [6:56:37<27:57,  3.39it/s]
 98%|█████████▊| 365779/371472 [6:56:37<29:46,  3.19it/s]
 98%|█████████▊| 365780/371472 [6:56:37<28:11,  3.37it/s]
                                                         
{'loss': 2.581, 'learning_rate': 1.1379797022129325e-07, 'epoch': 15.75}
+
 98%|█████████▊| 365780/371472 [6:56:37<28:11,  3.37it/s]
 98%|█████████▊| 365781/371472 [6:56:38<26:45,  3.54it/s]
 98%|█████████▊| 365782/371472 [6:56:38<26:01,  3.64it/s]
 98%|█████████▊| 365783/371472 [6:56:38<25:06,  3.78it/s]
 98%|█████████▊| 365784/371472 [6:56:38<25:12,  3.76it/s]
 98%|█████████▊| 365785/371472 [6:56:39<26:31,  3.57it/s]
 98%|█████████▊| 365786/371472 [6:56:39<25:51,  3.66it/s]
 98%|█████████▊| 365787/371472 [6:56:39<25:33,  3.71it/s]
 98%|█████████▊| 365788/371472 [6:56:39<26:00,  3.64it/s]
 98%|█████████▊| 365789/371472 [6:56:40<26:08,  3.62it/s]
 98%|█████████▊| 365790/371472 [6:56:40<27:25,  3.45it/s]
 98%|█████████▊| 365791/371472 [6:56:40<26:31,  3.57it/s]
 98%|█████████▊| 365792/371472 [6:56:41<26:50,  3.53it/s]
 98%|█████████▊| 365793/371472 [6:56:41<26:00,  3.64it/s]
 98%|█████████▊| 365794/371472 [6:56:41<29:57,  3.16it/s]
 98%|█████████▊| 365795/371472 [6:56:42<29:39,  3.19it/s]
 98%|█████████▊| 365796/371472 [6:56:42<33:59,  2.78it/s]
 98%|█████████▊| 365797/371472 [6:56:42<32:44,  2.89it/s]
 98%|█████████▊| 365798/371472 [6:56:43<29:58,  3.16it/s]
 98%|█████████▊| 365799/371472 [6:56:43<27:24,  3.45it/s]
 98%|███████��█▊| 365800/371472 [6:56:43<26:28,  3.57it/s]
                                                         
{'loss': 2.6662, 'learning_rate': 1.1374948824581442e-07, 'epoch': 15.76}
+
 98%|█████████▊| 365800/371472 [6:56:43<26:28,  3.57it/s]
 98%|█████████▊| 365801/371472 [6:56:43<26:07,  3.62it/s]
 98%|█████████▊| 365802/371472 [6:56:44<27:33,  3.43it/s]
 98%|█████████▊| 365803/371472 [6:56:44<27:21,  3.45it/s]
 98%|█████████▊| 365804/371472 [6:56:44<26:39,  3.54it/s]
 98%|█████████▊| 365805/371472 [6:56:44<26:18,  3.59it/s]
 98%|█████████▊| 365806/371472 [6:56:45<25:46,  3.66it/s]
 98%|█████████▊| 365807/371472 [6:56:45<26:34,  3.55it/s]
 98%|█████████▊| 365808/371472 [6:56:45<25:31,  3.70it/s]
 98%|█████████▊| 365809/371472 [6:56:46<25:41,  3.67it/s]
 98%|█████████▊| 365810/371472 [6:56:46<24:46,  3.81it/s]
 98%|█████████▊| 365811/371472 [6:56:46<25:50,  3.65it/s]
 98%|█████████▊| 365812/371472 [6:56:46<26:05,  3.62it/s]
 98%|█████████▊| 365813/371472 [6:56:47<25:27,  3.70it/s]
 98%|█████████▊| 365814/371472 [6:56:47<24:54,  3.79it/s]
 98%|█████████▊| 365815/371472 [6:56:47<26:55,  3.50it/s]
 98%|█████████▊| 365816/371472 [6:56:48<27:59,  3.37it/s]
 98%|█████████▊| 365817/371472 [6:56:48<27:37,  3.41it/s]
 98%|█████████▊| 365818/371472 [6:56:48<26:31,  3.55it/s]
 98%|█████████▊| 365819/371472 [6:56:48<26:35,  3.54it/s]
 98%|█████████▊| 365820/371472 [6:56:49<27:39,  3.40it/s]
                                                         
{'loss': 2.5288, 'learning_rate': 1.1370100627033548e-07, 'epoch': 15.76}
+
 98%|█████████▊| 365820/371472 [6:56:49<27:39,  3.40it/s]
 98%|█████████▊| 365821/371472 [6:56:49<28:19,  3.32it/s]
 98%|█████████▊| 365822/371472 [6:56:49<27:30,  3.42it/s]
 98%|█████████▊| 365823/371472 [6:56:50<26:35,  3.54it/s]
 98%|█████████▊| 365824/371472 [6:56:50<27:15,  3.45it/s]
 98%|█████████▊| 365825/371472 [6:56:50<29:16,  3.22it/s]
 98%|█████████▊| 365826/371472 [6:56:50<28:24,  3.31it/s]
 98%|█████████▊| 365827/371472 [6:56:51<27:59,  3.36it/s]
 98%|█████████▊| 365828/371472 [6:56:51<28:05,  3.35it/s]
 98%|█████████▊| 365829/371472 [6:56:51<26:44,  3.52it/s]
 98%|█████████▊| 365830/371472 [6:56:52<25:46,  3.65it/s]
 98%|█████████▊| 365831/371472 [6:56:52<27:26,  3.43it/s]
 98%|█████████▊| 365832/371472 [6:56:52<26:44,  3.51it/s]
 98%|█████████▊| 365833/371472 [6:56:52<26:36,  3.53it/s]
 98%|█████████▊| 365834/371472 [6:56:53<25:31,  3.68it/s]
 98%|█████████▊| 365835/371472 [6:56:53<24:45,  3.79it/s]
 98%|█████████▊| 365836/371472 [6:56:53<24:35,  3.82it/s]
 98%|█████████▊| 365837/371472 [6:56:53<24:54,  3.77it/s]
 98%|█████████▊| 365838/371472 [6:56:54<24:31,  3.83it/s]
 98%|█████████▊| 365839/371472 [6:56:54<24:14,  3.87it/s]
 98%|█████████▊| 365840/371472 [6:56:54<24:12,  3.88it/s]
                                                         
{'loss': 2.6536, 'learning_rate': 1.1365252429485663e-07, 'epoch': 15.76}
+
 98%|█████████▊| 365840/371472 [6:56:54<24:12,  3.88it/s]
 98%|█████████▊| 365841/371472 [6:56:54<24:25,  3.84it/s]
 98%|█████████▊| 365842/371472 [6:56:55<24:26,  3.84it/s]
 98%|█████████▊| 365843/371472 [6:56:55<24:24,  3.84it/s]
 98%|█████████▊| 365844/371472 [6:56:55<24:36,  3.81it/s]
 98%|█████████▊| 365845/371472 [6:56:56<25:40,  3.65it/s]
 98%|█████████▊| 365846/371472 [6:56:56<24:34,  3.82it/s]
 98%|█████████▊| 365847/371472 [6:56:56<26:28,  3.54it/s]
 98%|█████████▊| 365848/371472 [6:56:56<25:44,  3.64it/s]
 98%|█████████▊| 365849/371472 [6:56:57<25:12,  3.72it/s]
 98%|█████████▊| 365850/371472 [6:56:57<25:57,  3.61it/s]
 98%|█████████▊| 365851/371472 [6:56:57<25:57,  3.61it/s]
 98%|█████████▊| 365852/371472 [6:56:57<25:40,  3.65it/s]
 98%|█████████▊| 365853/371472 [6:56:58<26:02,  3.60it/s]
 98%|█████████▊| 365854/371472 [6:56:58<28:51,  3.25it/s]
 98%|█████████▊| 365855/371472 [6:56:58<27:40,  3.38it/s]
 98%|█████████▊| 365856/371472 [6:56:59<28:16,  3.31it/s]
 98%|█████████▊| 365857/371472 [6:56:59<27:27,  3.41it/s]
 98%|█████████▊| 365858/371472 [6:56:59<28:39,  3.26it/s]
 98%|█████████▊| 365859/371472 [6:57:00<26:57,  3.47it/s]
 98%|█████████▊| 365860/371472 [6:57:00<26:06,  3.58it/s]
                                                         
{'loss': 2.7102, 'learning_rate': 1.1360404231937769e-07, 'epoch': 15.76}
+
 98%|█████████▊| 365860/371472 [6:57:00<26:06,  3.58it/s]
 98%|█████████▊| 365861/371472 [6:57:00<25:31,  3.66it/s]
 98%|█████████▊| 365862/371472 [6:57:00<26:01,  3.59it/s]
 98%|█████████▊| 365863/371472 [6:57:01<26:02,  3.59it/s]
 98%|█████████▊| 365864/371472 [6:57:01<25:15,  3.70it/s]
 98%|█████████▊| 365865/371472 [6:57:01<25:28,  3.67it/s]
 98%|█████████▊| 365866/371472 [6:57:02<26:32,  3.52it/s]
 98%|█████████▊| 365867/371472 [6:57:02<27:25,  3.41it/s]
 98%|█████████▊| 365868/371472 [6:57:02<28:37,  3.26it/s]
 98%|█████████▊| 365869/371472 [6:57:03<29:08,  3.20it/s]
 98%|█████████▊| 365870/371472 [6:57:03<28:21,  3.29it/s]
 98%|█████████▊| 365871/371472 [6:57:03<32:02,  2.91it/s]
 98%|█████████▊| 365872/371472 [6:57:03<30:02,  3.11it/s]
 98%|█████████▊| 365873/371472 [6:57:04<31:13,  2.99it/s]
 98%|█████████▊| 365874/371472 [6:57:04<28:50,  3.23it/s]
 98%|█████████▊| 365875/371472 [6:57:04<27:38,  3.38it/s]
 98%|█████████▊| 365876/371472 [6:57:05<27:16,  3.42it/s]
 98%|█████████▊| 365877/371472 [6:57:05<26:22,  3.54it/s]
 98%|█████████▊| 365878/371472 [6:57:05<28:05,  3.32it/s]
 98%|█████████▊| 365879/371472 [6:57:06<27:25,  3.40it/s]
 98%|█████████▊| 365880/371472 [6:57:06<26:10,  3.56it/s]
                                                         
{'loss': 2.5644, 'learning_rate': 1.1355556034389885e-07, 'epoch': 15.76}
+
 98%|█████████▊| 365880/371472 [6:57:06<26:10,  3.56it/s]
 98%|█████████▊| 365881/371472 [6:57:06<26:07,  3.57it/s]
 98%|█████████▊| 365882/371472 [6:57:06<25:52,  3.60it/s]
 98%|█████████▊| 365883/371472 [6:57:07<24:51,  3.75it/s]
 98%|█████████▊| 365884/371472 [6:57:07<25:31,  3.65it/s]
 98%|█████████▊| 365885/371472 [6:57:07<24:27,  3.81it/s]
 98%|█████████▊| 365886/371472 [6:57:07<24:02,  3.87it/s]
 98%|█████████▊| 365887/371472 [6:57:08<24:00,  3.88it/s]
 98%|█████████▊| 365888/371472 [6:57:08<23:59,  3.88it/s]
 98%|█████████▊| 365889/371472 [6:57:08<26:11,  3.55it/s]
 98%|█████████▊| 365890/371472 [6:57:08<25:20,  3.67it/s]
 98%|█████████▊| 365891/371472 [6:57:09<24:41,  3.77it/s]
 98%|█████████▊| 365892/371472 [6:57:09<25:12,  3.69it/s]
 98%|█████████▊| 365893/371472 [6:57:09<24:32,  3.79it/s]
 98%|█████████▊| 365894/371472 [6:57:09<24:12,  3.84it/s]
 98%|█████████▊| 365895/371472 [6:57:10<24:30,  3.79it/s]
 98%|█████████▊| 365896/371472 [6:57:10<24:39,  3.77it/s]
 98%|█████████▊| 365897/371472 [6:57:10<26:52,  3.46it/s]
 98%|█████████▊| 365898/371472 [6:57:11<27:03,  3.43it/s]
 98%|█████████▊| 365899/371472 [6:57:11<29:28,  3.15it/s]
 99%|█████████▊| 365900/371472 [6:57:11<27:53,  3.33it/s]
                                                         
{'loss': 2.6228, 'learning_rate': 1.1350707836841991e-07, 'epoch': 15.76}
+
 99%|█████████▊| 365900/371472 [6:57:11<27:53,  3.33it/s]
 99%|█████████▊| 365901/371472 [6:57:12<28:50,  3.22it/s]
 99%|█████████▊| 365902/371472 [6:57:12<26:57,  3.44it/s]
 99%|█████████▊| 365903/371472 [6:57:12<28:06,  3.30it/s]
 99%|█████████▊| 365904/371472 [6:57:12<26:53,  3.45it/s]
 99%|█████████▊| 365905/371472 [6:57:13<26:30,  3.50it/s]
 99%|█████████▊| 365906/371472 [6:57:13<25:51,  3.59it/s]
 99%|█████████▊| 365907/371472 [6:57:13<26:26,  3.51it/s]
 99%|█████████▊| 365908/371472 [6:57:14<26:33,  3.49it/s]
 99%|█████████▊| 365909/371472 [6:57:14<25:09,  3.69it/s]
 99%|█████████▊| 365910/371472 [6:57:14<25:00,  3.71it/s]
 99%|█████████▊| 365911/371472 [6:57:14<24:34,  3.77it/s]
 99%|█████████▊| 365912/371472 [6:57:15<24:47,  3.74it/s]
 99%|█████████▊| 365913/371472 [6:57:15<24:26,  3.79it/s]
 99%|█████████▊| 365914/371472 [6:57:15<24:52,  3.72it/s]
 99%|█████████▊| 365915/371472 [6:57:15<24:27,  3.79it/s]
 99%|█████████▊| 365916/371472 [6:57:16<24:42,  3.75it/s]
 99%|█████████▊| 365917/371472 [6:57:16<24:40,  3.75it/s]
 99%|█████████▊| 365918/371472 [6:57:16<25:08,  3.68it/s]
 99%|█████████▊| 365919/371472 [6:57:16<24:32,  3.77it/s]
 99%|█████████▊| 365920/371472 [6:57:17<24:09,  3.83it/s]
                                                         
{'loss': 2.6566, 'learning_rate': 1.1345859639294106e-07, 'epoch': 15.76}
+
 99%|█████████▊| 365920/371472 [6:57:17<24:09,  3.83it/s]
 99%|█████████▊| 365921/371472 [6:57:17<24:03,  3.85it/s]
 99%|█████████▊| 365922/371472 [6:57:17<24:32,  3.77it/s]
 99%|█████████▊| 365923/371472 [6:57:18<25:12,  3.67it/s]
 99%|█████████▊| 365924/371472 [6:57:18<24:55,  3.71it/s]
 99%|█████████▊| 365925/371472 [6:57:18<26:11,  3.53it/s]
 99%|█████████▊| 365926/371472 [6:57:18<26:47,  3.45it/s]
 99%|█████████▊| 365927/371472 [6:57:19<26:03,  3.55it/s]
 99%|█████████▊| 365928/371472 [6:57:19<26:28,  3.49it/s]
 99%|█████████▊| 365929/371472 [6:57:19<26:27,  3.49it/s]
 99%|█████████▊| 365930/371472 [6:57:20<26:01,  3.55it/s]
 99%|█████████▊| 365931/371472 [6:57:20<25:45,  3.58it/s]
 99%|█████████▊| 365932/371472 [6:57:20<25:47,  3.58it/s]
 99%|█████████▊| 365933/371472 [6:57:20<26:04,  3.54it/s]
 99%|█████████▊| 365934/371472 [6:57:21<25:57,  3.55it/s]
 99%|█████████▊| 365935/371472 [6:57:21<26:16,  3.51it/s]
 99%|█████████▊| 365936/371472 [6:57:21<27:06,  3.40it/s]
 99%|█████████▊| 365937/371472 [6:57:22<26:57,  3.42it/s]
 99%|█████████▊| 365938/371472 [6:57:22<27:07,  3.40it/s]
 99%|█████████▊| 365939/371472 [6:57:22<25:37,  3.60it/s]
 99%|█████████▊| 365940/371472 [6:57:23<29:42,  3.10it/s]
                                                         
{'loss': 2.5764, 'learning_rate': 1.1341011441746212e-07, 'epoch': 15.76}
+
 99%|█████████▊| 365940/371472 [6:57:23<29:42,  3.10it/s]
 99%|█████████▊| 365941/371472 [6:57:23<27:37,  3.34it/s]
 99%|█████████▊| 365942/371472 [6:57:23<26:31,  3.48it/s]
 99%|█████████▊| 365943/371472 [6:57:23<25:39,  3.59it/s]
 99%|█████████▊| 365944/371472 [6:57:24<24:53,  3.70it/s]
 99%|█████████▊| 365945/371472 [6:57:24<28:12,  3.27it/s]
 99%|█████████▊| 365946/371472 [6:57:24<26:51,  3.43it/s]
 99%|█████████▊| 365947/371472 [6:57:24<25:47,  3.57it/s]
 99%|█████████▊| 365948/371472 [6:57:25<26:12,  3.51it/s]
 99%|█████████▊| 365949/371472 [6:57:25<25:10,  3.66it/s]
 99%|█████████▊| 365950/371472 [6:57:25<24:07,  3.81it/s]
 99%|█████████▊| 365951/371472 [6:57:26<23:57,  3.84it/s]
 99%|█████████▊| 365952/371472 [6:57:26<25:33,  3.60it/s]
 99%|█████████▊| 365953/371472 [6:57:26<26:49,  3.43it/s]
 99%|█████████▊| 365954/371472 [6:57:26<26:10,  3.51it/s]
 99%|█████████▊| 365955/371472 [6:57:27<25:11,  3.65it/s]
 99%|█████████▊| 365956/371472 [6:57:27<25:19,  3.63it/s]
 99%|█████████▊| 365957/371472 [6:57:27<24:27,  3.76it/s]
 99%|█████████▊| 365958/371472 [6:57:27<24:30,  3.75it/s]
 99%|█████████▊| 365959/371472 [6:57:28<24:45,  3.71it/s]
 99%|█████████▊| 365960/371472 [6:57:28<24:44,  3.71it/s]
                                                         
{'loss': 2.6615, 'learning_rate': 1.1336163244198328e-07, 'epoch': 15.76}
+
 99%|█████████▊| 365960/371472 [6:57:28<24:44,  3.71it/s]
 99%|█████████▊| 365961/371472 [6:57:28<24:03,  3.82it/s]
 99%|█████████▊| 365962/371472 [6:57:29<26:31,  3.46it/s]
 99%|█████████▊| 365963/371472 [6:57:29<26:38,  3.45it/s]
 99%|█████████▊| 365964/371472 [6:57:29<27:16,  3.37it/s]
 99%|█████████▊| 365965/371472 [6:57:30<27:57,  3.28it/s]
 99%|█████████▊| 365966/371472 [6:57:30<28:35,  3.21it/s]
 99%|█████████▊| 365967/371472 [6:57:30<27:37,  3.32it/s]
 99%|█████████▊| 365968/371472 [6:57:30<27:06,  3.38it/s]
 99%|█████████▊| 365969/371472 [6:57:31<26:31,  3.46it/s]
 99%|█████████▊| 365970/371472 [6:57:31<25:47,  3.56it/s]
 99%|█████████▊| 365971/371472 [6:57:31<24:53,  3.68it/s]
 99%|█████████▊| 365972/371472 [6:57:31<24:50,  3.69it/s]
 99%|█████████▊| 365973/371472 [6:57:32<24:31,  3.74it/s]
 99%|█████████▊| 365974/371472 [6:57:32<26:44,  3.43it/s]
 99%|█████████▊| 365975/371472 [6:57:32<26:38,  3.44it/s]
 99%|█████████▊| 365976/371472 [6:57:33<25:23,  3.61it/s]
 99%|█████████▊| 365977/371472 [6:57:33<25:31,  3.59it/s]
 99%|█████████▊| 365978/371472 [6:57:33<25:40,  3.57it/s]
 99%|█████████▊| 365979/371472 [6:57:34<28:53,  3.17it/s]
 99%|█████████▊| 365980/371472 [6:57:34<29:49,  3.07it/s]
                                                         
{'loss': 2.6155, 'learning_rate': 1.1331315046650434e-07, 'epoch': 15.76}
+
 99%|█████████▊| 365980/371472 [6:57:34<29:49,  3.07it/s]
 99%|█████████▊| 365981/371472 [6:57:34<28:06,  3.26it/s]
 99%|█████████▊| 365982/371472 [6:57:34<26:06,  3.50it/s]
 99%|█████████▊| 365983/371472 [6:57:35<29:13,  3.13it/s]
 99%|█████████▊| 365984/371472 [6:57:35<28:14,  3.24it/s]
 99%|█████████▊| 365985/371472 [6:57:35<27:24,  3.34it/s]
 99%|█████████▊| 365986/371472 [6:57:36<26:52,  3.40it/s]
 99%|█████████▊| 365987/371472 [6:57:36<27:01,  3.38it/s]
 99%|█████████▊| 365988/371472 [6:57:36<26:33,  3.44it/s]
 99%|█████████▊| 365989/371472 [6:57:37<25:56,  3.52it/s]
 99%|█████████▊| 365990/371472 [6:57:37<25:48,  3.54it/s]
 99%|█████████▊| 365991/371472 [6:57:37<25:27,  3.59it/s]
 99%|█████████▊| 365992/371472 [6:57:37<24:59,  3.66it/s]
 99%|█████████▊| 365993/371472 [6:57:38<28:12,  3.24it/s]
 99%|█████████▊| 365994/371472 [6:57:38<26:31,  3.44it/s]
 99%|█████████▊| 365995/371472 [6:57:38<25:33,  3.57it/s]
 99%|█████████▊| 365996/371472 [6:57:39<25:36,  3.56it/s]
 99%|█████████▊| 365997/371472 [6:57:39<25:32,  3.57it/s]
 99%|█████████▊| 365998/371472 [6:57:39<25:03,  3.64it/s]
 99%|█████████▊| 365999/371472 [6:57:39<24:25,  3.74it/s]
 99%|█████████▊| 366000/371472 [6:57:40<25:07,  3.63it/s]
                                                         
{'loss': 2.6458, 'learning_rate': 1.132646684910254e-07, 'epoch': 15.76}
+
 99%|█████████▊| 366000/371472 [6:57:40<25:07,  3.63it/s]
 99%|█████████▊| 366001/371472 [6:57:40<25:13,  3.62it/s]
 99%|█████████▊| 366002/371472 [6:57:40<25:04,  3.64it/s]
 99%|█████████▊| 366003/371472 [6:57:41<28:16,  3.22it/s]
 99%|█████████▊| 366004/371472 [6:57:41<28:56,  3.15it/s]
 99%|█████████▊| 366005/371472 [6:57:41<28:59,  3.14it/s]
 99%|█████████▊| 366006/371472 [6:57:42<30:17,  3.01it/s]
 99%|█████████▊| 366007/371472 [6:57:42<28:38,  3.18it/s]
 99%|█████████▊| 366008/371472 [6:57:42<29:27,  3.09it/s]
 99%|█████████▊| 366009/371472 [6:57:42<27:46,  3.28it/s]
 99%|█████████▊| 366010/371472 [6:57:43<26:07,  3.48it/s]
 99%|█████████▊| 366011/371472 [6:57:43<28:48,  3.16it/s]
 99%|█████████▊| 366012/371472 [6:57:43<26:26,  3.44it/s]
 99%|█████████▊| 366013/371472 [6:57:44<25:47,  3.53it/s]
 99%|█████████▊| 366014/371472 [6:57:44<24:24,  3.73it/s]
 99%|█████████▊| 366015/371472 [6:57:44<24:23,  3.73it/s]
 99%|█████████▊| 366016/371472 [6:57:44<24:01,  3.79it/s]
 99%|█████████▊| 366017/371472 [6:57:45<25:37,  3.55it/s]
 99%|█████████▊| 366018/371472 [6:57:45<25:22,  3.58it/s]
 99%|█████████▊| 366019/371472 [6:57:45<24:23,  3.73it/s]
 99%|█████████▊| 366020/371472 [6:57:45<24:47,  3.66it/s]
                                                         
{'loss': 2.4817, 'learning_rate': 1.1321618651554655e-07, 'epoch': 15.77}
+
 99%|█████████▊| 366020/371472 [6:57:45<24:47,  3.66it/s]
 99%|█████████▊| 366021/371472 [6:57:46<24:41,  3.68it/s]
 99%|█████████▊| 366022/371472 [6:57:46<28:00,  3.24it/s]
 99%|█████████▊| 366023/371472 [6:57:46<27:43,  3.28it/s]
 99%|█████████▊| 366024/371472 [6:57:47<26:22,  3.44it/s]
 99%|█████████▊| 366025/371472 [6:57:47<25:45,  3.52it/s]
 99%|█████████▊| 366026/371472 [6:57:47<26:01,  3.49it/s]
 99%|█████████▊| 366027/371472 [6:57:47<25:21,  3.58it/s]
 99%|█████████▊| 366028/371472 [6:57:48<24:43,  3.67it/s]
 99%|█████████▊| 366029/371472 [6:57:48<24:43,  3.67it/s]
 99%|██████��██▊| 366030/371472 [6:57:48<23:53,  3.80it/s]
 99%|█████████▊| 366031/371472 [6:57:49<23:54,  3.79it/s]
 99%|█████████▊| 366032/371472 [6:57:49<23:34,  3.85it/s]
 99%|█████████▊| 366033/371472 [6:57:49<23:08,  3.92it/s]
 99%|█████████▊| 366034/371472 [6:57:49<23:37,  3.84it/s]
 99%|█████████▊| 366035/371472 [6:57:50<23:52,  3.80it/s]
 99%|█████████▊| 366036/371472 [6:57:50<25:34,  3.54it/s]
 99%|█████████▊| 366037/371472 [6:57:50<26:09,  3.46it/s]
 99%|█████████▊| 366038/371472 [6:57:50<25:33,  3.54it/s]
 99%|█████████▊| 366039/371472 [6:57:51<24:53,  3.64it/s]
 99%|█████████▊| 366040/371472 [6:57:51<26:19,  3.44it/s]
                                                         
{'loss': 2.577, 'learning_rate': 1.1316770454006761e-07, 'epoch': 15.77}
+
 99%|█████████▊| 366040/371472 [6:57:51<26:19,  3.44it/s]
 99%|█████████▊| 366041/371472 [6:57:51<25:39,  3.53it/s]
 99%|█████████▊| 366042/371472 [6:57:52<24:56,  3.63it/s]
 99%|█████████▊| 366043/371472 [6:57:52<24:28,  3.70it/s]
 99%|█████████▊| 366044/371472 [6:57:52<25:00,  3.62it/s]
 99%|█████████▊| 366045/371472 [6:57:52<24:56,  3.63it/s]
 99%|█████████▊| 366046/371472 [6:57:53<25:20,  3.57it/s]
 99%|█████████▊| 366047/371472 [6:57:53<26:01,  3.47it/s]
 99%|█████████▊| 366048/371472 [6:57:53<24:55,  3.63it/s]
 99%|█████████▊| 366049/371472 [6:57:53<23:54,  3.78it/s]
 99%|█████████▊| 366050/371472 [6:57:54<25:30,  3.54it/s]
 99%|█████████▊| 366051/371472 [6:57:54<27:11,  3.32it/s]
 99%|█████████▊| 366052/371472 [6:57:54<26:23,  3.42it/s]
 99%|█████████▊| 366053/371472 [6:57:55<25:47,  3.50it/s]
 99%|█████████▊| 366054/371472 [6:57:55<25:03,  3.60it/s]
 99%|█████████▊| 366055/371472 [6:57:55<27:33,  3.28it/s]
 99%|█████████▊| 366056/371472 [6:57:56<26:51,  3.36it/s]
 99%|█████████▊| 366057/371472 [6:57:56<27:33,  3.28it/s]
 99%|█████████▊| 366058/371472 [6:57:56<26:53,  3.36it/s]
 99%|█████████▊| 366059/371472 [6:57:57<28:36,  3.15it/s]
 99%|█████████▊| 366060/371472 [6:57:57<27:05,  3.33it/s]
                                                         
{'loss': 2.5593, 'learning_rate': 1.1311922256458879e-07, 'epoch': 15.77}
+
 99%|█████████▊| 366060/371472 [6:57:57<27:05,  3.33it/s]
 99%|█████████▊| 366061/371472 [6:57:57<25:39,  3.52it/s]
 99%|█████████▊| 366062/371472 [6:57:57<25:09,  3.58it/s]
 99%|█████████▊| 366063/371472 [6:57:58<25:55,  3.48it/s]
 99%|█████████▊| 366064/371472 [6:57:58<26:20,  3.42it/s]
 99%|█████████▊| 366065/371472 [6:57:58<25:04,  3.59it/s]
 99%|█████████▊| 366066/371472 [6:57:58<25:22,  3.55it/s]
 99%|█████████▊| 366067/371472 [6:57:59<25:32,  3.53it/s]
 99%|█████████▊| 366068/371472 [6:57:59<24:34,  3.66it/s]
 99%|█████████▊| 366069/371472 [6:57:59<25:10,  3.58it/s]
 99%|█████████▊| 366070/371472 [6:58:00<25:29,  3.53it/s]
 99%|█████████▊| 366071/371472 [6:58:00<25:44,  3.50it/s]
 99%|█████████▊| 366072/371472 [6:58:00<26:04,  3.45it/s]
 99%|█████████▊| 366073/371472 [6:58:00<25:09,  3.58it/s]
 99%|█████████▊| 366074/371472 [6:58:01<25:05,  3.58it/s]
 99%|█████████▊| 366075/371472 [6:58:01<24:15,  3.71it/s]
 99%|█████████▊| 366076/371472 [6:58:01<26:08,  3.44it/s]
 99%|█████████▊| 366077/371472 [6:58:02<25:08,  3.58it/s]
 99%|█████████▊| 366078/371472 [6:58:02<26:43,  3.36it/s]
 99%|█████████▊| 366079/371472 [6:58:02<25:46,  3.49it/s]
 99%|█████████▊| 366080/371472 [6:58:02<27:00,  3.33it/s]
                                                         
{'loss': 2.4783, 'learning_rate': 1.1307074058910983e-07, 'epoch': 15.77}
+
 99%|█████████▊| 366080/371472 [6:58:02<27:00,  3.33it/s]
 99%|█████████▊| 366081/371472 [6:58:03<26:58,  3.33it/s]
 99%|█████████▊| 366082/371472 [6:58:03<27:37,  3.25it/s]
 99%|█████████▊| 366083/371472 [6:58:03<28:11,  3.19it/s]
 99%|█████████▊| 366084/371472 [6:58:04<27:17,  3.29it/s]
 99%|█████████▊| 366085/371472 [6:58:04<28:04,  3.20it/s]
 99%|█████████▊| 366086/371472 [6:58:04<28:06,  3.19it/s]
 99%|█████████▊| 366087/371472 [6:58:05<27:37,  3.25it/s]
 99%|█████████▊| 366088/371472 [6:58:05<27:05,  3.31it/s]
 99%|█████████▊| 366089/371472 [6:58:05<25:43,  3.49it/s]
 99%|█████████▊| 366090/371472 [6:58:05<25:27,  3.52it/s]
 99%|█████████▊| 366091/371472 [6:58:06<28:05,  3.19it/s]
 99%|█████████▊| 366092/371472 [6:58:06<27:26,  3.27it/s]
 99%|█████████▊| 366093/371472 [6:58:06<26:19,  3.41it/s]
 99%|█████████▊| 366094/371472 [6:58:07<25:30,  3.51it/s]
 99%|█████████▊| 366095/371472 [6:58:07<25:55,  3.46it/s]
 99%|█████████▊| 366096/371472 [6:58:07<27:00,  3.32it/s]
 99%|█████████▊| 366097/371472 [6:58:08<28:30,  3.14it/s]
 99%|█████████▊| 366098/371472 [6:58:08<26:59,  3.32it/s]
 99%|█████████▊| 366099/371472 [6:58:08<27:23,  3.27it/s]
 99%|█████████▊| 366100/371472 [6:58:09<26:31,  3.38it/s]
                                                         
{'loss': 2.6307, 'learning_rate': 1.1302225861363098e-07, 'epoch': 15.77}
+
 99%|█████████▊| 366100/371472 [6:58:09<26:31,  3.38it/s]
 99%|█████████▊| 366101/371472 [6:58:09<25:51,  3.46it/s]
 99%|█████████▊| 366102/371472 [6:58:09<25:26,  3.52it/s]
 99%|█████████▊| 366103/371472 [6:58:09<25:17,  3.54it/s]
 99%|█████████▊| 366104/371472 [6:58:10<27:16,  3.28it/s]
 99%|█████████▊| 366105/371472 [6:58:10<26:01,  3.44it/s]
 99%|█████████▊| 366106/371472 [6:58:10<25:36,  3.49it/s]
 99%|█████████▊| 366107/371472 [6:58:10<24:47,  3.61it/s]
 99%|█████████▊| 366108/371472 [6:58:11<25:16,  3.54it/s]
 99%|█████████▊| 366109/371472 [6:58:11<24:20,  3.67it/s]
 99%|█████████▊| 366110/371472 [6:58:11<24:02,  3.72it/s]
 99%|█████████▊| 366111/371472 [6:58:12<23:20,  3.83it/s]
 99%|█████████▊| 366112/371472 [6:58:12<23:34,  3.79it/s]
 99%|█████████▊| 366113/371472 [6:58:12<26:33,  3.36it/s]
 99%|█████████▊| 366114/371472 [6:58:12<26:00,  3.43it/s]
 99%|█████████▊| 366115/371472 [6:58:13<25:55,  3.44it/s]
 99%|█████████▊| 366116/371472 [6:58:13<26:25,  3.38it/s]
 99%|█████████▊| 366117/371472 [6:58:13<25:00,  3.57it/s]
 99%|█████████▊| 366118/371472 [6:58:14<25:21,  3.52it/s]
 99%|█████████▊| 366119/371472 [6:58:14<25:55,  3.44it/s]
 99%|█████████▊| 366120/371472 [6:58:14<24:59,  3.57it/s]
                                                         
{'loss': 2.5246, 'learning_rate': 1.1297377663815204e-07, 'epoch': 15.77}
+
 99%|█████████▊| 366120/371472 [6:58:14<24:59,  3.57it/s]
 99%|█████████▊| 366121/371472 [6:58:14<24:33,  3.63it/s]
 99%|█████████▊| 366122/371472 [6:58:15<23:38,  3.77it/s]
 99%|█████████▊| 366123/371472 [6:58:15<22:56,  3.89it/s]
 99%|█████████▊| 366124/371472 [6:58:15<24:29,  3.64it/s]
 99%|█████████▊| 366125/371472 [6:58:15<24:40,  3.61it/s]
 99%|█████████▊| 366126/371472 [6:58:16<24:20,  3.66it/s]
 99%|█████████▊| 366127/371472 [6:58:16<24:12,  3.68it/s]
 99%|█████████▊| 366128/371472 [6:58:16<25:18,  3.52it/s]
 99%|█████████▊| 366129/371472 [6:58:17<24:50,  3.59it/s]
 99%|█████████▊| 366130/371472 [6:58:17<24:22,  3.65it/s]
 99%|█████████▊| 366131/371472 [6:58:17<23:40,  3.76it/s]
 99%|█████████▊| 366132/371472 [6:58:17<23:50,  3.73it/s]
 99%|█████████▊| 366133/371472 [6:58:18<23:50,  3.73it/s]
 99%|█████████▊| 366134/371472 [6:58:18<24:37,  3.61it/s]
 99%|█████████▊| 366135/371472 [6:58:18<23:50,  3.73it/s]
 99%|█████████▊| 366136/371472 [6:58:19<26:02,  3.42it/s]
 99%|█████████▊| 366137/371472 [6:58:19<25:22,  3.50it/s]
 99%|█████████▊| 366138/371472 [6:58:19<26:44,  3.32it/s]
 99%|█████████▊| 366139/371472 [6:58:19<27:41,  3.21it/s]
 99%|█████████▊| 366140/371472 [6:58:20<29:24,  3.02it/s]
                                                         
{'loss': 2.4044, 'learning_rate': 1.129252946626732e-07, 'epoch': 15.77}
+
 99%|█████████▊| 366140/371472 [6:58:20<29:24,  3.02it/s]
 99%|█████████▊| 366141/371472 [6:58:20<27:28,  3.23it/s]
 99%|█████████▊| 366142/371472 [6:58:20<26:22,  3.37it/s]
 99%|█████████▊| 366143/371472 [6:58:21<25:21,  3.50it/s]
 99%|█████████▊| 366144/371472 [6:58:21<24:20,  3.65it/s]
 99%|██████��██▊| 366145/371472 [6:58:21<23:39,  3.75it/s]
 99%|█████████▊| 366146/371472 [6:58:21<23:40,  3.75it/s]
 99%|█████████▊| 366147/371472 [6:58:22<24:28,  3.63it/s]
 99%|█████████▊| 366148/371472 [6:58:22<25:12,  3.52it/s]
 99%|█████████▊| 366149/371472 [6:58:22<24:04,  3.69it/s]
 99%|█████████▊| 366150/371472 [6:58:23<23:32,  3.77it/s]
 99%|█████████▊| 366151/371472 [6:58:23<23:23,  3.79it/s]
 99%|█████████▊| 366152/371472 [6:58:23<22:39,  3.91it/s]
 99%|█████████▊| 366153/371472 [6:58:23<22:58,  3.86it/s]
 99%|█████████▊| 366154/371472 [6:58:24<23:42,  3.74it/s]
 99%|█████████▊| 366155/371472 [6:58:24<23:41,  3.74it/s]
 99%|█████████▊| 366156/371472 [6:58:24<25:02,  3.54it/s]
 99%|█████████▊| 366157/371472 [6:58:24<24:36,  3.60it/s]
 99%|█████████▊| 366158/371472 [6:58:25<24:00,  3.69it/s]
 99%|█████████▊| 366159/371472 [6:58:25<22:58,  3.85it/s]
 99%|█████████▊| 366160/371472 [6:58:25<23:00,  3.85it/s]
                                                         
{'loss': 2.6969, 'learning_rate': 1.1287681268719426e-07, 'epoch': 15.77}
+
 99%|█████████▊| 366160/371472 [6:58:25<23:00,  3.85it/s]
 99%|█████████▊| 366161/371472 [6:58:26<25:10,  3.52it/s]
 99%|█████████▊| 366162/371472 [6:58:26<27:32,  3.21it/s]
 99%|█████████▊| 366163/371472 [6:58:26<27:55,  3.17it/s]
 99%|█████████▊| 366164/371472 [6:58:26<26:42,  3.31it/s]
 99%|█████████▊| 366165/371472 [6:58:27<25:51,  3.42it/s]
 99%|█████████▊| 366166/371472 [6:58:27<26:07,  3.38it/s]
 99%|█████████▊| 366167/371472 [6:58:27<24:57,  3.54it/s]
 99%|█████████▊| 366168/371472 [6:58:28<24:11,  3.65it/s]
 99%|█████████▊| 366169/371472 [6:58:28<25:03,  3.53it/s]
 99%|█████████▊| 366170/371472 [6:58:28<23:51,  3.70it/s]
 99%|█████████▊| 366171/371472 [6:58:28<23:08,  3.82it/s]
 99%|█████████▊| 366172/371472 [6:58:29<23:28,  3.76it/s]
 99%|█████████▊| 366173/371472 [6:58:29<23:13,  3.80it/s]
 99%|█████████▊| 366174/371472 [6:58:29<23:43,  3.72it/s]
 99%|█████████▊| 366175/371472 [6:58:29<22:59,  3.84it/s]
 99%|█████████▊| 366176/371472 [6:58:30<23:13,  3.80it/s]
 99%|█████████▊| 366177/371472 [6:58:30<24:09,  3.65it/s]
 99%|█████████▊| 366178/371472 [6:58:30<24:02,  3.67it/s]
 99%|█████████▊| 366179/371472 [6:58:30<23:12,  3.80it/s]
 99%|█████████▊| 366180/371472 [6:58:31<23:22,  3.77it/s]
                                                         
{'loss': 2.5123, 'learning_rate': 1.1282833071171541e-07, 'epoch': 15.77}
+
 99%|█████████▊| 366180/371472 [6:58:31<23:22,  3.77it/s]
 99%|█████████▊| 366181/371472 [6:58:31<23:16,  3.79it/s]
 99%|█████████▊| 366182/371472 [6:58:31<23:51,  3.70it/s]
 99%|█████████▊| 366183/371472 [6:58:32<23:06,  3.82it/s]
 99%|█████████▊| 366184/371472 [6:58:32<23:54,  3.69it/s]
 99%|█████████▊| 366185/371472 [6:58:32<24:27,  3.60it/s]
 99%|█████████▊| 366186/371472 [6:58:32<23:38,  3.73it/s]
 99%|█████████▊| 366187/371472 [6:58:33<25:08,  3.50it/s]
 99%|█████████▊| 366188/371472 [6:58:33<24:14,  3.63it/s]
 99%|█████████▊| 366189/371472 [6:58:33<23:44,  3.71it/s]
 99%|█████████▊| 366190/371472 [6:58:33<23:29,  3.75it/s]
 99%|█████████▊| 366191/371472 [6:58:34<22:58,  3.83it/s]
 99%|█████████▊| 366192/371472 [6:58:34<26:17,  3.35it/s]
 99%|█████████▊| 366193/371472 [6:58:34<24:52,  3.54it/s]
 99%|█████████▊| 366194/371472 [6:58:35<26:30,  3.32it/s]
 99%|█████████▊| 366195/371472 [6:58:35<24:53,  3.53it/s]
 99%|█████████▊| 366196/371472 [6:58:35<24:38,  3.57it/s]
 99%|█████████▊| 366197/371472 [6:58:36<26:09,  3.36it/s]
 99%|█████████▊| 366198/371472 [6:58:36<28:31,  3.08it/s]
 99%|█████████▊| 366199/371472 [6:58:36<28:36,  3.07it/s]
 99%|█████████▊| 366200/371472 [6:58:37<27:06,  3.24it/s]
                                                         
{'loss': 2.6823, 'learning_rate': 1.1277984873623647e-07, 'epoch': 15.77}
+
 99%|█████████▊| 366200/371472 [6:58:37<27:06,  3.24it/s]
 99%|█████████▊| 366201/371472 [6:58:37<26:03,  3.37it/s]
 99%|█████████▊| 366202/371472 [6:58:37<25:31,  3.44it/s]
 99%|█████████▊| 366203/371472 [6:58:37<24:22,  3.60it/s]
 99%|█████████▊| 366204/371472 [6:58:38<25:14,  3.48it/s]
 99%|█████████▊| 366205/371472 [6:58:38<27:41,  3.17it/s]
 99%|█████████▊| 366206/371472 [6:58:38<27:06,  3.24it/s]
 99%|█████████▊| 366207/371472 [6:58:39<25:11,  3.48it/s]
 99%|█████████▊| 366208/371472 [6:58:39<25:13,  3.48it/s]
 99%|█████████▊| 366209/371472 [6:58:39<25:45,  3.40it/s]
 99%|█████████▊| 366210/371472 [6:58:39<26:05,  3.36it/s]
 99%|█████████▊| 366211/371472 [6:58:40<26:30,  3.31it/s]
 99%|█████████▊| 366212/371472 [6:58:40<26:59,  3.25it/s]
 99%|█████████▊| 366213/371472 [6:58:40<25:39,  3.42it/s]
 99%|█████████▊| 366214/371472 [6:58:41<26:15,  3.34it/s]
 99%|█████████▊| 366215/371472 [6:58:41<24:43,  3.54it/s]
 99%|█████████▊| 366216/371472 [6:58:41<24:43,  3.54it/s]
 99%|█████████▊| 366217/371472 [6:58:41<24:20,  3.60it/s]
 99%|█████████▊| 366218/371472 [6:58:42<24:40,  3.55it/s]
 99%|█████████▊| 366219/371472 [6:58:42<23:28,  3.73it/s]
 99%|█████████▊| 366220/371472 [6:58:42<25:28,  3.44it/s]
                                                         
{'loss': 2.4755, 'learning_rate': 1.1273136676075762e-07, 'epoch': 15.77}
+
 99%|█████████▊| 366220/371472 [6:58:42<25:28,  3.44it/s]
 99%|█████████▊| 366221/371472 [6:58:43<26:04,  3.36it/s]
 99%|█████████▊| 366222/371472 [6:58:43<25:00,  3.50it/s]
 99%|█████████▊| 366223/371472 [6:58:43<23:44,  3.68it/s]
 99%|█████████▊| 366224/371472 [6:58:43<23:06,  3.78it/s]
 99%|█████████▊| 366225/371472 [6:58:44<23:48,  3.67it/s]
 99%|█████████▊| 366226/371472 [6:58:44<26:17,  3.32it/s]
 99%|█████████▊| 366227/371472 [6:58:44<25:01,  3.49it/s]
 99%|█████████▊| 366228/371472 [6:58:45<24:28,  3.57it/s]
 99%|█████████▊| 366229/371472 [6:58:45<24:40,  3.54it/s]
 99%|█████████▊| 366230/371472 [6:58:45<23:48,  3.67it/s]
 99%|█████████▊| 366231/371472 [6:58:45<25:18,  3.45it/s]
 99%|█████████▊| 366232/371472 [6:58:46<24:12,  3.61it/s]
 99%|█████████▊| 366233/371472 [6:58:46<23:45,  3.67it/s]
 99%|█████████▊| 366234/371472 [6:58:46<23:46,  3.67it/s]
 99%|█████████▊| 366235/371472 [6:58:46<23:07,  3.77it/s]
 99%|█████████▊| 366236/371472 [6:58:47<23:59,  3.64it/s]
 99%|█████████▊| 366237/371472 [6:58:47<22:49,  3.82it/s]
 99%|█████████▊| 366238/371472 [6:58:47<22:45,  3.83it/s]
 99%|█████████▊| 366239/371472 [6:58:48<23:08,  3.77it/s]
 99%|█████████▊| 366240/371472 [6:58:48<23:17,  3.74it/s]
                                                         
{'loss': 2.5581, 'learning_rate': 1.1268288478527869e-07, 'epoch': 15.77}
+
 99%|█████████▊| 366240/371472 [6:58:48<23:17,  3.74it/s]
 99%|█████████▊| 366241/371472 [6:58:48<25:12,  3.46it/s]
 99%|█████████▊| 366242/371472 [6:58:48<24:35,  3.54it/s]
 99%|█████████▊| 366243/371472 [6:58:49<24:35,  3.54it/s]
 99%|█████████▊| 366244/371472 [6:58:49<23:48,  3.66it/s]
 99%|█████████▊| 366245/371472 [6:58:49<22:54,  3.80it/s]
 99%|█████████▊| 366246/371472 [6:58:49<24:01,  3.63it/s]
 99%|█████████▊| 366247/371472 [6:58:50<23:21,  3.73it/s]
 99%|█████████▊| 366248/371472 [6:58:50<23:29,  3.71it/s]
 99%|█████████▊| 366249/371472 [6:58:50<24:27,  3.56it/s]
 99%|█████████▊| 366250/371472 [6:58:51<23:50,  3.65it/s]
 99%|█████████▊| 366251/371472 [6:58:51<23:18,  3.73it/s]
 99%|█████████▊| 366252/371472 [6:58:51<24:06,  3.61it/s]
 99%|█████████▊| 366253/371472 [6:58:51<24:34,  3.54it/s]
 99%|█████████▊| 366254/371472 [6:58:52<26:30,  3.28it/s]
 99%|█████████▊| 366255/371472 [6:58:52<25:52,  3.36it/s]
 99%|█████████▊| 366256/371472 [6:58:52<24:34,  3.54it/s]
 99%|█████████▊| 366257/371472 [6:58:53<24:26,  3.56it/s]
 99%|█████████▊| 366258/371472 [6:58:53<24:28,  3.55it/s]
 99%|█████████▊| 366259/371472 [6:58:53<25:24,  3.42it/s]
 99%|█████████▊| 366260/371472 [6:58:54<27:48,  3.12it/s]
                                                         
{'loss': 2.546, 'learning_rate': 1.1263440280979984e-07, 'epoch': 15.78}
+
 99%|██████���██▊| 366260/371472 [6:58:54<27:48,  3.12it/s]
 99%|█████████▊| 366261/371472 [6:58:54<29:08,  2.98it/s]
 99%|█████████▊| 366262/371472 [6:58:54<27:18,  3.18it/s]
 99%|█████████▊| 366263/371472 [6:58:54<26:35,  3.26it/s]
 99%|█████████▊| 366264/371472 [6:58:55<25:23,  3.42it/s]
 99%|█████████▊| 366265/371472 [6:58:55<25:29,  3.40it/s]
 99%|█████████▊| 366266/371472 [6:58:55<24:28,  3.55it/s]
 99%|█████████▊| 366267/371472 [6:58:56<23:49,  3.64it/s]
 99%|█████████▊| 366268/371472 [6:58:56<22:46,  3.81it/s]
 99%|█████████▊| 366269/371472 [6:58:56<24:06,  3.60it/s]
 99%|█████████▊| 366270/371472 [6:58:56<23:33,  3.68it/s]
 99%|█████████▊| 366271/371472 [6:58:57<25:04,  3.46it/s]
 99%|█████████▊| 366272/371472 [6:58:57<24:38,  3.52it/s]
 99%|█████████▊| 366273/371472 [6:58:57<25:26,  3.40it/s]
 99%|█████████▊| 366274/371472 [6:58:58<25:14,  3.43it/s]
 99%|█████████▊| 366275/371472 [6:58:58<25:04,  3.45it/s]
 99%|█████████▊| 366276/371472 [6:58:58<23:58,  3.61it/s]
 99%|█████████▊| 366277/371472 [6:58:58<23:31,  3.68it/s]
 99%|█████████▊| 366278/371472 [6:58:59<23:09,  3.74it/s]
 99%|█████████▊| 366279/371472 [6:58:59<23:28,  3.69it/s]
 99%|█████████▊| 366280/371472 [6:58:59<23:32,  3.67it/s]
                                                         
{'loss': 2.7965, 'learning_rate': 1.125859208343209e-07, 'epoch': 15.78}
+
 99%|█████████▊| 366280/371472 [6:58:59<23:32,  3.67it/s]
 99%|█████████▊| 366281/371472 [6:58:59<23:31,  3.68it/s]
 99%|█████████▊| 366282/371472 [6:59:00<24:02,  3.60it/s]
 99%|█████████▊| 366283/371472 [6:59:00<24:02,  3.60it/s]
 99%|█████████▊| 366284/371472 [6:59:00<23:51,  3.62it/s]
 99%|█████████▊| 366285/371472 [6:59:01<28:01,  3.08it/s]
 99%|█████████▊| 366286/371472 [6:59:01<26:20,  3.28it/s]
 99%|█████████▊| 366287/371472 [6:59:01<26:29,  3.26it/s]
 99%|█████████▊| 366288/371472 [6:59:02<26:12,  3.30it/s]
 99%|█████████▊| 366289/371472 [6:59:02<26:36,  3.25it/s]
 99%|█████████▊| 366290/371472 [6:59:02<26:10,  3.30it/s]
 99%|█████████▊| 366291/371472 [6:59:03<27:06,  3.19it/s]
 99%|█████████▊| 366292/371472 [6:59:03<26:19,  3.28it/s]
 99%|█████████▊| 366293/371472 [6:59:03<24:53,  3.47it/s]
 99%|█████████▊| 366294/371472 [6:59:03<24:36,  3.51it/s]
 99%|█████████▊| 366295/371472 [6:59:04<25:55,  3.33it/s]
 99%|█████████▊| 366296/371472 [6:59:04<25:24,  3.39it/s]
 99%|█████████▊| 366297/371472 [6:59:04<24:46,  3.48it/s]
 99%|█████████▊| 366298/371472 [6:59:05<25:12,  3.42it/s]
 99%|█████████▊| 366299/371472 [6:59:05<24:41,  3.49it/s]
 99%|█████████▊| 366300/371472 [6:59:05<24:13,  3.56it/s]
                                                         
{'loss': 2.5936, 'learning_rate': 1.1253743885884207e-07, 'epoch': 15.78}
+
 99%|█████████▊| 366300/371472 [6:59:05<24:13,  3.56it/s]
 99%|█████████▊| 366301/371472 [6:59:05<23:53,  3.61it/s]
 99%|█████████▊| 366302/371472 [6:59:06<23:41,  3.64it/s]
 99%|█████████▊| 366303/371472 [6:59:06<24:35,  3.50it/s]
 99%|█████████▊| 366304/371472 [6:59:06<24:19,  3.54it/s]
 99%|█████████▊| 366305/371472 [6:59:06<23:23,  3.68it/s]
 99%|█████████▊| 366306/371472 [6:59:07<24:29,  3.51it/s]
 99%|█████████▊| 366307/371472 [6:59:07<24:17,  3.54it/s]
 99%|█████████▊| 366308/371472 [6:59:07<24:24,  3.53it/s]
 99%|█████████▊| 366309/371472 [6:59:08<25:46,  3.34it/s]
 99%|█████████▊| 366310/371472 [6:59:08<26:16,  3.27it/s]
 99%|█████████▊| 366311/371472 [6:59:08<25:40,  3.35it/s]
 99%|█████████▊| 366312/371472 [6:59:09<26:00,  3.31it/s]
 99%|█████████▊| 366313/371472 [6:59:09<24:57,  3.45it/s]
 99%|█████████▊| 366314/371472 [6:59:09<24:27,  3.51it/s]
 99%|█████████▊| 366315/371472 [6:59:09<24:49,  3.46it/s]
 99%|█████████▊| 366316/371472 [6:59:10<24:01,  3.58it/s]
 99%|█████████▊| 366317/371472 [6:59:10<22:55,  3.75it/s]
 99%|█████████▊| 366318/371472 [6:59:10<23:53,  3.60it/s]
 99%|█████████▊| 366319/371472 [6:59:10<23:17,  3.69it/s]
 99%|█████████▊| 366320/371472 [6:59:11<25:32,  3.36it/s]
                                                         
{'loss': 2.6067, 'learning_rate': 1.1248895688336311e-07, 'epoch': 15.78}
+
 99%|█████████▊| 366320/371472 [6:59:11<25:32,  3.36it/s]
 99%|█████████▊| 366321/371472 [6:59:11<26:47,  3.21it/s]
 99%|█████████▊| 366322/371472 [6:59:11<25:42,  3.34it/s]
 99%|█████████▊| 366323/371472 [6:59:12<25:23,  3.38it/s]
 99%|█████████▊| 366324/371472 [6:59:12<24:37,  3.48it/s]
 99%|█████████▊| 366325/371472 [6:59:12<23:42,  3.62it/s]
 99%|█████████▊| 366326/371472 [6:59:12<23:23,  3.67it/s]
 99%|█████████▊| 366327/371472 [6:59:13<24:02,  3.57it/s]
 99%|█████████▊| 366328/371472 [6:59:13<23:29,  3.65it/s]
 99%|█████████▊| 366329/371472 [6:59:13<23:15,  3.68it/s]
 99%|█████████▊| 366330/371472 [6:59:14<22:47,  3.76it/s]
 99%|█████████▊| 366331/371472 [6:59:14<24:39,  3.48it/s]
 99%|█████████▊| 366332/371472 [6:59:14<24:25,  3.51it/s]
 99%|█████████▊| 366333/371472 [6:59:14<24:34,  3.49it/s]
 99%|█████████▊| 366334/371472 [6:59:15<23:52,  3.59it/s]
 99%|█████████▊| 366335/371472 [6:59:15<23:01,  3.72it/s]
 99%|█████████▊| 366336/371472 [6:59:15<22:55,  3.73it/s]
 99%|█████████▊| 366337/371472 [6:59:16<24:16,  3.52it/s]
 99%|█████████▊| 366338/371472 [6:59:16<23:10,  3.69it/s]
 99%|█████████▊| 366339/371472 [6:59:16<23:34,  3.63it/s]
 99%|█████████▊| 366340/371472 [6:59:16<24:41,  3.46it/s]
                                                         
{'loss': 2.7544, 'learning_rate': 1.1244047490788428e-07, 'epoch': 15.78}
+
 99%|█████████▊| 366340/371472 [6:59:16<24:41,  3.46it/s]
 99%|█████████▊| 366341/371472 [6:59:17<25:50,  3.31it/s]
 99%|█████████▊| 366342/371472 [6:59:17<25:13,  3.39it/s]
 99%|█████████▊| 366343/371472 [6:59:17<24:59,  3.42it/s]
 99%|█████████▊| 366344/371472 [6:59:18<25:04,  3.41it/s]
 99%|█████████▊| 366345/371472 [6:59:18<24:00,  3.56it/s]
 99%|█████████▊| 366346/371472 [6:59:18<23:13,  3.68it/s]
 99%|█████████▊| 366347/371472 [6:59:18<24:34,  3.48it/s]
 99%|█████████▊| 366348/371472 [6:59:19<24:24,  3.50it/s]
 99%|█████████▊| 366349/371472 [6:59:19<24:46,  3.45it/s]
 99%|█████████▊| 366350/371472 [6:59:19<23:59,  3.56it/s]
 99%|█████████▊| 366351/371472 [6:59:20<23:33,  3.62it/s]
 99%|█████████▊| 366352/371472 [6:59:20<24:15,  3.52it/s]
 99%|█████████▊| 366353/371472 [6:59:20<24:35,  3.47it/s]
 99%|█████████▊| 366354/371472 [6:59:20<24:16,  3.51it/s]
 99%|█████████▊| 366355/371472 [6:59:21<24:06,  3.54it/s]
 99%|█████████▊| 366356/371472 [6:59:21<24:07,  3.53it/s]
 99%|█████████▊| 366357/371472 [6:59:21<23:44,  3.59it/s]
 99%|█████████▊| 366358/371472 [6:59:22<24:14,  3.52it/s]
 99%|█████████▊| 366359/371472 [6:59:22<24:10,  3.52it/s]
 99%|█████████▊| 366360/371472 [6:59:22<24:12,  3.52it/s]
                                                         
{'loss': 2.5254, 'learning_rate': 1.1239199293240533e-07, 'epoch': 15.78}
+
 99%|█████████▊| 366360/371472 [6:59:22<24:12,  3.52it/s]
 99%|█████████▊| 366361/371472 [6:59:22<23:39,  3.60it/s]
 99%|█████████▊| 366362/371472 [6:59:23<23:39,  3.60it/s]
 99%|█████████▊| 366363/371472 [6:59:23<23:12,  3.67it/s]
 99%|█████████▊| 366364/371472 [6:59:23<23:58,  3.55it/s]
 99%|█████████▊| 366365/371472 [6:59:23<22:40,  3.75it/s]
 99%|█████████▊| 366366/371472 [6:59:24<22:56,  3.71it/s]
 99%|█████████▊| 366367/371472 [6:59:24<22:35,  3.77it/s]
 99%|█████████▊| 366368/371472 [6:59:24<22:34,  3.77it/s]
 99%|█████████▊| 366369/371472 [6:59:25<22:41,  3.75it/s]
 99%|█████████▊| 366370/371472 [6:59:25<22:56,  3.71it/s]
 99%|█████████▊| 366371/371472 [6:59:25<22:55,  3.71it/s]
 99%|█████████▊| 366372/371472 [6:59:25<23:17,  3.65it/s]
 99%|█████████▊| 366373/371472 [6:59:26<22:21,  3.80it/s]
 99%|█████████▊| 366374/371472 [6:59:26<22:54,  3.71it/s]
 99%|█████████▊| 366375/371472 [6:59:26<24:01,  3.54it/s]
 99%|█████████▊| 366376/371472 [6:59:27<24:59,  3.40it/s]
 99%|█████████▊| 366377/371472 [6:59:27<27:21,  3.10it/s]
 99%|█████████▊| 366378/371472 [6:59:27<26:09,  3.25it/s]
 99%|█████████▊| 366379/371472 [6:59:27<25:53,  3.28it/s]
 99%|█████████▊| 366380/371472 [6:59:28<26:02,  3.26it/s]
                                                         
{'loss': 2.6236, 'learning_rate': 1.1234351095692648e-07, 'epoch': 15.78}
+
 99%|█████████▊| 366380/371472 [6:59:28<26:02,  3.26it/s]
 99%|█████████▊| 366381/371472 [6:59:28<25:35,  3.31it/s]
 99%|█████████▊| 366382/371472 [6:59:28<24:36,  3.45it/s]
 99%|█████████▊| 366383/371472 [6:59:29<25:21,  3.34it/s]
 99%|█████████▊| 366384/371472 [6:59:29<24:12,  3.50it/s]
 99%|█████████▊| 366385/371472 [6:59:29<23:28,  3.61it/s]
 99%|█████████▊| 366386/371472 [6:59:29<23:21,  3.63it/s]
 99%|█████████▊| 366387/371472 [6:59:30<22:33,  3.76it/s]
 99%|█████████▊| 366388/371472 [6:59:30<25:11,  3.36it/s]
 99%|█████████▊| 366389/371472 [6:59:30<24:16,  3.49it/s]
 99%|█████████▊| 366390/371472 [6:59:31<23:12,  3.65it/s]
 99%|█████████▊| 366391/371472 [6:59:31<22:38,  3.74it/s]
 99%|█████████▊| 366392/371472 [6:59:31<23:17,  3.64it/s]
 99%|█████████▊| 366393/371472 [6:59:31<24:32,  3.45it/s]
 99%|█████████▊| 366394/371472 [6:59:32<23:48,  3.55it/s]
 99%|█████████▊| 366395/371472 [6:59:32<23:00,  3.68it/s]
 99%|█████████▊| 366396/371472 [6:59:32<22:49,  3.71it/s]
 99%|█████████▊| 366397/371472 [6:59:32<22:33,  3.75it/s]
 99%|█████████▊| 366398/371472 [6:59:33<21:36,  3.91it/s]
 99%|█████████▊| 366399/371472 [6:59:33<21:45,  3.89it/s]
 99%|█████████▊| 366400/371472 [6:59:33<22:43,  3.72it/s]
                                                         
{'loss': 2.5303, 'learning_rate': 1.1229502898144754e-07, 'epoch': 15.78}
+
 99%|█████████▊| 366400/371472 [6:59:33<22:43,  3.72it/s]
 99%|█████████▊| 366401/371472 [6:59:33<22:01,  3.84it/s]
 99%|█████████▊| 366402/371472 [6:59:34<21:39,  3.90it/s]
 99%|█████████▊| 366403/371472 [6:59:34<21:27,  3.94it/s]
 99%|█████████▊| 366404/371472 [6:59:34<21:53,  3.86it/s]
 99%|█████████▊| 366405/371472 [6:59:35<21:48,  3.87it/s]
 99%|█████████▊| 366406/371472 [6:59:35<21:56,  3.85it/s]
 99%|█████████▊| 366407/371472 [6:59:35<22:41,  3.72it/s]
 99%|█████████▊| 366408/371472 [6:59:35<22:26,  3.76it/s]
 99%|█████████▊| 366409/371472 [6:59:36<22:04,  3.82it/s]
 99%|█████████▊| 366410/371472 [6:59:36<21:51,  3.86it/s]
 99%|█████████▊| 366411/371472 [6:59:36<21:59,  3.83it/s]
 99%|█████████▊| 366412/371472 [6:59:36<22:39,  3.72it/s]
 99%|█████████▊| 366413/371472 [6:59:37<22:39,  3.72it/s]
 99%|█████████▊| 366414/371472 [6:59:37<22:33,  3.74it/s]
 99%|█████████▊| 366415/371472 [6:59:37<22:38,  3.72it/s]
 99%|█████████▊| 366416/371472 [6:59:37<23:18,  3.62it/s]
 99%|█████████▊| 366417/371472 [6:59:38<23:48,  3.54it/s]
 99%|█████████▊| 366418/371472 [6:59:38<24:29,  3.44it/s]
 99%|█████████▊| 366419/371472 [6:59:38<24:20,  3.46it/s]
 99%|█████████▊| 366420/371472 [6:59:39<25:22,  3.32it/s]
                                                         
{'loss': 2.5965, 'learning_rate': 1.1224654700596871e-07, 'epoch': 15.78}
+
 99%|█████████▊| 366420/371472 [6:59:39<25:22,  3.32it/s]
 99%|█████████▊| 366421/371472 [6:59:39<25:14,  3.34it/s]
 99%|█████████▊| 366422/371472 [6:59:39<25:35,  3.29it/s]
 99%|█████████▊| 366423/371472 [6:59:40<24:42,  3.41it/s]
 99%|█████████▊| 366424/371472 [6:59:40<24:11,  3.48it/s]
 99%|█████████▊| 366425/371472 [6:59:40<23:53,  3.52it/s]
 99%|█████████▊| 366426/371472 [6:59:40<22:48,  3.69it/s]
 99%|█████████▊| 366427/371472 [6:59:41<22:48,  3.69it/s]
 99%|█████████▊| 366428/371472 [6:59:41<23:19,  3.61it/s]
 99%|█████████▊| 366429/371472 [6:59:41<23:08,  3.63it/s]
 99%|█████████▊| 366430/371472 [6:59:42<24:21,  3.45it/s]
 99%|█████████▊| 366431/371472 [6:59:42<23:23,  3.59it/s]
 99%|█████████▊| 366432/371472 [6:59:42<22:32,  3.73it/s]
 99%|█████████▊| 366433/371472 [6:59:42<22:04,  3.80it/s]
 99%|█████████▊| 366434/371472 [6:59:43<22:38,  3.71it/s]
 99%|█████████▊| 366435/371472 [6:59:43<22:34,  3.72it/s]
 99%|█████████▊| 366436/371472 [6:59:43<22:24,  3.75it/s]
 99%|█████████▊| 366437/371472 [6:59:43<22:01,  3.81it/s]
 99%|█████████▊| 366438/371472 [6:59:44<22:08,  3.79it/s]
 99%|█████████▊| 366439/371472 [6:59:44<21:53,  3.83it/s]
 99%|█████████▊| 366440/371472 [6:59:44<24:11,  3.47it/s]
                                                         
{'loss': 2.6287, 'learning_rate': 1.1219806503048977e-07, 'epoch': 15.78}
+
 99%|█████████▊| 366440/371472 [6:59:44<24:11,  3.47it/s]
 99%|█████████▊| 366441/371472 [6:59:44<22:55,  3.66it/s]
 99%|█████████▊| 366442/371472 [6:59:45<22:31,  3.72it/s]
 99%|█████████▊| 366443/371472 [6:59:45<23:03,  3.63it/s]
 99%|█████████▊| 366444/371472 [6:59:45<22:47,  3.68it/s]
 99%|█████████▊| 366445/371472 [6:59:46<22:46,  3.68it/s]
 99%|█████████▊| 366446/371472 [6:59:46<22:54,  3.66it/s]
 99%|█████████▊| 366447/371472 [6:59:46<23:40,  3.54it/s]
 99%|█████████▊| 366448/371472 [6:59:46<23:05,  3.62it/s]
 99%|█████████▊| 366449/371472 [6:59:47<24:13,  3.46it/s]
 99%|█████████▊| 366450/371472 [6:59:47<23:26,  3.57it/s]
 99%|█████████▊| 366451/371472 [6:59:47<23:01,  3.63it/s]
 99%|█████████▊| 366452/371472 [6:59:48<24:33,  3.41it/s]
 99%|█████████▊| 366453/371472 [6:59:48<23:33,  3.55it/s]
 99%|█████████▊| 366454/371472 [6:59:48<22:44,  3.68it/s]
 99%|█████████▊| 366455/371472 [6:59:48<22:53,  3.65it/s]
 99%|█████████▊| 366456/371472 [6:59:49<22:45,  3.67it/s]
 99%|█████████▊| 366457/371472 [6:59:49<22:26,  3.72it/s]
 99%|█████████▊| 366458/371472 [6:59:49<23:10,  3.61it/s]
 99%|█████████▊| 366459/371472 [6:59:49<22:26,  3.72it/s]
 99%|█████████▊| 366460/371472 [6:59:50<22:48,  3.66it/s]
                                                         
{'loss': 2.7425, 'learning_rate': 1.1214958305501092e-07, 'epoch': 15.78}
+
 99%|█████████▊| 366460/371472 [6:59:50<22:48,  3.66it/s]
 99%|█████████▊| 366461/371472 [6:59:50<22:52,  3.65it/s]
 99%|█████████▊| 366462/371472 [6:59:50<23:00,  3.63it/s]
 99%|█████████▊| 366463/371472 [6:59:51<24:36,  3.39it/s]
 99%|█████████▊| 366464/371472 [6:59:51<24:41,  3.38it/s]
 99%|█████████▊| 366465/371472 [6:59:51<23:31,  3.55it/s]
 99%|█████████▊| 366466/371472 [6:59:51<22:34,  3.69it/s]
 99%|█████████▊| 366467/371472 [6:59:52<22:29,  3.71it/s]
 99%|█████████▊| 366468/371472 [6:59:52<22:51,  3.65it/s]
 99%|█████████▊| 366469/371472 [6:59:52<23:06,  3.61it/s]
 99%|█████████▊| 366470/371472 [6:59:52<22:31,  3.70it/s]
 99%|█████████▊| 366471/371472 [6:59:53<22:32,  3.70it/s]
 99%|█████████▊| 366472/371472 [6:59:53<23:13,  3.59it/s]
 99%|█████████▊| 366473/371472 [6:59:53<22:37,  3.68it/s]
 99%|█████████▊| 366474/371472 [6:59:54<22:40,  3.67it/s]
 99%|█████████▊| 366475/371472 [6:59:54<22:20,  3.73it/s]
 99%|█████████▊| 366476/371472 [6:59:54<24:13,  3.44it/s]
 99%|█████████▊| 366477/371472 [6:59:54<22:57,  3.63it/s]
 99%|█████████▊| 366478/371472 [6:59:55<24:43,  3.37it/s]
 99%|█████████▊| 366479/371472 [6:59:55<23:59,  3.47it/s]
 99%|█████████▊| 366480/371472 [6:59:55<23:32,  3.54it/s]
                                                         
{'loss': 2.7305, 'learning_rate': 1.1210110107953199e-07, 'epoch': 15.78}
+
 99%|█████████▊| 366480/371472 [6:59:55<23:32,  3.54it/s]
 99%|█████████▊| 366481/371472 [6:59:56<23:01,  3.61it/s]
 99%|█████████▊| 366482/371472 [6:59:56<23:13,  3.58it/s]
 99%|█████████▊| 366483/371472 [6:59:56<23:20,  3.56it/s]
 99%|█████████▊| 366484/371472 [6:59:56<25:15,  3.29it/s]
 99%|█████████▊| 366485/371472 [6:59:57<25:09,  3.30it/s]
 99%|█████████▊| 366486/371472 [6:59:57<23:58,  3.47it/s]
 99%|█████████▊| 366487/371472 [6:59:57<23:23,  3.55it/s]
 99%|█████████▊| 366488/371472 [6:59:58<23:25,  3.55it/s]
 99%|█████████▊| 366489/371472 [6:59:58<23:02,  3.60it/s]
 99%|█████████▊| 366490/371472 [6:59:58<22:25,  3.70it/s]
 99%|█████████▊| 366491/371472 [6:59:58<21:52,  3.79it/s]
 99%|█████████▊| 366492/371472 [6:59:59<21:30,  3.86it/s]
 99%|█████████▊| 366493/371472 [6:59:59<22:40,  3.66it/s]
 99%|█████████▊| 366494/371472 [6:59:59<22:13,  3.73it/s]
 99%|█████████▊| 366495/371472 [6:59:59<23:09,  3.58it/s]
 99%|█████████▊| 366496/371472 [7:00:00<22:44,  3.65it/s]
 99%|█████████▊| 366497/371472 [7:00:00<22:20,  3.71it/s]
 99%|█████████▊| 366498/371472 [7:00:00<22:59,  3.61it/s]
 99%|█████████▊| 366499/371472 [7:00:01<22:31,  3.68it/s]
 99%|█████████▊| 366500/371472 [7:00:01<23:05,  3.59it/s]
                                                         
{'loss': 2.6572, 'learning_rate': 1.1205261910405314e-07, 'epoch': 15.79}
+
 99%|█████████▊| 366500/371472 [7:00:01<23:05,  3.59it/s]
 99%|█████████▊| 366501/371472 [7:00:01<22:37,  3.66it/s]
 99%|█████████▊| 366502/371472 [7:00:01<22:44,  3.64it/s]
 99%|█████████▊| 366503/371472 [7:00:02<22:10,  3.73it/s]
 99%|█████████▊| 366504/371472 [7:00:02<23:15,  3.56it/s]
 99%|█████████▊| 366505/371472 [7:00:02<23:06,  3.58it/s]
 99%|█████████▊| 366506/371472 [7:00:03<24:08,  3.43it/s]
 99%|█████████▊| 366507/371472 [7:00:03<24:27,  3.38it/s]
 99%|█████████▊| 366508/371472 [7:00:03<23:49,  3.47it/s]
 99%|█████████▊| 366509/371472 [7:00:03<23:08,  3.57it/s]
 99%|█████████▊| 366510/371472 [7:00:04<23:33,  3.51it/s]
 99%|█████████▊| 366511/371472 [7:00:04<22:56,  3.60it/s]
 99%|█████████▊| 366512/371472 [7:00:04<22:10,  3.73it/s]
 99%|█████████▊| 366513/371472 [7:00:04<22:03,  3.75it/s]
 99%|█████████▊| 366514/371472 [7:00:05<21:54,  3.77it/s]
 99%|█████████▊| 366515/371472 [7:00:05<23:06,  3.58it/s]
 99%|█████████▊| 366516/371472 [7:00:05<22:23,  3.69it/s]
 99%|█████████▊| 366517/371472 [7:00:06<23:12,  3.56it/s]
 99%|█████████▊| 366518/371472 [7:00:06<22:30,  3.67it/s]
 99%|█████████▊| 366519/371472 [7:00:06<24:53,  3.32it/s]
 99%|█████████▊| 366520/371472 [7:00:06<23:30,  3.51it/s]
                                                         
{'loss': 2.6309, 'learning_rate': 1.120041371285742e-07, 'epoch': 15.79}
+
 99%|█████████▊| 366520/371472 [7:00:06<23:30,  3.51it/s]
 99%|█████████▊| 366521/371472 [7:00:07<22:52,  3.61it/s]
 99%|█████████▊| 366522/371472 [7:00:07<22:23,  3.69it/s]
 99%|█████████▊| 366523/371472 [7:00:07<21:55,  3.76it/s]
 99%|█████████▊| 366524/371472 [7:00:07<21:49,  3.78it/s]
 99%|█████████▊| 366525/371472 [7:00:08<22:20,  3.69it/s]
 99%|█████████▊| 366526/371472 [7:00:08<21:33,  3.82it/s]
 99%|█████████▊| 366527/371472 [7:00:08<21:50,  3.77it/s]
 99%|█████████▊| 366528/371472 [7:00:09<23:19,  3.53it/s]
 99%|█████████▊| 366529/371472 [7:00:09<22:51,  3.60it/s]
 99%|█████████▊| 366530/371472 [7:00:09<21:59,  3.74it/s]
 99%|█████████▊| 366531/371472 [7:00:09<22:00,  3.74it/s]
 99%|█████████▊| 366532/371472 [7:00:10<23:38,  3.48it/s]
 99%|█████████▊| 366533/371472 [7:00:10<23:35,  3.49it/s]
 99%|█████████▊| 366534/371472 [7:00:10<24:31,  3.36it/s]
 99%|█████████▊| 366535/371472 [7:00:11<24:10,  3.40it/s]
 99%|█████████▊| 366536/371472 [7:00:11<23:43,  3.47it/s]
 99%|█████████▊| 366537/371472 [7:00:11<22:46,  3.61it/s]
 99%|█████████▊| 366538/371472 [7:00:11<23:07,  3.56it/s]
 99%|█████████▊| 366539/371472 [7:00:12<24:50,  3.31it/s]
 99%|█████████▊| 366540/371472 [7:00:12<24:00,  3.42it/s]
                                                         
{'loss': 2.5205, 'learning_rate': 1.1195565515309526e-07, 'epoch': 15.79}
+
 99%|█████████▊| 366540/371472 [7:00:12<24:00,  3.42it/s]
 99%|█████████▊| 366541/371472 [7:00:12<27:57,  2.94it/s]
 99%|█████████▊| 366542/371472 [7:00:13<26:01,  3.16it/s]
 99%|█████████▊| 366543/371472 [7:00:13<24:23,  3.37it/s]
 99%|█████████▊| 366544/371472 [7:00:13<23:50,  3.45it/s]
 99%|█████████▊| 366545/371472 [7:00:14<23:32,  3.49it/s]
 99%|█████████▊| 366546/371472 [7:00:14<23:30,  3.49it/s]
 99%|█████████▊| 366547/371472 [7:00:14<23:24,  3.51it/s]
 99%|█████████▊| 366548/371472 [7:00:14<23:05,  3.55it/s]
 99%|█████████▊| 366549/371472 [7:00:15<23:09,  3.54it/s]
 99%|█████████▊| 366550/371472 [7:00:15<23:09,  3.54it/s]
 99%|█████████▊| 366551/371472 [7:00:15<22:44,  3.61it/s]
 99%|█████████▊| 366552/371472 [7:00:16<23:00,  3.56it/s]
 99%|█████████▊| 366553/371472 [7:00:16<24:59,  3.28it/s]
 99%|█████████▊| 366554/371472 [7:00:16<25:05,  3.27it/s]
 99%|█████████▊| 366555/371472 [7:00:16<23:30,  3.49it/s]
 99%|█████████▊| 366556/371472 [7:00:17<22:58,  3.57it/s]
 99%|█████████▊| 366557/371472 [7:00:17<22:35,  3.62it/s]
 99%|█████████▊| 366558/371472 [7:00:17<24:15,  3.38it/s]
 99%|█████████▊| 366559/371472 [7:00:18<22:41,  3.61it/s]
 99%|█████████▊| 366560/371472 [7:00:18<22:17,  3.67it/s]
                                                         
{'loss': 2.6014, 'learning_rate': 1.119071731776164e-07, 'epoch': 15.79}
+
 99%|█████████▊| 366560/371472 [7:00:18<22:17,  3.67it/s]
 99%|█████████▊| 366561/371472 [7:00:18<23:59,  3.41it/s]
 99%|█████████▊| 366562/371472 [7:00:18<23:05,  3.54it/s]
 99%|█████████▊| 366563/371472 [7:00:19<24:55,  3.28it/s]
 99%|█████████▊| 366564/371472 [7:00:19<25:23,  3.22it/s]
 99%|█████████▊| 366565/371472 [7:00:19<24:45,  3.30it/s]
 99%|█████████▊| 366566/371472 [7:00:20<23:21,  3.50it/s]
 99%|█████████▊| 366567/371472 [7:00:20<23:59,  3.41it/s]
 99%|█████████▊| 366568/371472 [7:00:20<23:35,  3.46it/s]
 99%|█████████▊| 366569/371472 [7:00:20<23:14,  3.52it/s]
 99%|█████████▊| 366570/371472 [7:00:21<24:44,  3.30it/s]
 99%|█████████▊| 366571/371472 [7:00:21<23:51,  3.42it/s]
 99%|█████████▊| 366572/371472 [7:00:21<22:52,  3.57it/s]
 99%|█████████▊| 366573/371472 [7:00:22<23:51,  3.42it/s]
 99%|█████████▊| 366574/371472 [7:00:22<24:09,  3.38it/s]
 99%|█████████▊| 366575/371472 [7:00:22<23:25,  3.48it/s]
 99%|█████████▊| 366576/371472 [7:00:22<22:57,  3.55it/s]
 99%|█████████▊| 366577/371472 [7:00:23<22:27,  3.63it/s]
 99%|█████████▊| 366578/371472 [7:00:23<23:56,  3.41it/s]
 99%|█████████▊| 366579/371472 [7:00:23<23:12,  3.51it/s]
 99%|█████████▊| 366580/371472 [7:00:24<22:42,  3.59it/s]
                                                         
{'loss': 2.6644, 'learning_rate': 1.1185869120213746e-07, 'epoch': 15.79}
+
 99%|█████████▊| 366580/371472 [7:00:24<22:42,  3.59it/s]
 99%|█████████▊| 366581/371472 [7:00:24<22:12,  3.67it/s]
 99%|█████████▊| 366582/371472 [7:00:24<22:40,  3.60it/s]
 99%|█████████▊| 366583/371472 [7:00:24<21:58,  3.71it/s]
 99%|█████████▊| 366584/371472 [7:00:25<22:08,  3.68it/s]
 99%|█████████▊| 366585/371472 [7:00:25<22:39,  3.59it/s]
 99%|█████████▊| 366586/371472 [7:00:25<22:09,  3.68it/s]
 99%|█████████▊| 366587/371472 [7:00:26<22:43,  3.58it/s]
 99%|█████████▊| 366588/371472 [7:00:26<22:45,  3.58it/s]
 99%|█████████▊| 366589/371472 [7:00:26<22:28,  3.62it/s]
 99%|█████████▊| 366590/371472 [7:00:26<23:10,  3.51it/s]
 99%|█████████▊| 366591/371472 [7:00:27<23:19,  3.49it/s]
 99%|█████████▊| 366592/371472 [7:00:27<23:30,  3.46it/s]
 99%|█████████▊| 366593/371472 [7:00:27<23:28,  3.46it/s]
 99%|█████████▊| 366594/371472 [7:00:28<22:39,  3.59it/s]
 99%|█████████▊| 366595/371472 [7:00:28<22:10,  3.67it/s]
 99%|█████████▊| 366596/371472 [7:00:28<22:42,  3.58it/s]
 99%|█████████▊| 366597/371472 [7:00:28<23:41,  3.43it/s]
 99%|█████████▊| 366598/371472 [7:00:29<22:53,  3.55it/s]
 99%|█████████▊| 366599/371472 [7:00:29<22:32,  3.60it/s]
 99%|█████████▊| 366600/371472 [7:00:29<21:50,  3.72it/s]
                                                         
{'loss': 2.4946, 'learning_rate': 1.1181020922665863e-07, 'epoch': 15.79}
+
 99%|█████████▊| 366600/371472 [7:00:29<21:50,  3.72it/s]
 99%|█████████▊| 366601/371472 [7:00:30<25:41,  3.16it/s]
 99%|█████████▊| 366602/371472 [7:00:30<25:09,  3.23it/s]
 99%|█████████▊| 366603/371472 [7:00:30<23:59,  3.38it/s]
 99%|█████████▊| 366604/371472 [7:00:30<22:55,  3.54it/s]
 99%|█████████▊| 366605/371472 [7:00:31<22:12,  3.65it/s]
 99%|█████████▊| 366606/371472 [7:00:31<22:34,  3.59it/s]
 99%|█████████▊| 366607/371472 [7:00:31<22:23,  3.62it/s]
 99%|█████████▊| 366608/371472 [7:00:31<22:05,  3.67it/s]
 99%|█████████▊| 366609/371472 [7:00:32<21:31,  3.77it/s]
 99%|█████████▊| 366610/371472 [7:00:32<22:40,  3.57it/s]
 99%|█████████▊| 366611/371472 [7:00:32<21:48,  3.72it/s]
 99%|█████████▊| 366612/371472 [7:00:33<21:16,  3.81it/s]
 99%|█████████▊| 366613/371472 [7:00:33<21:35,  3.75it/s]
 99%|█████████▊| 366614/371472 [7:00:33<21:48,  3.71it/s]
 99%|█████████▊| 366615/371472 [7:00:33<21:37,  3.74it/s]
 99%|█████████▊| 366616/371472 [7:00:34<20:51,  3.88it/s]
 99%|█████████▊| 366617/371472 [7:00:34<21:16,  3.80it/s]
 99%|█████████▊| 366618/371472 [7:00:34<21:01,  3.85it/s]
 99%|█████████▊| 366619/371472 [7:00:34<20:48,  3.89it/s]
 99%|█████████▊| 366620/371472 [7:00:35<21:55,  3.69it/s]
                                                         
{'loss': 2.5698, 'learning_rate': 1.1176172725117969e-07, 'epoch': 15.79}
+
 99%|█████████▊| 366620/371472 [7:00:35<21:55,  3.69it/s]
 99%|█████████▊| 366621/371472 [7:00:35<21:23,  3.78it/s]
 99%|█████████▊| 366622/371472 [7:00:35<20:57,  3.86it/s]
 99%|█████████▊| 366623/371472 [7:00:35<20:32,  3.93it/s]
 99%|█████████▊| 366624/371472 [7:00:36<20:26,  3.95it/s]
 99%|█████████▊| 366625/371472 [7:00:36<22:11,  3.64it/s]
 99%|█████████▊| 366626/371472 [7:00:36<22:13,  3.63it/s]
 99%|█████████▊| 366627/371472 [7:00:37<22:22,  3.61it/s]
 99%|█████████▊| 366628/371472 [7:00:37<21:38,  3.73it/s]
 99%|█████████▊| 366629/371472 [7:00:37<22:59,  3.51it/s]
 99%|█████████▊| 366630/371472 [7:00:37<22:52,  3.53it/s]
 99%|█████████▊| 366631/371472 [7:00:38<22:12,  3.63it/s]
 99%|█████████▊| 366632/371472 [7:00:38<22:12,  3.63it/s]
 99%|█████████▊| 366633/371472 [7:00:38<22:03,  3.66it/s]
 99%|█████████▊| 366634/371472 [7:00:39<22:43,  3.55it/s]
 99%|█████████▊| 366635/371472 [7:00:39<22:35,  3.57it/s]
 99%|█████████▊| 366636/371472 [7:00:39<21:53,  3.68it/s]
 99%|█████████▊| 366637/371472 [7:00:39<22:20,  3.61it/s]
 99%|█████████▊| 366638/371472 [7:00:40<21:21,  3.77it/s]
 99%|█████████▊| 366639/371472 [7:00:40<21:49,  3.69it/s]
 99%|█████████▊| 366640/371472 [7:00:40<21:55,  3.67it/s]
                                                         
{'loss': 2.7821, 'learning_rate': 1.1171324527570084e-07, 'epoch': 15.79}
+
 99%|█████████▊| 366640/371472 [7:00:40<21:55,  3.67it/s]
 99%|█████████▊| 366641/371472 [7:00:40<22:44,  3.54it/s]
 99%|█████████▊| 366642/371472 [7:00:41<22:26,  3.59it/s]
 99%|█████████▊| 366643/371472 [7:00:41<22:44,  3.54it/s]
 99%|█████████▊| 366644/371472 [7:00:41<22:25,  3.59it/s]
 99%|█████████▊| 366645/371472 [7:00:42<22:43,  3.54it/s]
 99%|█████████▊| 366646/371472 [7:00:42<23:25,  3.43it/s]
 99%|█████████▊| 366647/371472 [7:00:42<22:47,  3.53it/s]
 99%|█████████▊| 366648/371472 [7:00:42<23:13,  3.46it/s]
 99%|█████████▊| 366649/371472 [7:00:43<22:23,  3.59it/s]
 99%|█████████▊| 366650/371472 [7:00:43<23:08,  3.47it/s]
 99%|█████████▊| 366651/371472 [7:00:43<22:36,  3.55it/s]
 99%|█████████▊| 366652/371472 [7:00:44<22:33,  3.56it/s]
 99%|█████████▊| 366653/371472 [7:00:44<22:00,  3.65it/s]
 99%|█████████▊| 366654/371472 [7:00:44<21:59,  3.65it/s]
 99%|█████████▊| 366655/371472 [7:00:44<21:28,  3.74it/s]
 99%|█████████▊| 366656/371472 [7:00:45<21:15,  3.77it/s]
 99%|█████████▊| 366657/371472 [7:00:45<21:36,  3.71it/s]
 99%|█████████▊| 366658/371472 [7:00:45<21:38,  3.71it/s]
 99%|█████████▊| 366659/371472 [7:00:45<21:11,  3.79it/s]
 99%|█████████▊| 366660/371472 [7:00:46<20:34,  3.90it/s]
                                                         
{'loss': 2.5324, 'learning_rate': 1.116647633002219e-07, 'epoch': 15.79}
+
 99%|█████████▊| 366660/371472 [7:00:46<20:34,  3.90it/s]
 99%|█████████▊| 366661/371472 [7:00:46<20:40,  3.88it/s]
 99%|█████████▊| 366662/371472 [7:00:46<21:36,  3.71it/s]
 99%|█████████▊| 366663/371472 [7:00:46<22:13,  3.61it/s]
 99%|█████████▊| 366664/371472 [7:00:47<21:55,  3.65it/s]
 99%|█████████▊| 366665/371472 [7:00:47<22:19,  3.59it/s]
 99%|█████████▊| 366666/371472 [7:00:47<21:59,  3.64it/s]
 99%|█████████▊| 366667/371472 [7:00:48<22:07,  3.62it/s]
 99%|█████████▊| 366668/371472 [7:00:48<23:09,  3.46it/s]
 99%|█████████▊| 366669/371472 [7:00:48<22:39,  3.53it/s]
 99%|█████████▊| 366670/371472 [7:00:48<21:53,  3.66it/s]
 99%|█████████▊| 366671/371472 [7:00:49<21:36,  3.70it/s]
 99%|█████████▊| 366672/371472 [7:00:49<21:22,  3.74it/s]
 99%|█████████▊| 366673/371472 [7:00:49<22:02,  3.63it/s]
 99%|█████████▊| 366674/371472 [7:00:49<21:39,  3.69it/s]
 99%|█████████▊| 366675/371472 [7:00:50<21:42,  3.68it/s]
 99%|█████████▊| 366676/371472 [7:00:50<21:13,  3.77it/s]
 99%|█████████▊| 366677/371472 [7:00:50<21:23,  3.74it/s]
 99%|█████████▊| 366678/371472 [7:00:51<21:29,  3.72it/s]
 99%|█████████▊| 366679/371472 [7:00:51<23:49,  3.35it/s]
 99%|█████████▊| 366680/371472 [7:00:51<22:48,  3.50it/s]
                                                         
{'loss': 2.6013, 'learning_rate': 1.1161628132474306e-07, 'epoch': 15.79}
+
 99%|█████████▊| 366680/371472 [7:00:51<22:48,  3.50it/s]
 99%|█████████▊| 366681/371472 [7:00:51<22:42,  3.52it/s]
 99%|█████████▊| 366682/371472 [7:00:52<21:44,  3.67it/s]
 99%|█████████▊| 366683/371472 [7:00:52<22:10,  3.60it/s]
 99%|█████████▊| 366684/371472 [7:00:52<21:32,  3.70it/s]
 99%|█████████▊| 366685/371472 [7:00:53<21:51,  3.65it/s]
 99%|█████████▊| 366686/371472 [7:00:53<21:52,  3.65it/s]
 99%|█████████▊| 366687/371472 [7:00:53<23:07,  3.45it/s]
 99%|█████████▊| 366688/371472 [7:00:53<22:11,  3.59it/s]
 99%|█████████▊| 366689/371472 [7:00:54<24:17,  3.28it/s]
 99%|█████████▊| 366690/371472 [7:00:54<23:33,  3.38it/s]
 99%|█████████▊| 366691/371472 [7:00:54<23:03,  3.45it/s]
 99%|█████████▊| 366692/371472 [7:00:55<23:13,  3.43it/s]
 99%|█████████▊| 366693/371472 [7:00:55<22:18,  3.57it/s]
 99%|█████████▊| 366694/371472 [7:00:55<21:40,  3.67it/s]
 99%|█████████▊| 366695/371472 [7:00:55<23:05,  3.45it/s]
 99%|█████████▊| 366696/371472 [7:00:56<23:41,  3.36it/s]
 99%|█████████▊| 366697/371472 [7:00:56<22:26,  3.55it/s]
 99%|█████████▊| 366698/371472 [7:00:56<23:07,  3.44it/s]
 99%|█████████▊| 366699/371472 [7:00:57<23:03,  3.45it/s]
 99%|█████████▊| 366700/371472 [7:00:57<22:44,  3.50it/s]
                                                         
{'loss': 2.5211, 'learning_rate': 1.1156779934926412e-07, 'epoch': 15.79}
+
 99%|█████████▊| 366700/371472 [7:00:57<22:44,  3.50it/s]
 99%|█████████▊| 366701/371472 [7:00:57<22:38,  3.51it/s]
 99%|█████████▊| 366702/371472 [7:00:57<22:40,  3.51it/s]
 99%|█████████▊| 366703/371472 [7:00:58<21:57,  3.62it/s]
 99%|█████████▊| 366704/371472 [7:00:58<24:39,  3.22it/s]
 99%|█████████▊| 366705/371472 [7:00:58<25:28,  3.12it/s]
 99%|█████████▊| 366706/371472 [7:00:59<25:25,  3.12it/s]
 99%|█████████▊| 366707/371472 [7:00:59<23:48,  3.34it/s]
 99%|█████████▊| 366708/371472 [7:00:59<22:48,  3.48it/s]
 99%|█████████▊| 366709/371472 [7:01:00<22:33,  3.52it/s]
 99%|█████████▊| 366710/371472 [7:01:00<23:15,  3.41it/s]
 99%|█████████▊| 366711/371472 [7:01:00<23:34,  3.36it/s]
 99%|█████████▊| 366712/371472 [7:01:00<22:04,  3.59it/s]
 99%|█████████▊| 366713/371472 [7:01:01<21:10,  3.74it/s]
 99%|█████████▊| 366714/371472 [7:01:01<21:06,  3.76it/s]
 99%|█████████▊| 366715/371472 [7:01:01<20:28,  3.87it/s]
 99%|█████████▊| 366716/371472 [7:01:01<21:01,  3.77it/s]
 99%|█████████▊| 366717/371472 [7:01:02<22:13,  3.57it/s]
 99%|█████████▊| 366718/371472 [7:01:02<22:22,  3.54it/s]
 99%|█████████▊| 366719/371472 [7:01:02<22:20,  3.55it/s]
 99%|█████████▊| 366720/371472 [7:01:03<21:14,  3.73it/s]
                                                         
{'loss': 2.741, 'learning_rate': 1.1151931737378527e-07, 'epoch': 15.8}
+
 99%|█████████▊| 366720/371472 [7:01:03<21:14,  3.73it/s]
 99%|█████████▊| 366721/371472 [7:01:03<24:25,  3.24it/s]
 99%|█████████▊| 366722/371472 [7:01:03<23:09,  3.42it/s]
 99%|█████████▊| 366723/371472 [7:01:03<22:12,  3.56it/s]
 99%|█████████▊| 366724/371472 [7:01:04<21:49,  3.63it/s]
 99%|█████████▊| 366725/371472 [7:01:04<22:08,  3.57it/s]
 99%|█████████▊| 366726/371472 [7:01:04<21:39,  3.65it/s]
 99%|█████████▊| 366727/371472 [7:01:05<22:18,  3.54it/s]
 99%|█████████▊| 366728/371472 [7:01:05<22:22,  3.53it/s]
 99%|█████████▊| 366729/371472 [7:01:05<22:25,  3.52it/s]
 99%|█████████▊| 366730/371472 [7:01:05<23:18,  3.39it/s]
 99%|█████████▊| 366731/371472 [7:01:06<22:10,  3.56it/s]
 99%|█████████▊| 366732/371472 [7:01:06<21:59,  3.59it/s]
 99%|█████████▊| 366733/371472 [7:01:06<21:40,  3.64it/s]
 99%|█████████▊| 366734/371472 [7:01:06<20:57,  3.77it/s]
 99%|█████████▊| 366735/371472 [7:01:07<22:25,  3.52it/s]
 99%|█████████▊| 366736/371472 [7:01:07<22:08,  3.57it/s]
 99%|█████████▊| 366737/371472 [7:01:07<21:49,  3.62it/s]
 99%|█████████▊| 366738/371472 [7:01:08<21:12,  3.72it/s]
 99%|█████████▊| 366739/371472 [7:01:08<20:48,  3.79it/s]
 99%|█████████▊| 366740/371472 [7:01:08<20:44,  3.80it/s]
                                                         
{'loss': 2.6795, 'learning_rate': 1.1147083539830633e-07, 'epoch': 15.8}
+
 99%|█████████▊| 366740/371472 [7:01:08<20:44,  3.80it/s]
 99%|█████████▊| 366741/371472 [7:01:08<21:51,  3.61it/s]
 99%|█████████▊| 366742/371472 [7:01:09<20:50,  3.78it/s]
 99%|█████████▊| 366743/371472 [7:01:09<20:17,  3.88it/s]
 99%|█████████▊| 366744/371472 [7:01:09<20:12,  3.90it/s]
 99%|█████████▊| 366745/371472 [7:01:09<21:09,  3.72it/s]
 99%|█████████▊| 366746/371472 [7:01:10<22:04,  3.57it/s]
 99%|█████████▊| 366747/371472 [7:01:10<23:59,  3.28it/s]
 99%|█████████▊| 366748/371472 [7:01:10<23:03,  3.41it/s]
 99%|█████████▊| 366749/371472 [7:01:11<22:30,  3.50it/s]
 99%|█████████▊| 366750/371472 [7:01:11<21:21,  3.68it/s]
 99%|█████████▊| 366751/371472 [7:01:11<21:23,  3.68it/s]
 99%|█████████▊| 366752/371472 [7:01:11<20:57,  3.75it/s]
 99%|█████████▊| 366753/371472 [7:01:12<20:46,  3.78it/s]
 99%|█████████▊| 366754/371472 [7:01:12<20:18,  3.87it/s]
 99%|█████████▊| 366755/371472 [7:01:12<21:46,  3.61it/s]
 99%|█████████▊| 366756/371472 [7:01:13<22:04,  3.56it/s]
 99%|█████████▊| 366757/371472 [7:01:13<21:50,  3.60it/s]
 99%|█████████▊| 366758/371472 [7:01:13<22:18,  3.52it/s]
 99%|█████████▊| 366759/371472 [7:01:13<22:54,  3.43it/s]
 99%|█████████▊| 366760/371472 [7:01:14<24:08,  3.25it/s]
                                                         
{'loss': 2.4631, 'learning_rate': 1.1142235342282749e-07, 'epoch': 15.8}
+
 99%|█████████▊| 366760/371472 [7:01:14<24:08,  3.25it/s]
 99%|█████████▊| 366761/371472 [7:01:14<23:45,  3.30it/s]
 99%|█████████▊| 366762/371472 [7:01:14<23:50,  3.29it/s]
 99%|█████████▊| 366763/371472 [7:01:15<23:33,  3.33it/s]
 99%|█████████▊| 366764/371472 [7:01:15<22:10,  3.54it/s]
 99%|█████████▊| 366765/371472 [7:01:15<22:25,  3.50it/s]
 99%|█████████▊| 366766/371472 [7:01:16<23:58,  3.27it/s]
 99%|█████████▊| 366767/371472 [7:01:16<24:39,  3.18it/s]
 99%|█████████▊| 366768/371472 [7:01:16<24:06,  3.25it/s]
 99%|█████████▊| 366769/371472 [7:01:16<22:21,  3.51it/s]
 99%|█████████▊| 366770/371472 [7:01:17<22:01,  3.56it/s]
 99%|█████████▊| 366771/371472 [7:01:17<21:50,  3.59it/s]
 99%|█████████▊| 366772/371472 [7:01:17<22:26,  3.49it/s]
 99%|█████████▊| 366773/371472 [7:01:18<22:21,  3.50it/s]
 99%|█████████▊| 366774/371472 [7:01:18<23:02,  3.40it/s]
 99%|█████████▊| 366775/371472 [7:01:18<22:25,  3.49it/s]
 99%|█████████▊| 366776/371472 [7:01:18<23:52,  3.28it/s]
 99%|█████████▊| 366777/371472 [7:01:19<22:59,  3.40it/s]
 99%|█████████▊| 366778/371472 [7:01:19<22:28,  3.48it/s]
 99%|█████████▊| 366779/371472 [7:01:19<23:05,  3.39it/s]
 99%|█████████▊| 366780/371472 [7:01:20<21:44,  3.60it/s]
                                                         
{'loss': 2.5864, 'learning_rate': 1.1137387144734855e-07, 'epoch': 15.8}
+
 99%|█████████▊| 366780/371472 [7:01:20<21:44,  3.60it/s]
 99%|█████████▊| 366781/371472 [7:01:20<21:41,  3.61it/s]
 99%|█████████▊| 366782/371472 [7:01:20<22:15,  3.51it/s]
 99%|█████████▊| 366783/371472 [7:01:20<22:06,  3.53it/s]
 99%|█████████▊| 366784/371472 [7:01:21<22:32,  3.47it/s]
 99%|█████████▊| 366785/371472 [7:01:21<23:12,  3.37it/s]
 99%|█████████▊| 366786/371472 [7:01:21<23:00,  3.39it/s]
 99%|█████████▊| 366787/371472 [7:01:22<22:44,  3.43it/s]
 99%|█████████▊| 366788/371472 [7:01:22<22:34,  3.46it/s]
 99%|█████████▊| 366789/371472 [7:01:22<21:14,  3.68it/s]
 99%|█████████▊| 366790/371472 [7:01:22<20:59,  3.72it/s]
 99%|█████████▊| 366791/371472 [7:01:23<21:49,  3.57it/s]
 99%|█████████▊| 366792/371472 [7:01:23<21:56,  3.55it/s]
 99%|█████████▊| 366793/371472 [7:01:23<21:32,  3.62it/s]
 99%|█████████▊| 366794/371472 [7:01:23<21:10,  3.68it/s]
 99%|█████████▊| 366795/371472 [7:01:24<20:53,  3.73it/s]
 99%|█████████▊| 366796/371472 [7:01:24<21:10,  3.68it/s]
 99%|█████████▊| 366797/371472 [7:01:24<21:08,  3.68it/s]
 99%|█████████▊| 366798/371472 [7:01:25<21:01,  3.71it/s]
 99%|█████████▊| 366799/371472 [7:01:25<22:15,  3.50it/s]
 99%|█████████▊| 366800/371472 [7:01:25<22:02,  3.53it/s]
                                                         
{'loss': 2.6844, 'learning_rate': 1.113253894718697e-07, 'epoch': 15.8}
+
 99%|█████████▊| 366800/371472 [7:01:25<22:02,  3.53it/s]
 99%|█████████▊| 366801/371472 [7:01:25<21:10,  3.68it/s]
 99%|█████████▊| 366802/371472 [7:01:26<20:43,  3.76it/s]
 99%|█████████▊| 366803/371472 [7:01:26<20:25,  3.81it/s]
 99%|█████████▊| 366804/371472 [7:01:26<20:41,  3.76it/s]
 99%|█████████▊| 366805/371472 [7:01:27<26:05,  2.98it/s]
 99%|█████████▊| 366806/371472 [7:01:27<24:32,  3.17it/s]
 99%|█████████▊| 366807/371472 [7:01:27<24:01,  3.24it/s]
 99%|█████████▊| 366808/371472 [7:01:27<22:31,  3.45it/s]
 99%|█████████▊| 366809/371472 [7:01:28<21:09,  3.67it/s]
 99%|█████████▊| 366810/371472 [7:01:28<21:15,  3.66it/s]
 99%|█████████▊| 366811/371472 [7:01:28<22:26,  3.46it/s]
 99%|█████████▊| 366812/371472 [7:01:29<22:43,  3.42it/s]
 99%|█████████▊| 366813/371472 [7:01:29<21:47,  3.56it/s]
 99%|█████████▊| 366814/371472 [7:01:29<22:05,  3.51it/s]
 99%|█████████▊| 366815/371472 [7:01:29<21:59,  3.53it/s]
 99%|█████████▊| 366816/371472 [7:01:30<21:45,  3.57it/s]
 99%|█████████▊| 366817/371472 [7:01:30<24:42,  3.14it/s]
 99%|█████████▊| 366818/371472 [7:01:30<23:28,  3.30it/s]
 99%|█████████▊| 366819/371472 [7:01:31<25:06,  3.09it/s]
 99%|█████████▊| 366820/371472 [7:01:31<23:49,  3.25it/s]
                                                         
{'loss': 2.6174, 'learning_rate': 1.1127690749639076e-07, 'epoch': 15.8}
+
 99%|█████████▊| 366820/371472 [7:01:31<23:49,  3.25it/s]
 99%|█████████▊| 366821/371472 [7:01:31<22:26,  3.45it/s]
 99%|█████████▊| 366822/371472 [7:01:32<23:25,  3.31it/s]
 99%|█████████▊| 366823/371472 [7:01:32<22:50,  3.39it/s]
 99%|█████████▊| 366824/371472 [7:01:32<22:29,  3.44it/s]
 99%|█████████▊| 366825/371472 [7:01:32<21:53,  3.54it/s]
 99%|█████████▊| 366826/371472 [7:01:33<21:12,  3.65it/s]
 99%|█████████▊| 366827/371472 [7:01:33<20:41,  3.74it/s]
 99%|█████████▊| 366828/371472 [7:01:33<22:07,  3.50it/s]
 99%|█████████▉| 366829/371472 [7:01:34<21:36,  3.58it/s]
 99%|█████████▉| 366830/371472 [7:01:34<24:23,  3.17it/s]
 99%|█████████▉| 366831/371472 [7:01:34<23:10,  3.34it/s]
 99%|█████████▉| 366832/371472 [7:01:35<26:12,  2.95it/s]
 99%|█████████▉| 366833/371472 [7:01:35<25:05,  3.08it/s]
 99%|█████████▉| 366834/371472 [7:01:35<24:40,  3.13it/s]
 99%|█████████▉| 366835/371472 [7:01:35<23:15,  3.32it/s]
 99%|█████████▉| 366836/371472 [7:01:36<22:30,  3.43it/s]
 99%|█████████▉| 366837/371472 [7:01:36<22:17,  3.46it/s]
 99%|█████████▉| 366838/371472 [7:01:36<21:54,  3.52it/s]
 99%|█████████▉| 366839/371472 [7:01:37<22:07,  3.49it/s]
 99%|█████████▉| 366840/371472 [7:01:37<22:41,  3.40it/s]
                                                         
{'loss': 2.6589, 'learning_rate': 1.1122842552091192e-07, 'epoch': 15.8}
+
 99%|█████████▉| 366840/371472 [7:01:37<22:41,  3.40it/s]
 99%|█████████▉| 366841/371472 [7:01:37<22:36,  3.41it/s]
 99%|█████████▉| 366842/371472 [7:01:37<21:51,  3.53it/s]
 99%|█████████▉| 366843/371472 [7:01:38<21:11,  3.64it/s]
 99%|█████████▉| 366844/371472 [7:01:38<21:26,  3.60it/s]
 99%|█████████▉| 366845/371472 [7:01:38<20:42,  3.72it/s]
 99%|█████████▉| 366846/371472 [7:01:39<20:37,  3.74it/s]
 99%|█████████▉| 366847/371472 [7:01:39<26:56,  2.86it/s]
 99%|█████████▉| 366848/371472 [7:01:39<24:19,  3.17it/s]
 99%|█████████▉| 366849/371472 [7:01:40<22:59,  3.35it/s]
 99%|█████████▉| 366850/371472 [7:01:40<24:28,  3.15it/s]
 99%|█████████▉| 366851/371472 [7:01:40<25:07,  3.06it/s]
 99%|█████████▉| 366852/371472 [7:01:41<23:54,  3.22it/s]
 99%|█████████▉| 366853/371472 [7:01:41<22:45,  3.38it/s]
 99%|█████████▉| 366854/371472 [7:01:41<21:50,  3.52it/s]
 99%|█████████▉| 366855/371472 [7:01:41<21:12,  3.63it/s]
 99%|█████████▉| 366856/371472 [7:01:42<22:15,  3.46it/s]
 99%|█████████▉| 366857/371472 [7:01:42<22:33,  3.41it/s]
 99%|█████████▉| 366858/371472 [7:01:42<21:39,  3.55it/s]
 99%|█████████▉| 366859/371472 [7:01:42<20:53,  3.68it/s]
 99%|█████████▉| 366860/371472 [7:01:43<20:30,  3.75it/s]
                                                         
{'loss': 2.6067, 'learning_rate': 1.1117994354543297e-07, 'epoch': 15.8}
+
 99%|█████████▉| 366860/371472 [7:01:43<20:30,  3.75it/s]
 99%|█████████▉| 366861/371472 [7:01:43<20:16,  3.79it/s]
 99%|█████████▉| 366862/371472 [7:01:43<20:40,  3.72it/s]
 99%|█████████▉| 366863/371472 [7:01:43<19:55,  3.86it/s]
 99%|█████████▉| 366864/371472 [7:01:44<20:09,  3.81it/s]
 99%|█████████▉| 366865/371472 [7:01:44<20:59,  3.66it/s]
 99%|█████████▉| 366866/371472 [7:01:44<22:32,  3.40it/s]
 99%|█████████▉| 366867/371472 [7:01:45<22:20,  3.44it/s]
 99%|█████████▉| 366868/371472 [7:01:45<21:33,  3.56it/s]
 99%|█████████▉| 366869/371472 [7:01:45<22:36,  3.39it/s]
 99%|█████████▉| 366870/371472 [7:01:46<21:51,  3.51it/s]
 99%|█████████▉| 366871/371472 [7:01:46<21:59,  3.49it/s]
 99%|█████████▉| 366872/371472 [7:01:46<23:50,  3.22it/s]
 99%|█████████▉| 366873/371472 [7:01:46<22:52,  3.35it/s]
 99%|█████████▉| 366874/371472 [7:01:47<22:19,  3.43it/s]
 99%|█████████▉| 366875/371472 [7:01:47<22:16,  3.44it/s]
 99%|█████████▉| 366876/371472 [7:01:47<21:27,  3.57it/s]
 99%|█████████▉| 366877/371472 [7:01:48<21:55,  3.49it/s]
 99%|█████████▉| 366878/371472 [7:01:48<21:35,  3.55it/s]
 99%|█████████▉| 366879/371472 [7:01:48<20:51,  3.67it/s]
 99%|█████████▉| 366880/371472 [7:01:48<20:58,  3.65it/s]
                                                         
{'loss': 2.4082, 'learning_rate': 1.1113146156995413e-07, 'epoch': 15.8}
+
 99%|█████████▉| 366880/371472 [7:01:48<20:58,  3.65it/s]
 99%|█████████▉| 366881/371472 [7:01:49<20:28,  3.74it/s]
 99%|█████████▉| 366882/371472 [7:01:49<20:13,  3.78it/s]
 99%|█████████▉| 366883/371472 [7:01:49<21:03,  3.63it/s]
 99%|█████████▉| 366884/371472 [7:01:49<20:13,  3.78it/s]
 99%|█████████▉| 366885/371472 [7:01:50<20:05,  3.81it/s]
 99%|█████████▉| 366886/371472 [7:01:50<20:09,  3.79it/s]
 99%|█████████▉| 366887/371472 [7:01:50<21:16,  3.59it/s]
 99%|█████████▉| 366888/371472 [7:01:51<21:27,  3.56it/s]
 99%|█████████▉| 366889/371472 [7:01:51<20:56,  3.65it/s]
 99%|█████████▉| 366890/371472 [7:01:51<22:41,  3.37it/s]
 99%|█████████▉| 366891/371472 [7:01:51<22:20,  3.42it/s]
 99%|█████████▉| 366892/371472 [7:01:52<21:56,  3.48it/s]
 99%|█████████▉| 366893/371472 [7:01:52<22:36,  3.38it/s]
 99%|█████████▉| 366894/371472 [7:01:52<22:34,  3.38it/s]
 99%|█████████▉| 366895/371472 [7:01:53<22:04,  3.46it/s]
 99%|█████████▉| 366896/371472 [7:01:53<21:44,  3.51it/s]
 99%|█████████▉| 366897/371472 [7:01:53<22:00,  3.47it/s]
 99%|█████████▉| 366898/371472 [7:01:53<21:47,  3.50it/s]
 99%|█████████▉| 366899/371472 [7:01:54<23:20,  3.27it/s]
 99%|█████████▉| 366900/371472 [7:01:54<22:32,  3.38it/s]
                                                         
{'loss': 2.5794, 'learning_rate': 1.1108297959447519e-07, 'epoch': 15.8}
+
 99%|█████████▉| 366900/371472 [7:01:54<22:32,  3.38it/s]
 99%|█████████▉| 366901/371472 [7:01:54<21:45,  3.50it/s]
 99%|█████████▉| 366902/371472 [7:01:55<22:46,  3.34it/s]
 99%|█████████▉| 366903/371472 [7:01:55<22:00,  3.46it/s]
 99%|█████████▉| 366904/371472 [7:01:55<21:03,  3.62it/s]
 99%|█████████▉| 366905/371472 [7:01:55<20:21,  3.74it/s]
 99%|█████████▉| 366906/371472 [7:01:56<19:38,  3.88it/s]
 99%|█████████▉| 366907/371472 [7:01:56<21:21,  3.56it/s]
 99%|█████████▉| 366908/371472 [7:01:56<21:43,  3.50it/s]
 99%|█████████▉| 366909/371472 [7:01:57<21:22,  3.56it/s]
 99%|█████████▉| 366910/371472 [7:01:57<21:12,  3.58it/s]
 99%|█████████▉| 366911/371472 [7:01:57<21:35,  3.52it/s]
 99%|█████████▉| 366912/371472 [7:01:57<21:58,  3.46it/s]
 99%|█████████▉| 366913/371472 [7:01:58<21:17,  3.57it/s]
 99%|█████████▉| 366914/371472 [7:01:58<20:58,  3.62it/s]
 99%|█████████▉| 366915/371472 [7:01:58<21:12,  3.58it/s]
 99%|█████████▉| 366916/371472 [7:01:59<20:51,  3.64it/s]
 99%|█████████▉| 366917/371472 [7:01:59<21:07,  3.59it/s]
 99%|█████████▉| 366918/371472 [7:01:59<20:30,  3.70it/s]
 99%|█████████▉| 366919/371472 [7:01:59<19:47,  3.83it/s]
 99%|█████████▉| 366920/371472 [7:02:00<23:06,  3.28it/s]
                                                         
{'loss': 2.5175, 'learning_rate': 1.1103449761899634e-07, 'epoch': 15.8}
+
 99%|█████████▉| 366920/371472 [7:02:00<23:06,  3.28it/s]
 99%|█████████▉| 366921/371472 [7:02:00<22:04,  3.44it/s]
 99%|█████████▉| 366922/371472 [7:02:00<21:30,  3.53it/s]
 99%|█████████▉| 366923/371472 [7:02:00<20:47,  3.65it/s]
 99%|█████████▉| 366924/371472 [7:02:01<21:21,  3.55it/s]
 99%|█████████▉| 366925/371472 [7:02:01<21:43,  3.49it/s]
 99%|█████████▉| 366926/371472 [7:02:01<21:23,  3.54it/s]
 99%|█████████▉| 366927/371472 [7:02:02<22:34,  3.36it/s]
 99%|█████████▉| 366928/371472 [7:02:02<24:17,  3.12it/s]
 99%|█████████▉| 366929/371472 [7:02:02<23:31,  3.22it/s]
 99%|█████████▉| 366930/371472 [7:02:03<23:20,  3.24it/s]
 99%|█████████▉| 366931/371472 [7:02:03<22:15,  3.40it/s]
 99%|█████████▉| 366932/371472 [7:02:03<24:05,  3.14it/s]
 99%|█████████▉| 366933/371472 [7:02:04<22:24,  3.38it/s]
 99%|█████████▉| 366934/371472 [7:02:04<21:16,  3.56it/s]
 99%|█████████▉| 366935/371472 [7:02:04<21:09,  3.57it/s]
 99%|█████████▉| 366936/371472 [7:02:04<20:55,  3.61it/s]
 99%|█████████▉| 366937/371472 [7:02:05<20:02,  3.77it/s]
 99%|█████████▉| 366938/371472 [7:02:05<19:58,  3.78it/s]
 99%|█████████▉| 366939/371472 [7:02:05<19:50,  3.81it/s]
 99%|█████████▉| 366940/371472 [7:02:05<19:16,  3.92it/s]
                                                         
{'loss': 2.5756, 'learning_rate': 1.109860156435174e-07, 'epoch': 15.8}
+
 99%|█████████▉| 366940/371472 [7:02:05<19:16,  3.92it/s]
 99%|█████████▉| 366941/371472 [7:02:06<19:35,  3.85it/s]
 99%|█████████▉| 366942/371472 [7:02:06<19:20,  3.90it/s]
 99%|█████████▉| 366943/371472 [7:02:06<20:13,  3.73it/s]
 99%|█████████▉| 366944/371472 [7:02:06<19:50,  3.80it/s]
 99%|█████████▉| 366945/371472 [7:02:07<19:27,  3.88it/s]
 99%|█████████▉| 366946/371472 [7:02:07<19:38,  3.84it/s]
 99%|█████████▉| 366947/371472 [7:02:07<20:10,  3.74it/s]
 99%|█████████▉| 366948/371472 [7:02:07<20:00,  3.77it/s]
 99%|█████████▉| 366949/371472 [7:02:08<20:25,  3.69it/s]
 99%|█████████▉| 366950/371472 [7:02:08<19:50,  3.80it/s]
 99%|█████████▉| 366951/371472 [7:02:08<20:27,  3.68it/s]
 99%|█████████▉| 366952/371472 [7:02:09<20:33,  3.66it/s]
 99%|█████████▉| 366953/371472 [7:02:09<19:36,  3.84it/s]
 99%|█████████▉| 366954/371472 [7:02:09<20:49,  3.62it/s]
 99%|█████████▉| 366955/371472 [7:02:09<20:08,  3.74it/s]
 99%|█████████▉| 366956/371472 [7:02:10<19:57,  3.77it/s]
 99%|█████████▉| 366957/371472 [7:02:10<20:44,  3.63it/s]
 99%|█████████▉| 366958/371472 [7:02:10<20:53,  3.60it/s]
 99%|█████████▉| 366959/371472 [7:02:10<20:16,  3.71it/s]
 99%|█████████▉| 366960/371472 [7:02:11<20:03,  3.75it/s]
                                                         
{'loss': 2.5556, 'learning_rate': 1.1093753366803856e-07, 'epoch': 15.81}
+
 99%|█████████▉| 366960/371472 [7:02:11<20:03,  3.75it/s]
 99%|█████████▉| 366961/371472 [7:02:11<20:20,  3.69it/s]
 99%|█████████▉| 366962/371472 [7:02:11<20:53,  3.60it/s]
 99%|█████████▉| 366963/371472 [7:02:12<20:47,  3.61it/s]
 99%|█████████▉| 366964/371472 [7:02:12<20:24,  3.68it/s]
 99%|█████████▉| 366965/371472 [7:02:12<20:37,  3.64it/s]
 99%|█████████▉| 366966/371472 [7:02:12<21:47,  3.45it/s]
 99%|█████████▉| 366967/371472 [7:02:13<21:28,  3.50it/s]
 99%|█████████▉| 366968/371472 [7:02:13<22:40,  3.31it/s]
 99%|█████████▉| 366969/371472 [7:02:13<22:58,  3.27it/s]
 99%|█████████▉| 366970/371472 [7:02:14<22:51,  3.28it/s]
 99%|█████████▉| 366971/371472 [7:02:14<22:01,  3.41it/s]
 99%|█████████▉| 366972/371472 [7:02:14<22:09,  3.39it/s]
 99%|█████████▉| 366973/371472 [7:02:14<21:32,  3.48it/s]
 99%|█████████▉| 366974/371472 [7:02:15<20:44,  3.61it/s]
 99%|█████████▉| 366975/371472 [7:02:15<20:04,  3.73it/s]
 99%|█████████▉| 366976/371472 [7:02:15<20:39,  3.63it/s]
 99%|█████████▉| 366977/371472 [7:02:16<20:54,  3.58it/s]
 99%|█████████▉| 366978/371472 [7:02:16<21:33,  3.48it/s]
 99%|█████████▉| 366979/371472 [7:02:16<21:51,  3.42it/s]
 99%|█████████▉| 366980/371472 [7:02:16<21:49,  3.43it/s]
                                                         
{'loss': 2.442, 'learning_rate': 1.1088905169255962e-07, 'epoch': 15.81}
+
 99%|█████████▉| 366980/371472 [7:02:16<21:49,  3.43it/s]
 99%|█████████▉| 366981/371472 [7:02:17<21:45,  3.44it/s]
 99%|█████████▉| 366982/371472 [7:02:17<21:06,  3.54it/s]
 99%|█████████▉| 366983/371472 [7:02:17<20:58,  3.57it/s]
 99%|█████████▉| 366984/371472 [7:02:18<21:00,  3.56it/s]
 99%|█████████▉| 366985/371472 [7:02:18<21:07,  3.54it/s]
 99%|█████████▉| 366986/371472 [7:02:18<20:12,  3.70it/s]
 99%|█████████▉| 366987/371472 [7:02:18<19:49,  3.77it/s]
 99%|█████████▉| 366988/371472 [7:02:19<19:13,  3.89it/s]
 99%|█████████▉| 366989/371472 [7:02:19<20:09,  3.71it/s]
 99%|█████████▉| 366990/371472 [7:02:19<20:35,  3.63it/s]
 99%|█████████▉| 366991/371472 [7:02:19<20:28,  3.65it/s]
 99%|█████████▉| 366992/371472 [7:02:20<20:17,  3.68it/s]
 99%|█████████▉| 366993/371472 [7:02:20<20:03,  3.72it/s]
 99%|█████████▉| 366994/371472 [7:02:20<20:00,  3.73it/s]
 99%|█████████▉| 366995/371472 [7:02:20<19:31,  3.82it/s]
 99%|█████████▉| 366996/371472 [7:02:21<19:09,  3.89it/s]
 99%|█████████▉| 366997/371472 [7:02:21<19:18,  3.86it/s]
 99%|█████████▉| 366998/371472 [7:02:21<20:04,  3.71it/s]
 99%|█████████▉| 366999/371472 [7:02:22<20:16,  3.68it/s]
 99%|█████████▉| 367000/371472 [7:02:22<20:13,  3.69it/s]
                                                         
{'loss': 2.5211, 'learning_rate': 1.1084056971708077e-07, 'epoch': 15.81}
+
 99%|█████████▉| 367000/371472 [7:02:22<20:13,  3.69it/s]
 99%|█████████▉| 367001/371472 [7:02:22<21:04,  3.54it/s]
 99%|█████████▉| 367002/371472 [7:02:22<21:25,  3.48it/s]
 99%|█████████▉| 367003/371472 [7:02:23<20:50,  3.58it/s]
 99%|█████████▉| 367004/371472 [7:02:23<21:16,  3.50it/s]
 99%|█████████▉| 367005/371472 [7:02:23<20:53,  3.56it/s]
 99%|█████████▉| 367006/371472 [7:02:24<20:45,  3.59it/s]
 99%|█████████▉| 367007/371472 [7:02:24<20:11,  3.69it/s]
 99%|█████████▉| 367008/371472 [7:02:24<20:03,  3.71it/s]
 99%|█████████▉| 367009/371472 [7:02:24<20:27,  3.64it/s]
 99%|█████████▉| 367010/371472 [7:02:25<21:03,  3.53it/s]
 99%|█████████▉| 367011/371472 [7:02:25<20:11,  3.68it/s]
 99%|█████████▉| 367012/371472 [7:02:25<20:24,  3.64it/s]
 99%|█████████▉| 367013/371472 [7:02:25<20:14,  3.67it/s]
 99%|█████████▉| 367014/371472 [7:02:26<20:24,  3.64it/s]
 99%|█████████▉| 367015/371472 [7:02:26<19:38,  3.78it/s]
 99%|█████████▉| 367016/371472 [7:02:26<20:19,  3.65it/s]
 99%|█████████▉| 367017/371472 [7:02:27<20:03,  3.70it/s]
 99%|█████████▉| 367018/371472 [7:02:27<20:44,  3.58it/s]
 99%|█████████▉| 367019/371472 [7:02:27<20:36,  3.60it/s]
 99%|█████████▉| 367020/371472 [7:02:27<20:25,  3.63it/s]
                                                         
{'loss': 2.5588, 'learning_rate': 1.1079208774160183e-07, 'epoch': 15.81}
+
 99%|█████████▉| 367020/371472 [7:02:27<20:25,  3.63it/s]
 99%|█████████▉| 367021/371472 [7:02:28<20:16,  3.66it/s]
 99%|█████████▉| 367022/371472 [7:02:28<20:03,  3.70it/s]
 99%|█████████▉| 367023/371472 [7:02:28<19:46,  3.75it/s]
 99%|█████████▉| 367024/371472 [7:02:28<20:21,  3.64it/s]
 99%|█████████▉| 367025/371472 [7:02:29<20:05,  3.69it/s]
 99%|█████████▉| 367026/371472 [7:02:29<19:57,  3.71it/s]
 99%|█████████▉| 367027/371472 [7:02:29<21:40,  3.42it/s]
 99%|█████████▉| 367028/371472 [7:02:30<21:33,  3.44it/s]
 99%|█████████▉| 367029/371472 [7:02:30<21:07,  3.51it/s]
 99%|█████████▉| 367030/371472 [7:02:30<23:02,  3.21it/s]
 99%|█████████▉| 367031/371472 [7:02:30<21:56,  3.37it/s]
 99%|█████████▉| 367032/371472 [7:02:31<22:35,  3.27it/s]
 99%|█████████▉| 367033/371472 [7:02:31<22:19,  3.31it/s]
 99%|█████████▉| 367034/371472 [7:02:31<20:56,  3.53it/s]
 99%|█████████▉| 367035/371472 [7:02:32<21:14,  3.48it/s]
 99%|█████████▉| 367036/371472 [7:02:32<20:28,  3.61it/s]
 99%|█████████▉| 367037/371472 [7:02:32<20:30,  3.61it/s]
 99%|█████████▉| 367038/371472 [7:02:32<20:08,  3.67it/s]
 99%|█████████▉| 367039/371472 [7:02:33<20:10,  3.66it/s]
 99%|█████████▉| 367040/371472 [7:02:33<21:12,  3.48it/s]
                                                         
{'loss': 2.6073, 'learning_rate': 1.10743605766123e-07, 'epoch': 15.81}
+
 99%|█████████▉| 367040/371472 [7:02:33<21:12,  3.48it/s]
 99%|█████████▉| 367041/371472 [7:02:33<20:33,  3.59it/s]
 99%|█████████▉| 367042/371472 [7:02:34<21:06,  3.50it/s]
 99%|█████████▉| 367043/371472 [7:02:34<21:11,  3.48it/s]
 99%|█████████▉| 367044/371472 [7:02:34<21:11,  3.48it/s]
 99%|█████████▉| 367045/371472 [7:02:34<21:23,  3.45it/s]
 99%|█████████▉| 367046/371472 [7:02:35<20:44,  3.56it/s]
 99%|█████████▉| 367047/371472 [7:02:35<21:08,  3.49it/s]
 99%|█████████▉| 367048/371472 [7:02:35<20:30,  3.59it/s]
 99%|█████████▉| 367049/371472 [7:02:36<20:14,  3.64it/s]
 99%|█████████▉| 367050/371472 [7:02:36<20:38,  3.57it/s]
 99%|█████████▉| 367051/371472 [7:02:36<21:03,  3.50it/s]
 99%|█████████▉| 367052/371472 [7:02:36<20:19,  3.62it/s]
 99%|█████████▉| 367053/371472 [7:02:37<23:33,  3.13it/s]
 99%|█████████▉| 367054/371472 [7:02:37<22:31,  3.27it/s]
 99%|█████████▉| 367055/371472 [7:02:37<21:18,  3.46it/s]
 99%|█████████▉| 367056/371472 [7:02:38<22:29,  3.27it/s]
 99%|█████████▉| 367057/371472 [7:02:38<21:34,  3.41it/s]
 99%|█████████▉| 367058/371472 [7:02:38<21:53,  3.36it/s]
 99%|█████████▉| 367059/371472 [7:02:39<20:54,  3.52it/s]
 99%|█████████▉| 367060/371472 [7:02:39<21:28,  3.42it/s]
                                                         
{'loss': 2.5385, 'learning_rate': 1.1069512379064405e-07, 'epoch': 15.81}
+
 99%|█████████▉| 367060/371472 [7:02:39<21:28,  3.42it/s]
 99%|█████████▉| 367061/371472 [7:02:39<21:27,  3.43it/s]
 99%|█████████▉| 367062/371472 [7:02:39<21:01,  3.50it/s]
 99%|█████████▉| 367063/371472 [7:02:40<20:11,  3.64it/s]
 99%|█████████▉| 367064/371472 [7:02:40<19:46,  3.72it/s]
 99%|█████████▉| 367065/371472 [7:02:40<20:42,  3.55it/s]
 99%|█████████▉| 367066/371472 [7:02:40<20:12,  3.63it/s]
 99%|█████████▉| 367067/371472 [7:02:41<19:43,  3.72it/s]
 99%|█████████▉| 367068/371472 [7:02:41<19:22,  3.79it/s]
 99%|█████████▉| 367069/371472 [7:02:41<20:34,  3.57it/s]
 99%|█████████▉| 367070/371472 [7:02:42<20:44,  3.54it/s]
 99%|█████████▉| 367071/371472 [7:02:42<20:06,  3.65it/s]
 99%|█████████▉| 367072/371472 [7:02:42<20:00,  3.66it/s]
 99%|█████████▉| 367073/371472 [7:02:42<19:36,  3.74it/s]
 99%|█████████▉| 367074/371472 [7:02:43<19:35,  3.74it/s]
 99%|█████████▉| 367075/371472 [7:02:43<20:20,  3.60it/s]
 99%|█████████▉| 367076/371472 [7:02:43<21:27,  3.41it/s]
 99%|█████████▉| 367077/371472 [7:02:44<21:03,  3.48it/s]
 99%|█████████▉| 367078/371472 [7:02:44<20:59,  3.49it/s]
 99%|█████████▉| 367079/371472 [7:02:44<22:04,  3.32it/s]
 99%|█████████▉| 367080/371472 [7:02:44<20:59,  3.49it/s]
                                                         
{'loss': 2.7344, 'learning_rate': 1.1064664181516511e-07, 'epoch': 15.81}
+
 99%|█████████▉| 367080/371472 [7:02:44<20:59,  3.49it/s]
 99%|█████████▉| 367081/371472 [7:02:45<20:47,  3.52it/s]
 99%|█████████▉| 367082/371472 [7:02:45<20:20,  3.60it/s]
 99%|█████████▉| 367083/371472 [7:02:45<20:21,  3.59it/s]
 99%|█████████▉| 367084/371472 [7:02:45<20:07,  3.64it/s]
 99%|█████████▉| 367085/371472 [7:02:46<21:04,  3.47it/s]
 99%|█████████▉| 367086/371472 [7:02:46<21:42,  3.37it/s]
 99%|█████████▉| 367087/371472 [7:02:46<21:50,  3.35it/s]
 99%|█████████▉| 367088/371472 [7:02:47<21:34,  3.39it/s]
 99%|█████████▉| 367089/371472 [7:02:47<21:28,  3.40it/s]
 99%|█████████▉| 367090/371472 [7:02:47<21:29,  3.40it/s]
 99%|█████████▉| 367091/371472 [7:02:48<21:04,  3.47it/s]
 99%|█████████▉| 367092/371472 [7:02:48<20:16,  3.60it/s]
 99%|█████████▉| 367093/371472 [7:02:48<20:05,  3.63it/s]
 99%|█████████▉| 367094/371472 [7:02:48<20:57,  3.48it/s]
 99%|█████████▉| 367095/371472 [7:02:49<20:16,  3.60it/s]
 99%|█████████▉| 367096/371472 [7:02:49<20:41,  3.52it/s]
 99%|█████████▉| 367097/371472 [7:02:49<20:48,  3.50it/s]
 99%|█████████▉| 367098/371472 [7:02:50<21:26,  3.40it/s]
 99%|█████████▉| 367099/371472 [7:02:50<21:10,  3.44it/s]
 99%|█████████▉| 367100/371472 [7:02:50<20:33,  3.55it/s]
                                                         
{'loss': 2.6858, 'learning_rate': 1.1059815983968626e-07, 'epoch': 15.81}
+
 99%|█████████▉| 367100/371472 [7:02:50<20:33,  3.55it/s]
 99%|█████████▉| 367101/371472 [7:02:50<21:53,  3.33it/s]
 99%|█████████▉| 367102/371472 [7:02:51<21:50,  3.33it/s]
 99%|█████████▉| 367103/371472 [7:02:51<21:49,  3.34it/s]
 99%|█████████▉| 367104/371472 [7:02:51<21:39,  3.36it/s]
 99%|█████████▉| 367105/371472 [7:02:52<21:00,  3.46it/s]
 99%|█████████▉| 367106/371472 [7:02:52<20:15,  3.59it/s]
 99%|█████████▉| 367107/371472 [7:02:52<19:36,  3.71it/s]
 99%|█████████▉| 367108/371472 [7:02:52<19:56,  3.65it/s]
 99%|█████████▉| 367109/371472 [7:02:53<19:36,  3.71it/s]
 99%|█████████▉| 367110/371472 [7:02:53<19:41,  3.69it/s]
 99%|█████████▉| 367111/371472 [7:02:53<19:27,  3.73it/s]
 99%|█████████▉| 367112/371472 [7:02:53<19:55,  3.65it/s]
 99%|█████████▉| 367113/371472 [7:02:54<19:55,  3.65it/s]
 99%|█████████▉| 367114/371472 [7:02:54<20:06,  3.61it/s]
 99%|█████████▉| 367115/371472 [7:02:54<22:05,  3.29it/s]
 99%|█████████▉| 367116/371472 [7:02:55<21:22,  3.40it/s]
 99%|█████████▉| 367117/371472 [7:02:55<20:32,  3.53it/s]
 99%|█████████▉| 367118/371472 [7:02:55<21:58,  3.30it/s]
 99%|█████████▉| 367119/371472 [7:02:56<22:30,  3.22it/s]
 99%|█████████▉| 367120/371472 [7:02:56<21:03,  3.45it/s]
                                                         
{'loss': 2.5419, 'learning_rate': 1.1054967786420732e-07, 'epoch': 15.81}
+
 99%|█████████▉| 367120/371472 [7:02:56<21:03,  3.45it/s]
 99%|█████████▉| 367121/371472 [7:02:56<21:31,  3.37it/s]
 99%|█████████▉| 367122/371472 [7:02:56<21:14,  3.41it/s]
 99%|█████████▉| 367123/371472 [7:02:57<20:33,  3.53it/s]
 99%|█████████▉| 367124/371472 [7:02:57<20:05,  3.61it/s]
 99%|█████████▉| 367125/371472 [7:02:57<21:19,  3.40it/s]
 99%|█████████▉| 367126/371472 [7:02:58<21:23,  3.39it/s]
 99%|█████████▉| 367127/371472 [7:02:58<21:42,  3.33it/s]
 99%|█████████▉| 367128/371472 [7:02:58<21:40,  3.34it/s]
 99%|█████████▉| 367129/371472 [7:02:59<21:57,  3.30it/s]
 99%|█████████▉| 367130/371472 [7:02:59<21:57,  3.29it/s]
 99%|█████████▉| 367131/371472 [7:02:59<22:39,  3.19it/s]
 99%|█████████▉| 367132/371472 [7:02:59<21:18,  3.39it/s]
 99%|█████████▉| 367133/371472 [7:03:00<20:19,  3.56it/s]
 99%|█████████▉| 367134/371472 [7:03:00<19:57,  3.62it/s]
 99%|█████████▉| 367135/371472 [7:03:00<20:46,  3.48it/s]
 99%|█████████▉| 367136/371472 [7:03:01<20:03,  3.60it/s]
 99%|█████████▉| 367137/371472 [7:03:01<20:16,  3.56it/s]
 99%|█████████▉| 367138/371472 [7:03:01<20:08,  3.59it/s]
 99%|█████████▉| 367139/371472 [7:03:01<21:01,  3.43it/s]
 99%|█████████▉| 367140/371472 [7:03:02<20:14,  3.57it/s]
                                                         
{'loss': 2.7576, 'learning_rate': 1.1050119588872849e-07, 'epoch': 15.81}
+
 99%|█████████▉| 367140/371472 [7:03:02<20:14,  3.57it/s]
 99%|█████████▉| 367141/371472 [7:03:02<21:03,  3.43it/s]
 99%|█████████▉| 367142/371472 [7:03:02<22:16,  3.24it/s]
 99%|█████████▉| 367143/371472 [7:03:03<21:48,  3.31it/s]
 99%|█████████▉| 367144/371472 [7:03:03<20:51,  3.46it/s]
 99%|█████████▉| 367145/371472 [7:03:03<20:44,  3.48it/s]
 99%|█████████▉| 367146/371472 [7:03:03<20:13,  3.56it/s]
 99%|█████████▉| 367147/371472 [7:03:04<19:46,  3.65it/s]
 99%|█████████▉| 367148/371472 [7:03:04<19:56,  3.62it/s]
 99%|█████████▉| 367149/371472 [7:03:04<19:57,  3.61it/s]
 99%|█████████▉| 367150/371472 [7:03:05<19:57,  3.61it/s]
 99%|█████████▉| 367151/371472 [7:03:05<19:34,  3.68it/s]
 99%|█████████▉| 367152/371472 [7:03:05<19:37,  3.67it/s]
 99%|█████████▉| 367153/371472 [7:03:05<19:10,  3.75it/s]
 99%|█████████▉| 367154/371472 [7:03:06<19:59,  3.60it/s]
 99%|█████████▉| 367155/371472 [7:03:06<19:52,  3.62it/s]
 99%|█████████▉| 367156/371472 [7:03:06<20:30,  3.51it/s]
 99%|█████████▉| 367157/371472 [7:03:06<19:55,  3.61it/s]
 99%|█████████▉| 367158/371472 [7:03:07<19:18,  3.72it/s]
 99%|█████████▉| 367159/371472 [7:03:07<19:29,  3.69it/s]
 99%|█████████▉| 367160/371472 [7:03:07<19:01,  3.78it/s]
                                                         
{'loss': 2.5102, 'learning_rate': 1.1045271391324954e-07, 'epoch': 15.81}
+
 99%|█████████▉| 367160/371472 [7:03:07<19:01,  3.78it/s]
 99%|█████████▉| 367161/371472 [7:03:08<20:28,  3.51it/s]
 99%|█████████▉| 367162/371472 [7:03:08<20:42,  3.47it/s]
 99%|█████████▉| 367163/371472 [7:03:08<19:46,  3.63it/s]
 99%|█████████▉| 367164/371472 [7:03:08<19:38,  3.66it/s]
 99%|█████████▉| 367165/371472 [7:03:09<19:22,  3.70it/s]
 99%|█████████▉| 367166/371472 [7:03:09<19:31,  3.68it/s]
 99%|█████████▉| 367167/371472 [7:03:09<20:03,  3.58it/s]
 99%|█████████▉| 367168/371472 [7:03:10<23:04,  3.11it/s]
 99%|█████████▉| 367169/371472 [7:03:10<24:14,  2.96it/s]
 99%|█████████▉| 367170/371472 [7:03:10<26:22,  2.72it/s]
 99%|█████████▉| 367171/371472 [7:03:11<24:46,  2.89it/s]
 99%|█████████▉| 367172/371472 [7:03:11<22:35,  3.17it/s]
 99%|█████████▉| 367173/371472 [7:03:11<21:13,  3.38it/s]
 99%|█████████▉| 367174/371472 [7:03:11<20:43,  3.46it/s]
 99%|█████████▉| 367175/371472 [7:03:12<19:49,  3.61it/s]
 99%|█████████▉| 367176/371472 [7:03:12<20:22,  3.51it/s]
 99%|█████████▉| 367177/371472 [7:03:12<20:50,  3.44it/s]
 99%|█████████▉| 367178/371472 [7:03:13<20:30,  3.49it/s]
 99%|█████████▉| 367179/371472 [7:03:13<19:49,  3.61it/s]
 99%|█████████▉| 367180/371472 [7:03:13<21:02,  3.40it/s]
                                                         
{'loss': 2.4908, 'learning_rate': 1.104042319377707e-07, 'epoch': 15.82}
+
 99%|█████████▉| 367180/371472 [7:03:13<21:02,  3.40it/s]
 99%|█████████▉| 367181/371472 [7:03:13<20:35,  3.47it/s]
 99%|█████████▉| 367182/371472 [7:03:14<20:07,  3.55it/s]
 99%|█████████▉| 367183/371472 [7:03:14<19:46,  3.61it/s]
 99%|█████████▉| 367184/371472 [7:03:14<19:35,  3.65it/s]
 99%|█████████▉| 367185/371472 [7:03:15<19:04,  3.75it/s]
 99%|█████████▉| 367186/371472 [7:03:15<19:55,  3.58it/s]
 99%|█████████▉| 367187/371472 [7:03:15<21:27,  3.33it/s]
 99%|█████████▉| 367188/371472 [7:03:15<20:35,  3.47it/s]
 99%|█████████▉| 367189/371472 [7:03:16<19:30,  3.66it/s]
 99%|█████████▉| 367190/371472 [7:03:16<20:35,  3.47it/s]
 99%|█████████▉| 367191/371472 [7:03:16<21:10,  3.37it/s]
 99%|█████████▉| 367192/371472 [7:03:17<21:22,  3.34it/s]
 99%|█████████▉| 367193/371472 [7:03:17<20:49,  3.43it/s]
 99%|█████████▉| 367194/371472 [7:03:17<21:09,  3.37it/s]
 99%|█████████▉| 367195/371472 [7:03:17<20:28,  3.48it/s]
 99%|█████████▉| 367196/371472 [7:03:18<22:01,  3.23it/s]
 99%|█████████▉| 367197/371472 [7:03:18<23:22,  3.05it/s]
 99%|█████████▉| 367198/371472 [7:03:18<21:53,  3.25it/s]
 99%|█████████▉| 367199/371472 [7:03:19<21:14,  3.35it/s]
 99%|█████████▉| 367200/371472 [7:03:19<20:22,  3.49it/s]
                                                         
{'loss': 2.5595, 'learning_rate': 1.1035574996229175e-07, 'epoch': 15.82}
+
 99%|█████████▉| 367200/371472 [7:03:19<20:22,  3.49it/s]
 99%|█████████▉| 367201/371472 [7:03:19<20:06,  3.54it/s]
 99%|█████████▉| 367202/371472 [7:03:20<20:29,  3.47it/s]
 99%|█████████▉| 367203/371472 [7:03:20<20:04,  3.55it/s]
 99%|█████████▉| 367204/371472 [7:03:20<21:00,  3.39it/s]
 99%|█████████▉| 367205/371472 [7:03:20<20:20,  3.50it/s]
 99%|█████████▉| 367206/371472 [7:03:21<20:10,  3.52it/s]
 99%|█████████▉| 367207/371472 [7:03:21<19:13,  3.70it/s]
 99%|█████████▉| 367208/371472 [7:03:21<19:14,  3.69it/s]
 99%|█████████▉| 367209/371472 [7:03:22<19:20,  3.67it/s]
 99%|█████████▉| 367210/371472 [7:03:22<19:44,  3.60it/s]
 99%|█████████▉| 367211/371472 [7:03:22<20:36,  3.45it/s]
 99%|█████████▉| 367212/371472 [7:03:22<20:54,  3.40it/s]
 99%|█████████▉| 367213/371472 [7:03:23<20:10,  3.52it/s]
 99%|█████████▉| 367214/371472 [7:03:23<19:52,  3.57it/s]
 99%|█████████▉| 367215/371472 [7:03:23<20:44,  3.42it/s]
 99%|█████████▉| 367216/371472 [7:03:24<21:40,  3.27it/s]
 99%|█████████▉| 367217/371472 [7:03:24<21:15,  3.34it/s]
 99%|█████████▉| 367218/371472 [7:03:24<21:15,  3.33it/s]
 99%|█████████▉| 367219/371472 [7:03:24<20:23,  3.48it/s]
 99%|█████████▉| 367220/371472 [7:03:25<20:20,  3.48it/s]
                                                         
{'loss': 2.6025, 'learning_rate': 1.1030726798681292e-07, 'epoch': 15.82}
+
 99%|█████████▉| 367220/371472 [7:03:25<20:20,  3.48it/s]
 99%|█████████▉| 367221/371472 [7:03:25<20:38,  3.43it/s]
 99%|█████████▉| 367222/371472 [7:03:25<20:16,  3.49it/s]
 99%|█████████▉| 367223/371472 [7:03:26<20:22,  3.47it/s]
 99%|█████████▉| 367224/371472 [7:03:26<20:19,  3.48it/s]
 99%|█████████▉| 367225/371472 [7:03:26<21:05,  3.36it/s]
 99%|█████████▉| 367226/371472 [7:03:27<20:49,  3.40it/s]
 99%|█████████▉| 367227/371472 [7:03:27<20:59,  3.37it/s]
 99%|█████████▉| 367228/371472 [7:03:27<20:13,  3.50it/s]
 99%|█████████▉| 367229/371472 [7:03:27<19:48,  3.57it/s]
 99%|█████████▉| 367230/371472 [7:03:28<20:07,  3.51it/s]
 99%|█████████▉| 367231/371472 [7:03:28<19:53,  3.55it/s]
 99%|█████████▉| 367232/371472 [7:03:28<19:24,  3.64it/s]
 99%|█████████▉| 367233/371472 [7:03:28<20:38,  3.42it/s]
 99%|█████████▉| 367234/371472 [7:03:29<20:28,  3.45it/s]
 99%|█████████▉| 367235/371472 [7:03:29<20:01,  3.53it/s]
 99%|█████████▉| 367236/371472 [7:03:29<19:27,  3.63it/s]
 99%|█████████▉| 367237/371472 [7:03:30<19:01,  3.71it/s]
 99%|█████████▉| 367238/371472 [7:03:30<19:03,  3.70it/s]
 99%|█████████▉| 367239/371472 [7:03:30<19:29,  3.62it/s]
 99%|█████████▉| 367240/371472 [7:03:30<19:53,  3.55it/s]
                                                         
{'loss': 2.7819, 'learning_rate': 1.1025878601133398e-07, 'epoch': 15.82}
+
 99%|█████████▉| 367240/371472 [7:03:30<19:53,  3.55it/s]
 99%|█████████▉| 367241/371472 [7:03:31<19:31,  3.61it/s]
 99%|█████████▉| 367242/371472 [7:03:31<19:32,  3.61it/s]
 99%|██████��██▉| 367243/371472 [7:03:31<19:32,  3.61it/s]
 99%|█████████▉| 367244/371472 [7:03:32<19:26,  3.62it/s]
 99%|█████████▉| 367245/371472 [7:03:32<20:58,  3.36it/s]
 99%|█████████▉| 367246/371472 [7:03:32<20:19,  3.47it/s]
 99%|█████████▉| 367247/371472 [7:03:32<20:16,  3.47it/s]
 99%|█████████▉| 367248/371472 [7:03:33<19:55,  3.53it/s]
 99%|█████████▉| 367249/371472 [7:03:33<19:31,  3.60it/s]
 99%|█████████▉| 367250/371472 [7:03:33<19:26,  3.62it/s]
 99%|█████████▉| 367251/371472 [7:03:33<19:01,  3.70it/s]
 99%|█████████▉| 367252/371472 [7:03:34<18:32,  3.79it/s]
 99%|█████████▉| 367253/371472 [7:03:34<19:41,  3.57it/s]
 99%|█████████▉| 367254/371472 [7:03:34<19:26,  3.62it/s]
 99%|█████████▉| 367255/371472 [7:03:35<19:00,  3.70it/s]
 99%|█████████▉| 367256/371472 [7:03:35<19:29,  3.60it/s]
 99%|█████████▉| 367257/371472 [7:03:35<19:11,  3.66it/s]
 99%|█████████▉| 367258/371472 [7:03:35<19:15,  3.65it/s]
 99%|█████████▉| 367259/371472 [7:03:36<18:45,  3.74it/s]
 99%|█████████▉| 367260/371472 [7:03:36<18:51,  3.72it/s]
                                                         
{'loss': 2.6057, 'learning_rate': 1.1021030403585513e-07, 'epoch': 15.82}
+
 99%|█████████▉| 367260/371472 [7:03:36<18:51,  3.72it/s]
 99%|█████████▉| 367261/371472 [7:03:36<19:28,  3.60it/s]
 99%|█████████▉| 367262/371472 [7:03:36<19:20,  3.63it/s]
 99%|█████████▉| 367263/371472 [7:03:37<19:18,  3.63it/s]
 99%|█████████▉| 367264/371472 [7:03:37<20:04,  3.49it/s]
 99%|█████████▉| 367265/371472 [7:03:37<20:46,  3.38it/s]
 99%|█████████▉| 367266/371472 [7:03:38<20:55,  3.35it/s]
 99%|█████████▉| 367267/371472 [7:03:38<20:56,  3.35it/s]
 99%|█████████▉| 367268/371472 [7:03:38<20:20,  3.44it/s]
 99%|█████████▉| 367269/371472 [7:03:39<19:45,  3.55it/s]
 99%|█████████▉| 367270/371472 [7:03:39<22:46,  3.08it/s]
 99%|█████████▉| 367271/371472 [7:03:39<23:10,  3.02it/s]
 99%|█████████▉| 367272/371472 [7:03:40<21:57,  3.19it/s]
 99%|█████████▉| 367273/371472 [7:03:40<21:00,  3.33it/s]
 99%|█████████▉| 367274/371472 [7:03:40<20:35,  3.40it/s]
 99%|█████████▉| 367275/371472 [7:03:40<21:43,  3.22it/s]
 99%|█████████▉| 367276/371472 [7:03:41<20:47,  3.36it/s]
 99%|█████████▉| 367277/371472 [7:03:41<20:15,  3.45it/s]
 99%|█████████▉| 367278/371472 [7:03:41<20:22,  3.43it/s]
 99%|█████████▉| 367279/371472 [7:03:42<19:59,  3.50it/s]
 99%|█████████▉| 367280/371472 [7:03:42<19:44,  3.54it/s]
                                                         
{'loss': 2.5239, 'learning_rate': 1.1016182206037617e-07, 'epoch': 15.82}
+
 99%|█████████▉| 367280/371472 [7:03:42<19:44,  3.54it/s]
 99%|█████████▉| 367281/371472 [7:03:42<20:34,  3.40it/s]
 99%|█████████▉| 367282/371472 [7:03:43<25:30,  2.74it/s]
 99%|█████████▉| 367283/371472 [7:03:43<23:26,  2.98it/s]
 99%|█████████▉| 367284/371472 [7:03:43<22:54,  3.05it/s]
 99%|█████████▉| 367285/371472 [7:03:44<21:53,  3.19it/s]
 99%|█████████▉| 367286/371472 [7:03:44<21:00,  3.32it/s]
 99%|█████████▉| 367287/371472 [7:03:44<20:38,  3.38it/s]
 99%|█████████▉| 367288/371472 [7:03:44<20:51,  3.34it/s]
 99%|█████████▉| 367289/371472 [7:03:45<20:23,  3.42it/s]
 99%|█████████▉| 367290/371472 [7:03:45<20:46,  3.35it/s]
 99%|█████████▉| 367291/371472 [7:03:45<20:37,  3.38it/s]
 99%|█████████▉| 367292/371472 [7:03:46<22:26,  3.10it/s]
 99%|█████████▉| 367293/371472 [7:03:46<21:27,  3.24it/s]
 99%|█████████▉| 367294/371472 [7:03:46<21:30,  3.24it/s]
 99%|█████████▉| 367295/371472 [7:03:47<20:39,  3.37it/s]
 99%|█████████▉| 367296/371472 [7:03:47<19:39,  3.54it/s]
 99%|█████████▉| 367297/371472 [7:03:47<19:44,  3.52it/s]
 99%|█████████▉| 367298/371472 [7:03:47<19:48,  3.51it/s]
 99%|█████████▉| 367299/371472 [7:03:48<19:30,  3.57it/s]
 99%|█████████▉| 367300/371472 [7:03:48<19:39,  3.54it/s]
                                                         
{'loss': 2.5051, 'learning_rate': 1.1011334008489735e-07, 'epoch': 15.82}
+
 99%|█████████▉| 367300/371472 [7:03:48<19:39,  3.54it/s]
 99%|█████████▉| 367301/371472 [7:03:48<20:05,  3.46it/s]
 99%|█████████▉| 367302/371472 [7:03:49<20:35,  3.38it/s]
 99%|█████████▉| 367303/371472 [7:03:49<20:37,  3.37it/s]
 99%|█████████▉| 367304/371472 [7:03:49<20:12,  3.44it/s]
 99%|█████████▉| 367305/371472 [7:03:49<20:20,  3.41it/s]
 99%|█████████▉| 367306/371472 [7:03:50<20:56,  3.32it/s]
 99%|█████████▉| 367307/371472 [7:03:50<21:52,  3.17it/s]
 99%|█████████▉| 367308/371472 [7:03:50<23:05,  3.00it/s]
 99%|█████████▉| 367309/371472 [7:03:51<22:15,  3.12it/s]
 99%|█████████▉| 367310/371472 [7:03:51<22:10,  3.13it/s]
 99%|█████████▉| 367311/371472 [7:03:51<21:10,  3.27it/s]
 99%|█████████▉| 367312/371472 [7:03:52<20:37,  3.36it/s]
 99%|█████████▉| 367313/371472 [7:03:52<20:01,  3.46it/s]
 99%|█████████▉| 367314/371472 [7:03:52<20:15,  3.42it/s]
 99%|█████████▉| 367315/371472 [7:03:52<19:38,  3.53it/s]
 99%|█████████▉| 367316/371472 [7:03:53<19:03,  3.63it/s]
 99%|█████████▉| 367317/371472 [7:03:53<19:05,  3.63it/s]
 99%|█████████▉| 367318/371472 [7:03:53<19:03,  3.63it/s]
 99%|█████████▉| 367319/371472 [7:03:54<19:42,  3.51it/s]
 99%|█████████▉| 367320/371472 [7:03:54<19:03,  3.63it/s]
                                                         
{'loss': 2.8137, 'learning_rate': 1.1006485810941841e-07, 'epoch': 15.82}
+
 99%|█████████▉| 367320/371472 [7:03:54<19:03,  3.63it/s]
 99%|█████████▉| 367321/371472 [7:03:54<19:02,  3.63it/s]
 99%|█████████▉| 367322/371472 [7:03:54<20:33,  3.36it/s]
 99%|█████████▉| 367323/371472 [7:03:55<20:04,  3.44it/s]
 99%|█████████▉| 367324/371472 [7:03:55<19:18,  3.58it/s]
 99%|█████████▉| 367325/371472 [7:03:55<19:11,  3.60it/s]
 99%|█████████▉| 367326/371472 [7:03:56<21:05,  3.28it/s]
 99%|█████████▉| 367327/371472 [7:03:56<20:07,  3.43it/s]
 99%|█████████▉| 367328/371472 [7:03:56<19:53,  3.47it/s]
 99%|█████████▉| 367329/371472 [7:03:56<19:25,  3.56it/s]
 99%|█████████▉| 367330/371472 [7:03:57<19:19,  3.57it/s]
 99%|█████████▉| 367331/371472 [7:03:57<19:00,  3.63it/s]
 99%|█████████▉| 367332/371472 [7:03:57<19:30,  3.54it/s]
 99%|█████████▉| 367333/371472 [7:03:58<20:42,  3.33it/s]
 99%|█████████▉| 367334/371472 [7:03:58<20:46,  3.32it/s]
 99%|█████████▉| 367335/371472 [7:03:58<20:23,  3.38it/s]
 99%|█████████▉| 367336/371472 [7:03:58<20:14,  3.41it/s]
 99%|█████████▉| 367337/371472 [7:03:59<19:32,  3.53it/s]
 99%|█████████▉| 367338/371472 [7:03:59<20:17,  3.40it/s]
 99%|█████████▉| 367339/371472 [7:03:59<20:15,  3.40it/s]
 99%|█████████▉| 367340/371472 [7:04:00<20:40,  3.33it/s]
                                                         
{'loss': 2.769, 'learning_rate': 1.1001637613393956e-07, 'epoch': 15.82}
+
 99%|█████████▉| 367340/371472 [7:04:00<20:40,  3.33it/s]
 99%|█████████▉| 367341/371472 [7:04:00<20:26,  3.37it/s]
 99%|█████████▉| 367342/371472 [7:04:00<19:50,  3.47it/s]
 99%|█████████▉| 367343/371472 [7:04:00<19:19,  3.56it/s]
 99%|█████████▉| 367344/371472 [7:04:01<18:43,  3.67it/s]
 99%|█████████▉| 367345/371472 [7:04:01<18:29,  3.72it/s]
 99%|█████████▉| 367346/371472 [7:04:01<20:35,  3.34it/s]
 99%|█████████▉| 367347/371472 [7:04:02<20:13,  3.40it/s]
 99%|█████████▉| 367348/371472 [7:04:02<19:32,  3.52it/s]
 99%|█████████▉| 367349/371472 [7:04:02<19:15,  3.57it/s]
 99%|█████████▉| 367350/371472 [7:04:03<22:09,  3.10it/s]
 99%|█████████▉| 367351/371472 [7:04:03<22:15,  3.09it/s]
 99%|█████████▉| 367352/371472 [7:04:03<21:12,  3.24it/s]
 99%|█████████▉| 367353/371472 [7:04:03<20:27,  3.36it/s]
 99%|█████████▉| 367354/371472 [7:04:04<20:54,  3.28it/s]
 99%|█████████▉| 367355/371472 [7:04:04<20:00,  3.43it/s]
 99%|█████████▉| 367356/371472 [7:04:04<20:04,  3.42it/s]
 99%|█████████▉| 367357/371472 [7:04:05<19:20,  3.55it/s]
 99%|█████████▉| 367358/371472 [7:04:05<19:00,  3.61it/s]
 99%|█████████▉| 367359/371472 [7:04:05<18:56,  3.62it/s]
 99%|█████████▉| 367360/371472 [7:04:06<20:29,  3.35it/s]
                                                         
{'loss': 2.5455, 'learning_rate': 1.0996789415846062e-07, 'epoch': 15.82}
+
 99%|█████████▉| 367360/371472 [7:04:06<20:29,  3.35it/s]
 99%|█████████▉| 367361/371472 [7:04:06<20:39,  3.32it/s]
 99%|█████████▉| 367362/371472 [7:04:06<21:23,  3.20it/s]
 99%|█████████▉| 367363/371472 [7:04:06<20:36,  3.32it/s]
 99%|█████████▉| 367364/371472 [7:04:07<20:16,  3.38it/s]
 99%|█████████▉| 367365/371472 [7:04:07<20:04,  3.41it/s]
 99%|█████████▉| 367366/371472 [7:04:07<19:22,  3.53it/s]
 99%|█████████▉| 367367/371472 [7:04:08<20:48,  3.29it/s]
 99%|█████████▉| 367368/371472 [7:04:08<20:09,  3.39it/s]
 99%|█████████▉| 367369/371472 [7:04:08<20:50,  3.28it/s]
 99%|█████████▉| 367370/371472 [7:04:08<20:13,  3.38it/s]
 99%|█████████▉| 367371/371472 [7:04:09<22:31,  3.03it/s]
 99%|█████████▉| 367372/371472 [7:04:09<21:12,  3.22it/s]
 99%|█████████▉| 367373/371472 [7:04:09<20:38,  3.31it/s]
 99%|█████████▉| 367374/371472 [7:04:10<20:26,  3.34it/s]
 99%|█████████▉| 367375/371472 [7:04:10<20:00,  3.41it/s]
 99%|█████████▉| 367376/371472 [7:04:10<19:33,  3.49it/s]
 99%|█████████▉| 367377/371472 [7:04:11<18:52,  3.62it/s]
 99%|█████████▉| 367378/371472 [7:04:11<18:28,  3.69it/s]
 99%|█████████▉| 367379/371472 [7:04:11<17:58,  3.79it/s]
 99%|█████████▉| 367380/371472 [7:04:11<18:14,  3.74it/s]
                                                         
{'loss': 2.5242, 'learning_rate': 1.0991941218298178e-07, 'epoch': 15.82}
+
 99%|█████████▉| 367380/371472 [7:04:11<18:14,  3.74it/s]
 99%|█████████▉| 367381/371472 [7:04:12<19:13,  3.55it/s]
 99%|█████████▉| 367382/371472 [7:04:12<19:28,  3.50it/s]
 99%|█████████▉| 367383/371472 [7:04:12<18:50,  3.62it/s]
 99%|█████████▉| 367384/371472 [7:04:13<19:58,  3.41it/s]
 99%|█████████▉| 367385/371472 [7:04:13<19:23,  3.51it/s]
 99%|█████████▉| 367386/371472 [7:04:13<19:43,  3.45it/s]
 99%|█████████▉| 367387/371472 [7:04:13<19:03,  3.57it/s]
 99%|█████████▉| 367388/371472 [7:04:14<18:59,  3.58it/s]
 99%|█████████▉| 367389/371472 [7:04:14<18:20,  3.71it/s]
 99%|█████████▉| 367390/371472 [7:04:14<17:59,  3.78it/s]
 99%|█████████▉| 367391/371472 [7:04:14<18:22,  3.70it/s]
 99%|█████████▉| 367392/371472 [7:04:15<18:20,  3.71it/s]
 99%|█████████▉| 367393/371472 [7:04:15<18:36,  3.65it/s]
 99%|█████████▉| 367394/371472 [7:04:15<18:32,  3.67it/s]
 99%|█████████▉| 367395/371472 [7:04:15<18:35,  3.66it/s]
 99%|█████████▉| 367396/371472 [7:04:16<18:12,  3.73it/s]
 99%|█████████▉| 367397/371472 [7:04:16<19:03,  3.56it/s]
 99%|█████████▉| 367398/371472 [7:04:16<20:49,  3.26it/s]
 99%|█████████▉| 367399/371472 [7:04:17<19:34,  3.47it/s]
 99%|█████████▉| 367400/371472 [7:04:17<21:45,  3.12it/s]
                                                         
{'loss': 2.668, 'learning_rate': 1.0987093020750283e-07, 'epoch': 15.82}
+
 99%|█████████▉| 367400/371472 [7:04:17<21:45,  3.12it/s]
 99%|█████████▉| 367401/371472 [7:04:17<21:08,  3.21it/s]
 99%|█████████▉| 367402/371472 [7:04:18<20:29,  3.31it/s]
 99%|█████████▉| 367403/371472 [7:04:18<19:50,  3.42it/s]
 99%|█████████▉| 367404/371472 [7:04:18<19:07,  3.54it/s]
 99%|█████████▉| 367405/371472 [7:04:18<18:32,  3.66it/s]
 99%|█████████▉| 367406/371472 [7:04:19<20:45,  3.26it/s]
 99%|█████████▉| 367407/371472 [7:04:19<19:39,  3.45it/s]
 99%|█████████▉| 367408/371472 [7:04:19<19:39,  3.45it/s]
 99%|█████████▉| 367409/371472 [7:04:20<20:53,  3.24it/s]
 99%|█████████▉| 367410/371472 [7:04:20<20:06,  3.37it/s]
 99%|█████████▉| 367411/371472 [7:04:20<19:37,  3.45it/s]
 99%|█████████▉| 367412/371472 [7:04:21<19:39,  3.44it/s]
 99%|█████████▉| 367413/371472 [7:04:21<19:03,  3.55it/s]
 99%|█████████▉| 367414/371472 [7:04:21<19:14,  3.51it/s]
 99%|█████████▉| 367415/371472 [7:04:21<19:20,  3.50it/s]
 99%|█████████▉| 367416/371472 [7:04:22<19:11,  3.52it/s]
 99%|█████████▉| 367417/371472 [7:04:22<18:41,  3.62it/s]
 99%|█████████▉| 367418/371472 [7:04:22<18:53,  3.58it/s]
 99%|█████████▉| 367419/371472 [7:04:22<18:39,  3.62it/s]
 99%|█████████▉| 367420/371472 [7:04:23<18:22,  3.68it/s]
                                                         
{'loss': 2.5615, 'learning_rate': 1.0982244823202399e-07, 'epoch': 15.83}
+
 99%|█████████▉| 367420/371472 [7:04:23<18:22,  3.68it/s]
 99%|█████████▉| 367421/371472 [7:04:23<18:19,  3.68it/s]
 99%|█████████▉| 367422/371472 [7:04:23<18:29,  3.65it/s]
 99%|█████████▉| 367423/371472 [7:04:24<18:19,  3.68it/s]
 99%|█████████▉| 367424/371472 [7:04:24<17:56,  3.76it/s]
 99%|█████████▉| 367425/371472 [7:04:24<17:59,  3.75it/s]
 99%|█████████▉| 367426/371472 [7:04:24<18:41,  3.61it/s]
 99%|█████████▉| 367427/371472 [7:04:25<18:35,  3.63it/s]
 99%|█████████▉| 367428/371472 [7:04:25<18:37,  3.62it/s]
 99%|█████████▉| 367429/371472 [7:04:25<18:57,  3.55it/s]
 99%|█████████▉| 367430/371472 [7:04:25<18:51,  3.57it/s]
 99%|█████████▉| 367431/371472 [7:04:26<19:19,  3.49it/s]
 99%|█████████▉| 367432/371472 [7:04:26<19:46,  3.41it/s]
 99%|█████████▉| 367433/371472 [7:04:26<19:22,  3.47it/s]
 99%|█████████▉| 367434/371472 [7:04:27<19:54,  3.38it/s]
 99%|█████████▉| 367435/371472 [7:04:27<19:22,  3.47it/s]
 99%|█████████▉| 367436/371472 [7:04:27<19:07,  3.52it/s]
 99%|█████████▉| 367437/371472 [7:04:28<20:18,  3.31it/s]
 99%|█████████▉| 367438/371472 [7:04:28<19:40,  3.42it/s]
 99%|█████████▉| 367439/371472 [7:04:28<19:07,  3.51it/s]
 99%|█████████▉| 367440/371472 [7:04:28<19:44,  3.40it/s]
                                                         
{'loss': 2.492, 'learning_rate': 1.0977396625654505e-07, 'epoch': 15.83}
+
 99%|█████████▉| 367440/371472 [7:04:28<19:44,  3.40it/s]
 99%|█████████▉| 367441/371472 [7:04:29<19:21,  3.47it/s]
 99%|█████████▉| 367442/371472 [7:04:29<18:39,  3.60it/s]
 99%|█████████▉| 367443/371472 [7:04:29<19:09,  3.51it/s]
 99%|█████████▉| 367444/371472 [7:04:30<18:47,  3.57it/s]
 99%|█████████▉| 367445/371472 [7:04:30<18:33,  3.62it/s]
 99%|█████████▉| 367446/371472 [7:04:30<19:43,  3.40it/s]
 99%|█████████▉| 367447/371472 [7:04:30<18:58,  3.53it/s]
 99%|█████████▉| 367448/371472 [7:04:31<19:33,  3.43it/s]
 99%|█████████▉| 367449/371472 [7:04:31<19:03,  3.52it/s]
 99%|█████████▉| 367450/371472 [7:04:31<19:06,  3.51it/s]
 99%|█████████▉| 367451/371472 [7:04:32<18:19,  3.66it/s]
 99%|█████████▉| 367452/371472 [7:04:32<19:54,  3.37it/s]
 99%|█████████▉| 367453/371472 [7:04:32<19:05,  3.51it/s]
 99%|█████████▉| 367454/371472 [7:04:32<18:30,  3.62it/s]
 99%|█████████▉| 367455/371472 [7:04:33<19:17,  3.47it/s]
 99%|█████████▉| 367456/371472 [7:04:33<18:49,  3.56it/s]
 99%|█████████▉| 367457/371472 [7:04:33<19:02,  3.51it/s]
 99%|█████████▉| 367458/371472 [7:04:34<19:25,  3.44it/s]
 99%|█████████▉| 367459/371472 [7:04:34<19:20,  3.46it/s]
 99%|█████████▉| 367460/371472 [7:04:34<20:08,  3.32it/s]
                                                         
{'loss': 2.5693, 'learning_rate': 1.0972548428106621e-07, 'epoch': 15.83}
+
 99%|█████████▉| 367460/371472 [7:04:34<20:08,  3.32it/s]
 99%|█████████▉| 367461/371472 [7:04:34<19:25,  3.44it/s]
 99%|█████████▉| 367462/371472 [7:04:35<21:22,  3.13it/s]
 99%|█████████▉| 367463/371472 [7:04:35<20:15,  3.30it/s]
 99%|█████████▉| 367464/371472 [7:04:35<20:26,  3.27it/s]
 99%|█████████▉| 367465/371472 [7:04:36<19:38,  3.40it/s]
 99%|█████████▉| 367466/371472 [7:04:36<19:38,  3.40it/s]
 99%|█████████▉| 367467/371472 [7:04:36<19:36,  3.40it/s]
 99%|█████████▉| 367468/371472 [7:04:37<19:02,  3.50it/s]
 99%|█████████▉| 367469/371472 [7:04:37<19:36,  3.40it/s]
 99%|█████████▉| 367470/371472 [7:04:37<19:13,  3.47it/s]
 99%|█████████▉| 367471/371472 [7:04:37<19:26,  3.43it/s]
 99%|█████████▉| 367472/371472 [7:04:38<18:34,  3.59it/s]
 99%|█████████▉| 367473/371472 [7:04:38<23:03,  2.89it/s]
 99%|█████████▉| 367474/371472 [7:04:38<21:02,  3.17it/s]
 99%|█████████▉| 367475/371472 [7:04:39<20:53,  3.19it/s]
 99%|█████████▉| 367476/371472 [7:04:39<19:39,  3.39it/s]
 99%|█████████▉| 367477/371472 [7:04:39<19:15,  3.46it/s]
 99%|█████████▉| 367478/371472 [7:04:40<18:50,  3.53it/s]
 99%|█████████▉| 367479/371472 [7:04:40<19:22,  3.44it/s]
 99%|█████████▉| 367480/371472 [7:04:40<18:57,  3.51it/s]
                                                         
{'loss': 2.5658, 'learning_rate': 1.0967700230558727e-07, 'epoch': 15.83}
+
 99%|█████████▉| 367480/371472 [7:04:40<18:57,  3.51it/s]
 99%|█████████▉| 367481/371472 [7:04:40<19:09,  3.47it/s]
 99%|█████████▉| 367482/371472 [7:04:41<20:36,  3.23it/s]
 99%|█████████▉| 367483/371472 [7:04:41<19:26,  3.42it/s]
 99%|█████████▉| 367484/371472 [7:04:41<19:17,  3.45it/s]
 99%|█████████▉| 367485/371472 [7:04:42<18:40,  3.56it/s]
 99%|█████████▉| 367486/371472 [7:04:42<18:03,  3.68it/s]
 99%|█████████▉| 367487/371472 [7:04:42<17:33,  3.78it/s]
 99%|█████████▉| 367488/371472 [7:04:42<18:09,  3.66it/s]
 99%|█████████▉| 367489/371472 [7:04:43<17:59,  3.69it/s]
 99%|█████████▉| 367490/371472 [7:04:43<17:37,  3.77it/s]
 99%|█████████▉| 367491/371472 [7:04:43<17:19,  3.83it/s]
 99%|█████████▉| 367492/371472 [7:04:43<17:20,  3.83it/s]
 99%|█████████▉| 367493/371472 [7:04:44<17:22,  3.82it/s]
 99%|█████████▉| 367494/371472 [7:04:44<18:24,  3.60it/s]
 99%|█████████▉| 367495/371472 [7:04:44<19:22,  3.42it/s]
 99%|█████████▉| 367496/371472 [7:04:45<19:43,  3.36it/s]
 99%|█████████▉| 367497/371472 [7:04:45<18:51,  3.51it/s]
 99%|█████████▉| 367498/371472 [7:04:45<19:38,  3.37it/s]
 99%|█████████▉| 367499/371472 [7:04:45<19:17,  3.43it/s]
 99%|█████████▉| 367500/371472 [7:04:46<18:57,  3.49it/s]
                                                         
{'loss': 2.6753, 'learning_rate': 1.0962852033010842e-07, 'epoch': 15.83}
+
 99%|█████████▉| 367500/371472 [7:04:46<18:57,  3.49it/s]
 99%|█████████▉| 367501/371472 [7:04:46<18:43,  3.53it/s]
 99%|█████████▉| 367502/371472 [7:04:46<18:35,  3.56it/s]
 99%|█████████▉| 367503/371472 [7:04:47<19:20,  3.42it/s]
 99%|█████████▉| 367504/371472 [7:04:47<18:41,  3.54it/s]
 99%|█████████▉| 367505/371472 [7:04:47<18:52,  3.50it/s]
 99%|█████████▉| 367506/371472 [7:04:47<18:17,  3.61it/s]
 99%|█████████▉| 367507/371472 [7:04:48<18:03,  3.66it/s]
 99%|█████████▉| 367508/371472 [7:04:48<17:59,  3.67it/s]
 99%|█████████▉| 367509/371472 [7:04:48<18:00,  3.67it/s]
 99%|█████████▉| 367510/371472 [7:04:48<17:43,  3.72it/s]
 99%|█████████▉| 367511/371472 [7:04:49<17:48,  3.71it/s]
 99%|█████████▉| 367512/371472 [7:04:49<17:41,  3.73it/s]
 99%|█████████▉| 367513/371472 [7:04:49<17:42,  3.73it/s]
 99%|█████████▉| 367514/371472 [7:04:50<17:28,  3.77it/s]
 99%|█████████▉| 367515/371472 [7:04:50<17:56,  3.68it/s]
 99%|█████████▉| 367516/371472 [7:04:50<19:02,  3.46it/s]
 99%|█████████▉| 367517/371472 [7:04:50<18:44,  3.52it/s]
 99%|█████████▉| 367518/371472 [7:04:51<18:34,  3.55it/s]
 99%|█████████▉| 367519/371472 [7:04:51<18:18,  3.60it/s]
 99%|█████████▉| 367520/371472 [7:04:51<18:18,  3.60it/s]
                                                         
{'loss': 2.4462, 'learning_rate': 1.0958003835462948e-07, 'epoch': 15.83}
+
 99%|█████████▉| 367520/371472 [7:04:51<18:18,  3.60it/s]
 99%|█████████▉| 367521/371472 [7:04:52<18:10,  3.62it/s]
 99%|█████████▉| 367522/371472 [7:04:52<17:55,  3.67it/s]
 99%|█████████▉| 367523/371472 [7:04:52<19:07,  3.44it/s]
 99%|█████████▉| 367524/371472 [7:04:52<19:30,  3.37it/s]
 99%|█████████▉| 367525/371472 [7:04:53<19:26,  3.38it/s]
 99%|█████████▉| 367526/371472 [7:04:53<19:28,  3.38it/s]
 99%|█████████▉| 367527/371472 [7:04:53<19:06,  3.44it/s]
 99%|█████████▉| 367528/371472 [7:04:54<18:56,  3.47it/s]
 99%|█████████▉| 367529/371472 [7:04:54<19:13,  3.42it/s]
 99%|█████████▉| 367530/371472 [7:04:54<18:40,  3.52it/s]
 99%|█████████▉| 367531/371472 [7:04:54<18:11,  3.61it/s]
 99%|█████████▉| 367532/371472 [7:04:55<18:06,  3.63it/s]
 99%|█████████▉| 367533/371472 [7:04:55<19:13,  3.41it/s]
 99%|█████████▉| 367534/371472 [7:04:55<19:24,  3.38it/s]
 99%|█████████▉| 367535/371472 [7:04:56<18:37,  3.52it/s]
 99%|█████████▉| 367536/371472 [7:04:56<18:22,  3.57it/s]
 99%|█████████▉| 367537/371472 [7:04:56<18:31,  3.54it/s]
 99%|█████████▉| 367538/371472 [7:04:56<18:16,  3.59it/s]
 99%|█████████▉| 367539/371472 [7:04:57<18:04,  3.63it/s]
 99%|█████████▉| 367540/371472 [7:04:57<17:33,  3.73it/s]
                                                         
{'loss': 2.6037, 'learning_rate': 1.0953155637915064e-07, 'epoch': 15.83}
+
 99%|█████████▉| 367540/371472 [7:04:57<17:33,  3.73it/s]
 99%|█████████▉| 367541/371472 [7:04:57<18:34,  3.53it/s]
 99%|█████████▉| 367542/371472 [7:04:57<18:08,  3.61it/s]
 99%|█████████▉| 367543/371472 [7:04:58<17:54,  3.66it/s]
 99%|█████████▉| 367544/371472 [7:04:58<18:41,  3.50it/s]
 99%|█████████▉| 367545/371472 [7:04:58<19:57,  3.28it/s]
 99%|█████████▉| 367546/371472 [7:04:59<18:55,  3.46it/s]
 99%|█████████▉| 367547/371472 [7:04:59<19:42,  3.32it/s]
 99%|█████████▉| 367548/371472 [7:04:59<20:10,  3.24it/s]
 99%|█████████▉| 367549/371472 [7:05:00<20:09,  3.24it/s]
 99%|█████████▉| 367550/371472 [7:05:00<20:52,  3.13it/s]
 99%|█████████▉| 367551/371472 [7:05:00<19:25,  3.36it/s]
 99%|█████████▉| 367552/371472 [7:05:00<18:48,  3.47it/s]
 99%|█████████▉| 367553/371472 [7:05:01<19:51,  3.29it/s]
 99%|█████████▉| 367554/371472 [7:05:01<19:18,  3.38it/s]
 99%|█████████▉| 367555/371472 [7:05:01<18:32,  3.52it/s]
 99%|█████████▉| 367556/371472 [7:05:02<18:06,  3.60it/s]
 99%|█████████▉| 367557/371472 [7:05:02<18:13,  3.58it/s]
 99%|█████████▉| 367558/371472 [7:05:02<18:50,  3.46it/s]
 99%|█████████▉| 367559/371472 [7:05:02<18:40,  3.49it/s]
 99%|█████████▉| 367560/371472 [7:05:03<18:08,  3.59it/s]
                                                         
{'loss': 2.7087, 'learning_rate': 1.0948307440367169e-07, 'epoch': 15.83}
+
 99%|█████████▉| 367560/371472 [7:05:03<18:08,  3.59it/s]
 99%|█████████▉| 367561/371472 [7:05:03<17:56,  3.63it/s]
 99%|█████████▉| 367562/371472 [7:05:03<17:55,  3.64it/s]
 99%|█████████▉| 367563/371472 [7:05:04<17:47,  3.66it/s]
 99%|█████████▉| 367564/371472 [7:05:04<17:45,  3.67it/s]
 99%|█████████▉| 367565/371472 [7:05:04<20:07,  3.24it/s]
 99%|█████████▉| 367566/371472 [7:05:04<18:55,  3.44it/s]
 99%|█████████▉| 367567/371472 [7:05:05<20:14,  3.21it/s]
 99%|█████████▉| 367568/371472 [7:05:05<19:55,  3.26it/s]
 99%|█████████▉| 367569/371472 [7:05:05<19:32,  3.33it/s]
 99%|█████████▉| 367570/371472 [7:05:06<18:42,  3.48it/s]
 99%|█████████▉| 367571/371472 [7:05:06<19:12,  3.38it/s]
 99%|█████████▉| 367572/371472 [7:05:06<20:31,  3.17it/s]
 99%|█████████▉| 367573/371472 [7:05:07<19:27,  3.34it/s]
 99%|█████████▉| 367574/371472 [7:05:07<19:51,  3.27it/s]
 99%|█████████▉| 367575/371472 [7:05:07<19:01,  3.42it/s]
 99%|█████████▉| 367576/371472 [7:05:07<18:58,  3.42it/s]
 99%|█████████▉| 367577/371472 [7:05:08<18:47,  3.46it/s]
 99%|█████████▉| 367578/371472 [7:05:08<19:16,  3.37it/s]
 99%|█████████▉| 367579/371472 [7:05:08<19:03,  3.40it/s]
 99%|█████████▉| 367580/371472 [7:05:09<19:14,  3.37it/s]
                                                         
{'loss': 2.5989, 'learning_rate': 1.0943459242819285e-07, 'epoch': 15.83}
+
 99%|█████████▉| 367580/371472 [7:05:09<19:14,  3.37it/s]
 99%|█████████▉| 367581/371472 [7:05:09<19:28,  3.33it/s]
 99%|█████████▉| 367582/371472 [7:05:09<18:59,  3.41it/s]
 99%|█████████▉| 367583/371472 [7:05:10<19:54,  3.25it/s]
 99%|█████████▉| 367584/371472 [7:05:10<21:59,  2.95it/s]
 99%|█████████▉| 367585/371472 [7:05:10<20:24,  3.18it/s]
 99%|█████████▉| 367586/371472 [7:05:11<19:23,  3.34it/s]
 99%|█████████▉| 367587/371472 [7:05:11<18:29,  3.50it/s]
 99%|█████████▉| 367588/371472 [7:05:11<18:48,  3.44it/s]
 99%|█████████▉| 367589/371472 [7:05:11<18:41,  3.46it/s]
 99%|█████████▉| 367590/371472 [7:05:12<18:13,  3.55it/s]
 99%|█████████▉| 367591/371472 [7:05:12<17:29,  3.70it/s]
 99%|█████████▉| 367592/371472 [7:05:12<17:12,  3.76it/s]
 99%|█████████▉| 367593/371472 [7:05:12<17:52,  3.62it/s]
 99%|█████████▉| 367594/371472 [7:05:13<17:53,  3.61it/s]
 99%|█████████▉| 367595/371472 [7:05:13<17:39,  3.66it/s]
 99%|█████████▉| 367596/371472 [7:05:13<17:36,  3.67it/s]
 99%|█████████▉| 367597/371472 [7:05:14<18:35,  3.47it/s]
 99%|█████████▉| 367598/371472 [7:05:14<18:06,  3.57it/s]
 99%|█████████▉| 367599/371472 [7:05:14<19:14,  3.36it/s]
 99%|█████████▉| 367600/371472 [7:05:14<18:14,  3.54it/s]
                                                         
{'loss': 2.5077, 'learning_rate': 1.0938611045271391e-07, 'epoch': 15.83}
+
 99%|█████████▉| 367600/371472 [7:05:14<18:14,  3.54it/s]
 99%|█████████▉| 367601/371472 [7:05:15<18:43,  3.44it/s]
 99%|█████████▉| 367602/371472 [7:05:15<18:14,  3.54it/s]
 99%|█████████▉| 367603/371472 [7:05:15<17:52,  3.61it/s]
 99%|█████████▉| 367604/371472 [7:05:16<18:13,  3.54it/s]
 99%|█████████▉| 367605/371472 [7:05:16<17:35,  3.66it/s]
 99%|█████████▉| 367606/371472 [7:05:16<17:54,  3.60it/s]
 99%|█████████▉| 367607/371472 [7:05:16<17:17,  3.73it/s]
 99%|█████████▉| 367608/371472 [7:05:17<17:02,  3.78it/s]
 99%|█████████▉| 367609/371472 [7:05:17<17:12,  3.74it/s]
 99%|█████████▉| 367610/371472 [7:05:17<17:17,  3.72it/s]
 99%|█████████▉| 367611/371472 [7:05:17<17:48,  3.61it/s]
 99%|█████████▉| 367612/371472 [7:05:18<17:38,  3.65it/s]
 99%|█████████▉| 367613/371472 [7:05:18<17:17,  3.72it/s]
 99%|█████████▉| 367614/371472 [7:05:18<16:45,  3.84it/s]
 99%|█████████▉| 367615/371472 [7:05:18<17:12,  3.74it/s]
 99%|█████████▉| 367616/371472 [7:05:19<17:14,  3.73it/s]
 99%|█████████▉| 367617/371472 [7:05:19<16:54,  3.80it/s]
 99%|█████████▉| 367618/371472 [7:05:19<16:58,  3.79it/s]
 99%|█████████▉| 367619/371472 [7:05:20<18:29,  3.47it/s]
 99%|█████████▉| 367620/371472 [7:05:20<18:44,  3.43it/s]
                                                         
{'loss': 2.6157, 'learning_rate': 1.0933762847723497e-07, 'epoch': 15.83}
+
 99%|█████████▉| 367620/371472 [7:05:20<18:44,  3.43it/s]
 99%|█████████▉| 367621/371472 [7:05:20<18:25,  3.48it/s]
 99%|█████████▉| 367622/371472 [7:05:20<17:25,  3.68it/s]
 99%|█████████▉| 367623/371472 [7:05:21<17:15,  3.72it/s]
 99%|█████████▉| 367624/371472 [7:05:21<16:42,  3.84it/s]
 99%|█████████▉| 367625/371472 [7:05:21<18:16,  3.51it/s]
 99%|█████████▉| 367626/371472 [7:05:22<17:49,  3.60it/s]
 99%|█████████▉| 367627/371472 [7:05:22<17:26,  3.67it/s]
 99%|█████████▉| 367628/371472 [7:05:22<16:59,  3.77it/s]
 99%|█████████▉| 367629/371472 [7:05:22<17:41,  3.62it/s]
 99%|█████████▉| 367630/371472 [7:05:23<17:31,  3.65it/s]
 99%|█████████▉| 367631/371472 [7:05:23<17:17,  3.70it/s]
 99%|█████████▉| 367632/371472 [7:05:23<17:10,  3.73it/s]
 99%|█████████▉| 367633/371472 [7:05:23<17:17,  3.70it/s]
 99%|█████████▉| 367634/371472 [7:05:24<17:50,  3.59it/s]
 99%|█████████▉| 367635/371472 [7:05:24<17:53,  3.57it/s]
 99%|█████████▉| 367636/371472 [7:05:24<17:22,  3.68it/s]
 99%|█████████▉| 367637/371472 [7:05:25<17:44,  3.60it/s]
 99%|█████████▉| 367638/371472 [7:05:25<17:35,  3.63it/s]
 99%|█████████▉| 367639/371472 [7:05:25<17:03,  3.75it/s]
 99%|█████████▉| 367640/371472 [7:05:25<16:55,  3.77it/s]
                                                         
{'loss': 2.5226, 'learning_rate': 1.0928914650175612e-07, 'epoch': 15.83}
+
 99%|█████████▉| 367640/371472 [7:05:25<16:55,  3.77it/s]
 99%|█████████▉| 367641/371472 [7:05:26<17:08,  3.72it/s]
 99%|█████████▉| 367642/371472 [7:05:26<17:55,  3.56it/s]
 99%|█████████▉| 367643/371472 [7:05:26<19:53,  3.21it/s]
 99%|█████████▉| 367644/371472 [7:05:27<19:03,  3.35it/s]
 99%|█████████▉| 367645/371472 [7:05:27<18:43,  3.41it/s]
 99%|█████████▉| 367646/371472 [7:05:27<17:59,  3.54it/s]
 99%|█████████▉| 367647/371472 [7:05:27<17:33,  3.63it/s]
 99%|█████████▉| 367648/371472 [7:05:28<17:47,  3.58it/s]
 99%|█████████▉| 367649/371472 [7:05:28<19:13,  3.31it/s]
 99%|█████████▉| 367650/371472 [7:05:28<18:52,  3.38it/s]
 99%|█████████▉| 367651/371472 [7:05:29<18:13,  3.50it/s]
 99%|█████████▉| 367652/371472 [7:05:29<17:51,  3.56it/s]
 99%|█████████▉| 367653/371472 [7:05:29<18:11,  3.50it/s]
 99%|█████████▉| 367654/371472 [7:05:29<19:14,  3.31it/s]
 99%|█████████▉| 367655/371472 [7:05:30<18:38,  3.41it/s]
 99%|█████████▉| 367656/371472 [7:05:30<18:16,  3.48it/s]
 99%|█████████▉| 367657/371472 [7:05:30<17:54,  3.55it/s]
 99%|█████████▉| 367658/371472 [7:05:31<18:36,  3.42it/s]
 99%|█████████▉| 367659/371472 [7:05:31<20:12,  3.15it/s]
 99%|█████████▉| 367660/371472 [7:05:31<18:50,  3.37it/s]
                                                         
{'loss': 2.5803, 'learning_rate': 1.0924066452627719e-07, 'epoch': 15.84}
+
 99%|█████████▉| 367660/371472 [7:05:31<18:50,  3.37it/s]
 99%|█████████▉| 367661/371472 [7:05:31<18:33,  3.42it/s]
 99%|█████████▉| 367662/371472 [7:05:32<17:53,  3.55it/s]
 99%|█████████▉| 367663/371472 [7:05:32<17:38,  3.60it/s]
 99%|█████████▉| 367664/371472 [7:05:32<19:29,  3.26it/s]
 99%|█████████▉| 367665/371472 [7:05:33<18:42,  3.39it/s]
 99%|█████████▉| 367666/371472 [7:05:33<18:23,  3.45it/s]
 99%|█████████▉| 367667/371472 [7:05:33<18:13,  3.48it/s]
 99%|█████████▉| 367668/371472 [7:05:33<17:57,  3.53it/s]
 99%|█████████▉| 367669/371472 [7:05:34<19:18,  3.28it/s]
 99%|█████████▉| 367670/371472 [7:05:34<22:39,  2.80it/s]
 99%|█████████▉| 367671/371472 [7:05:35<20:53,  3.03it/s]
 99%|█████████▉| 367672/371472 [7:05:35<19:40,  3.22it/s]
 99%|█████████▉| 367673/371472 [7:05:35<18:35,  3.41it/s]
 99%|█████████▉| 367674/371472 [7:05:35<18:09,  3.49it/s]
 99%|█████████▉| 367675/371472 [7:05:36<17:33,  3.60it/s]
 99%|█████████▉| 367676/371472 [7:05:36<17:23,  3.64it/s]
 99%|█████████▉| 367677/371472 [7:05:36<17:41,  3.58it/s]
 99%|█████████▉| 367678/371472 [7:05:36<17:14,  3.67it/s]
 99%|█████████▉| 367679/371472 [7:05:37<16:59,  3.72it/s]
 99%|█████████▉| 367680/371472 [7:05:37<17:26,  3.62it/s]
                                                         
{'loss': 2.5906, 'learning_rate': 1.0919218255079834e-07, 'epoch': 15.84}
+
 99%|█████████▉| 367680/371472 [7:05:37<17:26,  3.62it/s]
 99%|█████████▉| 367681/371472 [7:05:37<18:59,  3.33it/s]
 99%|█████████▉| 367682/371472 [7:05:38<18:41,  3.38it/s]
 99%|█████████▉| 367683/371472 [7:05:38<18:27,  3.42it/s]
 99%|█████████▉| 367684/371472 [7:05:38<18:46,  3.36it/s]
 99%|█████████▉| 367685/371472 [7:05:38<17:37,  3.58it/s]
 99%|█████████▉| 367686/371472 [7:05:39<16:58,  3.72it/s]
 99%|█████████▉| 367687/371472 [7:05:39<16:31,  3.82it/s]
 99%|█████████▉| 367688/371472 [7:05:39<16:26,  3.84it/s]
 99%|█████████▉| 367689/371472 [7:05:40<16:43,  3.77it/s]
 99%|█████████▉| 367690/371472 [7:05:40<16:50,  3.74it/s]
 99%|█████████▉| 367691/371472 [7:05:40<17:47,  3.54it/s]
 99%|█████████▉| 367692/371472 [7:05:40<17:41,  3.56it/s]
 99%|█████████▉| 367693/371472 [7:05:41<17:29,  3.60it/s]
 99%|█████████▉| 367694/371472 [7:05:41<18:43,  3.36it/s]
 99%|█████████▉| 367695/371472 [7:05:41<18:21,  3.43it/s]
 99%|█████████▉| 367696/371472 [7:05:42<18:31,  3.40it/s]
 99%|█████████▉| 367697/371472 [7:05:42<19:16,  3.26it/s]
 99%|█████████▉| 367698/371472 [7:05:42<18:39,  3.37it/s]
 99%|█████████▉| 367699/371472 [7:05:42<17:46,  3.54it/s]
 99%|█████████▉| 367700/371472 [7:05:43<17:27,  3.60it/s]
                                                         
{'loss': 2.511, 'learning_rate': 1.091437005753194e-07, 'epoch': 15.84}
+
 99%|█████████▉| 367700/371472 [7:05:43<17:27,  3.60it/s]
 99%|█████████▉| 367701/371472 [7:05:43<18:33,  3.39it/s]
 99%|█████████▉| 367702/371472 [7:05:43<17:45,  3.54it/s]
 99%|█████████▉| 367703/371472 [7:05:44<17:08,  3.67it/s]
 99%|█████████▉| 367704/371472 [7:05:44<18:23,  3.42it/s]
 99%|█████████▉| 367705/371472 [7:05:44<17:47,  3.53it/s]
 99%|█████████▉| 367706/371472 [7:05:44<17:20,  3.62it/s]
 99%|█████████▉| 367707/371472 [7:05:45<17:18,  3.63it/s]
 99%|█████████▉| 367708/371472 [7:05:45<17:27,  3.59it/s]
 99%|█████████▉| 367709/371472 [7:05:45<17:06,  3.66it/s]
 99%|█████████▉| 367710/371472 [7:05:45<16:43,  3.75it/s]
 99%|█████████▉| 367711/371472 [7:05:46<18:46,  3.34it/s]
 99%|█████████▉| 367712/371472 [7:05:46<19:20,  3.24it/s]
 99%|█████████▉| 367713/371472 [7:05:46<19:06,  3.28it/s]
 99%|█████████▉| 367714/371472 [7:05:47<18:31,  3.38it/s]
 99%|█████████▉| 367715/371472 [7:05:47<19:11,  3.26it/s]
 99%|█████████▉| 367716/371472 [7:05:47<18:41,  3.35it/s]
 99%|█████████▉| 367717/371472 [7:05:48<17:30,  3.57it/s]
 99%|█████████▉| 367718/371472 [7:05:48<17:33,  3.56it/s]
 99%|█████████▉| 367719/371472 [7:05:48<17:10,  3.64it/s]
 99%|█████████▉| 367720/371472 [7:05:48<16:30,  3.79it/s]
                                                         
{'loss': 2.359, 'learning_rate': 1.0909521859984055e-07, 'epoch': 15.84}
+
 99%|█████████▉| 367720/371472 [7:05:48<16:30,  3.79it/s]
 99%|█████████▉| 367721/371472 [7:05:49<17:56,  3.48it/s]
 99%|█████████▉| 367722/371472 [7:05:49<17:06,  3.65it/s]
 99%|█████████▉| 367723/371472 [7:05:49<18:21,  3.40it/s]
 99%|█████████▉| 367724/371472 [7:05:50<18:04,  3.46it/s]
 99%|█████████▉| 367725/371472 [7:05:50<18:19,  3.41it/s]
 99%|█████████▉| 367726/371472 [7:05:50<17:53,  3.49it/s]
 99%|█████████▉| 367727/371472 [7:05:50<17:47,  3.51it/s]
 99%|█████████▉| 367728/371472 [7:05:51<17:11,  3.63it/s]
 99%|█████████▉| 367729/371472 [7:05:51<18:43,  3.33it/s]
 99%|█████████▉| 367730/371472 [7:05:51<19:08,  3.26it/s]
 99%|█████████▉| 367731/371472 [7:05:52<19:30,  3.20it/s]
 99%|█████████▉| 367732/371472 [7:05:52<19:10,  3.25it/s]
 99%|█████████▉| 367733/371472 [7:05:52<19:48,  3.15it/s]
 99%|█████████▉| 367734/371472 [7:05:53<18:44,  3.32it/s]
 99%|█████████▉| 367735/371472 [7:05:53<17:44,  3.51it/s]
 99%|█████████▉| 367736/371472 [7:05:53<17:45,  3.51it/s]
 99%|█████████▉| 367737/371472 [7:05:53<17:36,  3.54it/s]
 99%|█████████▉| 367738/371472 [7:05:54<16:43,  3.72it/s]
 99%|█████████▉| 367739/371472 [7:05:54<16:41,  3.73it/s]
 99%|█████████▉| 367740/371472 [7:05:54<17:52,  3.48it/s]
                                                         
{'loss': 2.4398, 'learning_rate': 1.0904673662436161e-07, 'epoch': 15.84}
+
 99%|█████████▉| 367740/371472 [7:05:54<17:52,  3.48it/s]
 99%|█████████▉| 367741/371472 [7:05:55<18:29,  3.36it/s]
 99%|█████████▉| 367742/371472 [7:05:55<17:53,  3.48it/s]
 99%|█████████▉| 367743/371472 [7:05:55<18:06,  3.43it/s]
 99%|█████████▉| 367744/371472 [7:05:55<18:10,  3.42it/s]
 99%|█████████▉| 367745/371472 [7:05:56<18:07,  3.43it/s]
 99%|█████████▉| 367746/371472 [7:05:56<17:58,  3.45it/s]
 99%|█████████▉| 367747/371472 [7:05:56<18:10,  3.42it/s]
 99%|█████████▉| 367748/371472 [7:05:57<18:47,  3.30it/s]
 99%|█████████▉| 367749/371472 [7:05:57<18:01,  3.44it/s]
 99%|█████████▉| 367750/371472 [7:05:57<17:21,  3.57it/s]
 99%|█████████▉| 367751/371472 [7:05:57<17:08,  3.62it/s]
 99%|█████████▉| 367752/371472 [7:05:58<16:34,  3.74it/s]
 99%|█████████▉| 367753/371472 [7:05:58<17:05,  3.63it/s]
 99%|█████████▉| 367754/371472 [7:05:58<17:53,  3.46it/s]
 99%|█████████▉| 367755/371472 [7:05:59<17:14,  3.59it/s]
 99%|█████████▉| 367756/371472 [7:05:59<17:39,  3.51it/s]
 99%|█████████▉| 367757/371472 [7:05:59<17:00,  3.64it/s]
 99%|█████████▉| 367758/371472 [7:05:59<16:44,  3.70it/s]
 99%|█████████▉| 367759/371472 [7:06:00<16:45,  3.69it/s]
 99%|█████████▉| 367760/371472 [7:06:00<17:08,  3.61it/s]
                                                         
{'loss': 2.5133, 'learning_rate': 1.0899825464888277e-07, 'epoch': 15.84}
+
 99%|█████████▉| 367760/371472 [7:06:00<17:08,  3.61it/s]
 99%|█████████▉| 367761/371472 [7:06:00<17:56,  3.45it/s]
 99%|█████████▉| 367762/371472 [7:06:00<17:29,  3.54it/s]
 99%|█████████▉| 367763/371472 [7:06:01<17:23,  3.56it/s]
 99%|█████████▉| 367764/371472 [7:06:01<17:04,  3.62it/s]
 99%|█████████▉| 367765/371472 [7:06:01<17:46,  3.48it/s]
 99%|█████████▉| 367766/371472 [7:06:02<17:25,  3.55it/s]
 99%|█████████▉| 367767/371472 [7:06:02<17:43,  3.48it/s]
 99%|█████████▉| 367768/371472 [7:06:02<17:33,  3.52it/s]
 99%|█████████▉| 367769/371472 [7:06:02<17:29,  3.53it/s]
 99%|█████████▉| 367770/371472 [7:06:03<18:26,  3.35it/s]
 99%|█████████▉| 367771/371472 [7:06:03<17:58,  3.43it/s]
 99%|█████████▉| 367772/371472 [7:06:03<18:58,  3.25it/s]
 99%|█████████▉| 367773/371472 [7:06:04<18:34,  3.32it/s]
 99%|█████████▉| 367774/371472 [7:06:04<18:06,  3.40it/s]
 99%|█████████▉| 367775/371472 [7:06:04<18:18,  3.37it/s]
 99%|█████████▉| 367776/371472 [7:06:05<17:51,  3.45it/s]
 99%|█████████▉| 367777/371472 [7:06:05<17:50,  3.45it/s]
 99%|█████████▉| 367778/371472 [7:06:05<19:14,  3.20it/s]
 99%|█████████▉| 367779/371472 [7:06:06<19:45,  3.12it/s]
 99%|█████████▉| 367780/371472 [7:06:06<18:31,  3.32it/s]
                                                         
{'loss': 2.6249, 'learning_rate': 1.0894977267340383e-07, 'epoch': 15.84}
+
 99%|█████████▉| 367780/371472 [7:06:06<18:31,  3.32it/s]
 99%|█████████▉| 367781/371472 [7:06:06<19:45,  3.11it/s]
 99%|█████████▉| 367782/371472 [7:06:06<19:10,  3.21it/s]
 99%|█████████▉| 367783/371472 [7:06:07<19:06,  3.22it/s]
 99%|█████████▉| 367784/371472 [7:06:07<19:05,  3.22it/s]
 99%|█████████▉| 367785/371472 [7:06:07<18:23,  3.34it/s]
 99%|█████████▉| 367786/371472 [7:06:08<18:49,  3.26it/s]
 99%|█████████▉| 367787/371472 [7:06:08<18:05,  3.39it/s]
 99%|█████████▉| 367788/371472 [7:06:08<17:58,  3.42it/s]
 99%|█████████▉| 367789/371472 [7:06:09<18:12,  3.37it/s]
 99%|█████████▉| 367790/371472 [7:06:09<17:42,  3.46it/s]
 99%|█████████▉| 367791/371472 [7:06:09<18:50,  3.26it/s]
 99%|█████████▉| 367792/371472 [7:06:09<18:38,  3.29it/s]
 99%|█████████▉| 367793/371472 [7:06:10<18:09,  3.38it/s]
 99%|█████████▉| 367794/371472 [7:06:10<17:38,  3.48it/s]
 99%|█████████▉| 367795/371472 [7:06:10<17:00,  3.60it/s]
 99%|█████████▉| 367796/371472 [7:06:11<16:31,  3.71it/s]
 99%|█████████▉| 367797/371472 [7:06:11<16:25,  3.73it/s]
 99%|█████████▉| 367798/371472 [7:06:11<16:38,  3.68it/s]
 99%|█████████▉| 367799/371472 [7:06:11<16:13,  3.77it/s]
 99%|█████████▉| 367800/371472 [7:06:12<17:11,  3.56it/s]
                                                         
{'loss': 2.6033, 'learning_rate': 1.0890129069792498e-07, 'epoch': 15.84}
+
 99%|█████████▉| 367800/371472 [7:06:12<17:11,  3.56it/s]
 99%|█████████▉| 367801/371472 [7:06:12<17:02,  3.59it/s]
 99%|█████████▉| 367802/371472 [7:06:12<16:48,  3.64it/s]
 99%|█████████▉| 367803/371472 [7:06:12<16:41,  3.66it/s]
 99%|█████████▉| 367804/371472 [7:06:13<16:32,  3.69it/s]
 99%|█████████▉| 367805/371472 [7:06:13<16:10,  3.78it/s]
 99%|█████████▉| 367806/371472 [7:06:13<16:34,  3.69it/s]
 99%|█████████▉| 367807/371472 [7:06:14<17:09,  3.56it/s]
 99%|█████████▉| 367808/371472 [7:06:14<18:10,  3.36it/s]
 99%|█████████▉| 367809/371472 [7:06:14<18:30,  3.30it/s]
 99%|█████████▉| 367810/371472 [7:06:15<18:51,  3.24it/s]
 99%|█████████▉| 367811/371472 [7:06:15<17:46,  3.43it/s]
 99%|█████████▉| 367812/371472 [7:06:15<18:06,  3.37it/s]
 99%|█████████▉| 367813/371472 [7:06:15<17:46,  3.43it/s]
 99%|█████████▉| 367814/371472 [7:06:16<16:59,  3.59it/s]
 99%|█████████▉| 367815/371472 [7:06:16<16:49,  3.62it/s]
 99%|█████████▉| 367816/371472 [7:06:16<16:46,  3.63it/s]
 99%|█████████▉| 367817/371472 [7:06:16<16:41,  3.65it/s]
 99%|█████████▉| 367818/371472 [7:06:17<17:07,  3.55it/s]
 99%|█████████▉| 367819/371472 [7:06:17<17:03,  3.57it/s]
 99%|█████████▉| 367820/371472 [7:06:17<18:44,  3.25it/s]
                                                         
{'loss': 2.5214, 'learning_rate': 1.0885280872244604e-07, 'epoch': 15.84}
+
 99%|█████████▉| 367820/371472 [7:06:17<18:44,  3.25it/s]
 99%|█████████▉| 367821/371472 [7:06:18<18:16,  3.33it/s]
 99%|█████████▉| 367822/371472 [7:06:18<19:08,  3.18it/s]
 99%|█████████▉| 367823/371472 [7:06:18<18:53,  3.22it/s]
 99%|█████████▉| 367824/371472 [7:06:19<17:46,  3.42it/s]
 99%|█████████▉| 367825/371472 [7:06:19<17:15,  3.52it/s]
 99%|█████████▉| 367826/371472 [7:06:19<16:59,  3.58it/s]
 99%|█████████▉| 367827/371472 [7:06:19<17:17,  3.51it/s]
 99%|█████████▉| 367828/371472 [7:06:20<17:14,  3.52it/s]
 99%|█████████▉| 367829/371472 [7:06:20<17:40,  3.44it/s]
 99%|█████████▉| 367830/371472 [7:06:20<17:12,  3.53it/s]
 99%|█████████▉| 367831/371472 [7:06:21<17:25,  3.48it/s]
 99%|█████████▉| 367832/371472 [7:06:21<17:46,  3.41it/s]
 99%|█████████▉| 367833/371472 [7:06:21<17:04,  3.55it/s]
 99%|█████████▉| 367834/371472 [7:06:21<16:34,  3.66it/s]
 99%|█████████▉| 367835/371472 [7:06:22<16:39,  3.64it/s]
 99%|█████████▉| 367836/371472 [7:06:22<16:44,  3.62it/s]
 99%|█████████▉| 367837/371472 [7:06:22<16:37,  3.64it/s]
 99%|█████████▉| 367838/371472 [7:06:22<16:23,  3.70it/s]
 99%|█████████▉| 367839/371472 [7:06:23<16:20,  3.71it/s]
 99%|█████████▉| 367840/371472 [7:06:23<16:37,  3.64it/s]
                                                         
{'loss': 2.5907, 'learning_rate': 1.0880432674696722e-07, 'epoch': 15.84}
+
 99%|█████████▉| 367840/371472 [7:06:23<16:37,  3.64it/s]
 99%|█████████▉| 367841/371472 [7:06:23<16:45,  3.61it/s]
 99%|█████████▉| 367842/371472 [7:06:24<16:52,  3.58it/s]
 99%|█████████▉| 367843/371472 [7:06:24<17:13,  3.51it/s]
 99%|█████████▉| 367844/371472 [7:06:24<17:09,  3.52it/s]
 99%|█████████▉| 367845/371472 [7:06:24<16:39,  3.63it/s]
 99%|█████████▉| 367846/371472 [7:06:25<18:20,  3.30it/s]
 99%|█████████▉| 367847/371472 [7:06:25<18:01,  3.35it/s]
 99%|█████████▉| 367848/371472 [7:06:25<17:54,  3.37it/s]
 99%|█████████▉| 367849/371472 [7:06:26<18:26,  3.28it/s]
 99%|█████████▉| 367850/371472 [7:06:26<17:34,  3.44it/s]
 99%|█████████▉| 367851/371472 [7:06:26<18:05,  3.34it/s]
 99%|█████████▉| 367852/371472 [7:06:27<18:06,  3.33it/s]
 99%|█████████▉| 367853/371472 [7:06:27<18:02,  3.34it/s]
 99%|█████████▉| 367854/371472 [7:06:27<17:29,  3.45it/s]
 99%|█████████▉| 367855/371472 [7:06:27<17:15,  3.49it/s]
 99%|█████████▉| 367856/371472 [7:06:28<16:40,  3.61it/s]
 99%|█████████▉| 367857/371472 [7:06:28<17:40,  3.41it/s]
 99%|█████████▉| 367858/371472 [7:06:28<17:06,  3.52it/s]
 99%|█████████▉| 367859/371472 [7:06:28<16:44,  3.60it/s]
 99%|█████████▉| 367860/371472 [7:06:29<17:04,  3.53it/s]
                                                         
{'loss': 2.4936, 'learning_rate': 1.0875584477148826e-07, 'epoch': 15.84}
+
 99%|█████████▉| 367860/371472 [7:06:29<17:04,  3.53it/s]
 99%|█████████▉| 367861/371472 [7:06:29<16:33,  3.63it/s]
 99%|█████████▉| 367862/371472 [7:06:29<17:45,  3.39it/s]
 99%|█████████▉| 367863/371472 [7:06:30<17:52,  3.37it/s]
 99%|█████████▉| 367864/371472 [7:06:30<17:30,  3.44it/s]
 99%|█████████▉| 367865/371472 [7:06:30<17:45,  3.39it/s]
 99%|█████████▉| 367866/371472 [7:06:31<17:37,  3.41it/s]
 99%|█████████▉| 367867/371472 [7:06:31<16:51,  3.56it/s]
 99%|█████████▉| 367868/371472 [7:06:31<16:19,  3.68it/s]
 99%|█████████▉| 367869/371472 [7:06:31<16:07,  3.72it/s]
 99%|█████████▉| 367870/371472 [7:06:32<16:07,  3.72it/s]
 99%|█████████▉| 367871/371472 [7:06:32<15:42,  3.82it/s]
 99%|█████████▉| 367872/371472 [7:06:32<16:10,  3.71it/s]
 99%|█████████▉| 367873/371472 [7:06:32<16:06,  3.72it/s]
 99%|█████████▉| 367874/371472 [7:06:33<15:52,  3.78it/s]
 99%|█████████▉| 367875/371472 [7:06:33<15:22,  3.90it/s]
 99%|█████████▉| 367876/371472 [7:06:33<17:30,  3.42it/s]
 99%|█████████▉| 367877/371472 [7:06:34<18:06,  3.31it/s]
 99%|█████████▉| 367878/371472 [7:06:34<17:12,  3.48it/s]
 99%|█████████▉| 367879/371472 [7:06:34<17:06,  3.50it/s]
 99%|█████████▉| 367880/371472 [7:06:34<16:47,  3.57it/s]
                                                         
{'loss': 2.6067, 'learning_rate': 1.0870736279600941e-07, 'epoch': 15.85}
+
 99%|█████████▉| 367880/371472 [7:06:34<16:47,  3.57it/s]
 99%|█████████▉| 367881/371472 [7:06:35<16:59,  3.52it/s]
 99%|█████████▉| 367882/371472 [7:06:35<17:12,  3.48it/s]
 99%|█████████▉| 367883/371472 [7:06:35<16:54,  3.54it/s]
 99%|█████████▉| 367884/371472 [7:06:36<16:27,  3.63it/s]
 99%|█████████▉| 367885/371472 [7:06:36<16:12,  3.69it/s]
 99%|█████████▉| 367886/371472 [7:06:36<16:15,  3.68it/s]
 99%|█████████▉| 367887/371472 [7:06:36<16:28,  3.63it/s]
 99%|█████████▉| 367888/371472 [7:06:37<16:24,  3.64it/s]
 99%|█████████▉| 367889/371472 [7:06:37<16:46,  3.56it/s]
 99%|█████████▉| 367890/371472 [7:06:37<16:21,  3.65it/s]
 99%|█████████▉| 367891/371472 [7:06:37<15:52,  3.76it/s]
 99%|█████████▉| 367892/371472 [7:06:38<15:58,  3.74it/s]
 99%|█████████▉| 367893/371472 [7:06:38<16:22,  3.64it/s]
 99%|█████████▉| 367894/371472 [7:06:38<16:17,  3.66it/s]
 99%|█████████▉| 367895/371472 [7:06:39<16:42,  3.57it/s]
 99%|█████████▉| 367896/371472 [7:06:39<16:34,  3.60it/s]
 99%|█████████▉| 367897/371472 [7:06:39<16:13,  3.67it/s]
 99%|█████████▉| 367898/371472 [7:06:39<16:59,  3.51it/s]
 99%|█████████▉| 367899/371472 [7:06:40<16:55,  3.52it/s]
 99%|█████████▉| 367900/371472 [7:06:40<16:56,  3.52it/s]
                                                         
{'loss': 2.6242, 'learning_rate': 1.0865888082053047e-07, 'epoch': 15.85}
+
 99%|█████████▉| 367900/371472 [7:06:40<16:56,  3.52it/s]
 99%|█████████▉| 367901/371472 [7:06:40<16:46,  3.55it/s]
 99%|█████████▉| 367902/371472 [7:06:40<16:38,  3.57it/s]
 99%|█████████▉| 367903/371472 [7:06:41<15:47,  3.77it/s]
 99%|█████████▉| 367904/371472 [7:06:41<15:57,  3.73it/s]
 99%|█████████▉| 367905/371472 [7:06:41<15:44,  3.78it/s]
 99%|█████████▉| 367906/371472 [7:06:42<16:18,  3.64it/s]
 99%|█████████▉| 367907/371472 [7:06:42<15:58,  3.72it/s]
 99%|█████████▉| 367908/371472 [7:06:42<16:11,  3.67it/s]
 99%|█████████▉| 367909/371472 [7:06:42<16:06,  3.69it/s]
 99%|█████████▉| 367910/371472 [7:06:43<15:46,  3.76it/s]
 99%|█████████▉| 367911/371472 [7:06:43<15:53,  3.73it/s]
 99%|█████████▉| 367912/371472 [7:06:43<16:00,  3.70it/s]
 99%|█████████▉| 367913/371472 [7:06:43<16:11,  3.67it/s]
 99%|█████████▉| 367914/371472 [7:06:44<15:39,  3.79it/s]
 99%|█████████▉| 367915/371472 [7:06:44<15:31,  3.82it/s]
 99%|█████████▉| 367916/371472 [7:06:44<15:04,  3.93it/s]
 99%|█████████▉| 367917/371472 [7:06:44<15:24,  3.85it/s]
 99%|█████████▉| 367918/371472 [7:06:45<15:56,  3.71it/s]
 99%|█████████▉| 367919/371472 [7:06:45<17:19,  3.42it/s]
 99%|█████████▉| 367920/371472 [7:06:45<17:33,  3.37it/s]
                                                         
{'loss': 2.833, 'learning_rate': 1.0861039884505164e-07, 'epoch': 15.85}
+
 99%|█████████▉| 367920/371472 [7:06:45<17:33,  3.37it/s]
 99%|█████████▉| 367921/371472 [7:06:46<16:46,  3.53it/s]
 99%|█████████▉| 367922/371472 [7:06:46<16:27,  3.60it/s]
 99%|█████████▉| 367923/371472 [7:06:46<16:33,  3.57it/s]
 99%|█████████▉| 367924/371472 [7:06:46<16:14,  3.64it/s]
 99%|█████████▉| 367925/371472 [7:06:47<16:21,  3.62it/s]
 99%|█████████▉| 367926/371472 [7:06:47<17:46,  3.33it/s]
 99%|█████████▉| 367927/371472 [7:06:47<17:31,  3.37it/s]
 99%|█████████▉| 367928/371472 [7:06:48<17:03,  3.46it/s]
 99%|█████████▉| 367929/371472 [7:06:48<17:02,  3.47it/s]
 99%|█████████▉| 367930/371472 [7:06:48<17:09,  3.44it/s]
 99%|█████████▉| 367931/371472 [7:06:49<18:11,  3.24it/s]
 99%|█████████▉| 367932/371472 [7:06:49<17:59,  3.28it/s]
 99%|█████████▉| 367933/371472 [7:06:49<17:38,  3.34it/s]
 99%|█████████▉| 367934/371472 [7:06:49<17:31,  3.36it/s]
 99%|█████████▉| 367935/371472 [7:06:50<18:30,  3.18it/s]
 99%|█████████▉| 367936/371472 [7:06:50<18:07,  3.25it/s]
 99%|█████████▉| 367937/371472 [7:06:50<17:23,  3.39it/s]
 99%|█████████▉| 367938/371472 [7:06:51<16:48,  3.50it/s]
 99%|█████████▉| 367939/371472 [7:06:51<16:54,  3.48it/s]
 99%|█████████▉| 367940/371472 [7:06:51<16:30,  3.57it/s]
                                                         
{'loss': 2.4719, 'learning_rate': 1.085619168695727e-07, 'epoch': 15.85}
+
 99%|█████████▉| 367940/371472 [7:06:51<16:30,  3.57it/s]
 99%|█████████▉| 367941/371472 [7:06:51<16:12,  3.63it/s]
 99%|█████████▉| 367942/371472 [7:06:52<16:04,  3.66it/s]
 99%|█████████▉| 367943/371472 [7:06:52<15:40,  3.75it/s]
 99%|█████████▉| 367944/371472 [7:06:52<17:06,  3.44it/s]
 99%|█████████▉| 367945/371472 [7:06:53<17:56,  3.28it/s]
 99%|█████████▉| 367946/371472 [7:06:53<17:08,  3.43it/s]
 99%|█████████▉| 367947/371472 [7:06:53<18:03,  3.25it/s]
 99%|█████████▉| 367948/371472 [7:06:54<17:24,  3.37it/s]
 99%|█████████▉| 367949/371472 [7:06:54<16:47,  3.50it/s]
 99%|█████████▉| 367950/371472 [7:06:54<16:08,  3.64it/s]
 99%|█████████▉| 367951/371472 [7:06:54<16:00,  3.67it/s]
 99%|█████████▉| 367952/371472 [7:06:55<15:42,  3.74it/s]
 99%|█████████▉| 367953/371472 [7:06:55<15:35,  3.76it/s]
 99%|█████████▉| 367954/371472 [7:06:55<15:56,  3.68it/s]
 99%|█████████▉| 367955/371472 [7:06:55<15:51,  3.70it/s]
 99%|█████████▉| 367956/371472 [7:06:56<16:19,  3.59it/s]
 99%|█████████▉| 367957/371472 [7:06:56<17:03,  3.43it/s]
 99%|█████████▉| 367958/371472 [7:06:56<16:52,  3.47it/s]
 99%|█████████▉| 367959/371472 [7:06:57<17:38,  3.32it/s]
 99%|█████████▉| 367960/371472 [7:06:57<16:59,  3.44it/s]
                                                         
{'loss': 2.8012, 'learning_rate': 1.0851343489409385e-07, 'epoch': 15.85}
+
 99%|█████████▉| 367960/371472 [7:06:57<16:59,  3.44it/s]
 99%|█████████▉| 367961/371472 [7:06:57<17:04,  3.43it/s]
 99%|█████████▉| 367962/371472 [7:06:57<16:21,  3.58it/s]
 99%|█████████▉| 367963/371472 [7:06:58<17:32,  3.33it/s]
 99%|█████████▉| 367964/371472 [7:06:58<16:58,  3.44it/s]
 99%|█████████▉| 367965/371472 [7:06:58<16:27,  3.55it/s]
 99%|█████████▉| 367966/371472 [7:06:59<16:22,  3.57it/s]
 99%|█████████▉| 367967/371472 [7:06:59<16:27,  3.55it/s]
 99%|█████████▉| 367968/371472 [7:06:59<17:25,  3.35it/s]
 99%|█████████▉| 367969/371472 [7:06:59<16:30,  3.54it/s]
 99%|█████████▉| 367970/371472 [7:07:00<16:57,  3.44it/s]
 99%|█████████▉| 367971/371472 [7:07:00<16:27,  3.55it/s]
 99%|█████████▉| 367972/371472 [7:07:00<16:08,  3.61it/s]
 99%|█████████▉| 367973/371472 [7:07:01<16:24,  3.55it/s]
 99%|█████████▉| 367974/371472 [7:07:01<16:42,  3.49it/s]
 99%|█████████▉| 367975/371472 [7:07:01<16:56,  3.44it/s]
 99%|█████████▉| 367976/371472 [7:07:01<17:09,  3.40it/s]
 99%|█████████▉| 367977/371472 [7:07:02<17:26,  3.34it/s]
 99%|█████████▉| 367978/371472 [7:07:02<18:04,  3.22it/s]
 99%|█████████▉| 367979/371472 [7:07:02<17:31,  3.32it/s]
 99%|█████████▉| 367980/371472 [7:07:03<17:28,  3.33it/s]
                                                         
{'loss': 2.5388, 'learning_rate': 1.084649529186149e-07, 'epoch': 15.85}
+
 99%|█████████▉| 367980/371472 [7:07:03<17:28,  3.33it/s]
 99%|█████████▉| 367981/371472 [7:07:03<17:25,  3.34it/s]
 99%|█████████▉| 367982/371472 [7:07:03<17:30,  3.32it/s]
 99%|█████████▉| 367983/371472 [7:07:04<16:34,  3.51it/s]
 99%|█████████▉| 367984/371472 [7:07:04<16:43,  3.48it/s]
 99%|█████████▉| 367985/371472 [7:07:04<17:40,  3.29it/s]
 99%|█████████▉| 367986/371472 [7:07:04<17:44,  3.28it/s]
 99%|█████████▉| 367987/371472 [7:07:05<17:45,  3.27it/s]
 99%|█████████▉| 367988/371472 [7:07:05<17:19,  3.35it/s]
 99%|█████████▉| 367989/371472 [7:07:05<16:34,  3.50it/s]
 99%|█████████▉| 367990/371472 [7:07:06<17:20,  3.35it/s]
 99%|█████████▉| 367991/371472 [7:07:06<16:43,  3.47it/s]
 99%|█████████▉| 367992/371472 [7:07:06<16:48,  3.45it/s]
 99%|█████████▉| 367993/371472 [7:07:07<16:53,  3.43it/s]
 99%|█████████▉| 367994/371472 [7:07:07<16:07,  3.59it/s]
 99%|█████████▉| 367995/371472 [7:07:07<16:50,  3.44it/s]
 99%|███��█████▉| 367996/371472 [7:07:07<16:03,  3.61it/s]
 99%|█████████▉| 367997/371472 [7:07:08<17:09,  3.38it/s]
 99%|█████████▉| 367998/371472 [7:07:08<17:27,  3.32it/s]
 99%|█████████▉| 367999/371472 [7:07:08<17:25,  3.32it/s]
 99%|█████████▉| 368000/371472 [7:07:09<18:03,  3.20it/s]
                                                         
{'loss': 2.6826, 'learning_rate': 1.0841647094313606e-07, 'epoch': 15.85}
+
 99%|█████████▉| 368000/371472 [7:07:09<18:03,  3.20it/s]
 99%|█████████▉| 368001/371472 [7:07:09<17:45,  3.26it/s]
 99%|█████████▉| 368002/371472 [7:07:09<17:30,  3.30it/s]
 99%|█████████▉| 368003/371472 [7:07:10<17:38,  3.28it/s]
 99%|█████████▉| 368004/371472 [7:07:10<17:05,  3.38it/s]
 99%|█████████▉| 368005/371472 [7:07:10<16:39,  3.47it/s]
 99%|█████████▉| 368006/371472 [7:07:10<16:02,  3.60it/s]
 99%|█████████▉| 368007/371472 [7:07:11<16:05,  3.59it/s]
 99%|█████████▉| 368008/371472 [7:07:11<15:35,  3.70it/s]
 99%|█████████▉| 368009/371472 [7:07:11<15:57,  3.62it/s]
 99%|█████████▉| 368010/371472 [7:07:11<15:47,  3.65it/s]
 99%|█████████▉| 368011/371472 [7:07:12<15:43,  3.67it/s]
 99%|█████████▉| 368012/371472 [7:07:12<15:52,  3.63it/s]
 99%|█████████▉| 368013/371472 [7:07:12<16:01,  3.60it/s]
 99%|█████████▉| 368014/371472 [7:07:13<16:24,  3.51it/s]
 99%|█████████▉| 368015/371472 [7:07:13<15:55,  3.62it/s]
 99%|█████████▉| 368016/371472 [7:07:13<16:45,  3.44it/s]
 99%|█████████▉| 368017/371472 [7:07:14<19:07,  3.01it/s]
 99%|█████████▉| 368018/371472 [7:07:14<18:11,  3.17it/s]
 99%|█████████▉| 368019/371472 [7:07:14<17:17,  3.33it/s]
 99%|█████████▉| 368020/371472 [7:07:14<16:57,  3.39it/s]
                                                         
{'loss': 2.5899, 'learning_rate': 1.0836798896765713e-07, 'epoch': 15.85}
+
 99%|█████████▉| 368020/371472 [7:07:14<16:57,  3.39it/s]
 99%|█████████▉| 368021/371472 [7:07:15<16:53,  3.40it/s]
 99%|█████████▉| 368022/371472 [7:07:15<19:12,  2.99it/s]
 99%|█████████▉| 368023/371472 [7:07:15<18:10,  3.16it/s]
 99%|█████████▉| 368024/371472 [7:07:16<17:23,  3.30it/s]
 99%|█████████▉| 368025/371472 [7:07:16<16:53,  3.40it/s]
 99%|█████████▉| 368026/371472 [7:07:16<16:47,  3.42it/s]
 99%|█████████▉| 368027/371472 [7:07:16<16:05,  3.57it/s]
 99%|█████████▉| 368028/371472 [7:07:17<16:12,  3.54it/s]
 99%|█████████▉| 368029/371472 [7:07:17<16:34,  3.46it/s]
 99%|█████████▉| 368030/371472 [7:07:17<15:44,  3.65it/s]
 99%|█████████▉| 368031/371472 [7:07:18<15:39,  3.66it/s]
 99%|█████████▉| 368032/371472 [7:07:18<15:35,  3.68it/s]
 99%|█████████▉| 368033/371472 [7:07:18<15:05,  3.80it/s]
 99%|█████████▉| 368034/371472 [7:07:18<15:00,  3.82it/s]
 99%|█████████▉| 368035/371472 [7:07:19<15:31,  3.69it/s]
 99%|█████████▉| 368036/371472 [7:07:19<14:54,  3.84it/s]
 99%|█████████▉| 368037/371472 [7:07:19<15:42,  3.64it/s]
 99%|█████████▉| 368038/371472 [7:07:20<17:04,  3.35it/s]
 99%|█████████▉| 368039/371472 [7:07:20<17:23,  3.29it/s]
 99%|█████████▉| 368040/371472 [7:07:20<17:10,  3.33it/s]
                                                         
{'loss': 2.6208, 'learning_rate': 1.0831950699217828e-07, 'epoch': 15.85}
+
 99%|█████████▉| 368040/371472 [7:07:20<17:10,  3.33it/s]
 99%|█████████▉| 368041/371472 [7:07:20<16:17,  3.51it/s]
 99%|█████████▉| 368042/371472 [7:07:21<15:52,  3.60it/s]
 99%|█████████▉| 368043/371472 [7:07:21<15:55,  3.59it/s]
 99%|█████████▉| 368044/371472 [7:07:21<16:13,  3.52it/s]
 99%|█████████▉| 368045/371472 [7:07:22<16:42,  3.42it/s]
 99%|█████████▉| 368046/371472 [7:07:22<16:55,  3.37it/s]
 99%|█████████▉| 368047/371472 [7:07:22<15:51,  3.60it/s]
 99%|█████████▉| 368048/371472 [7:07:22<15:49,  3.61it/s]
 99%|█████████▉| 368049/371472 [7:07:23<15:35,  3.66it/s]
 99%|█████████▉| 368050/371472 [7:07:23<16:27,  3.47it/s]
 99%|█████████▉| 368051/371472 [7:07:23<16:02,  3.56it/s]
 99%|█████████▉| 368052/371472 [7:07:24<16:25,  3.47it/s]
 99%|█████████▉| 368053/371472 [7:07:24<16:23,  3.48it/s]
 99%|█████████▉| 368054/371472 [7:07:24<15:37,  3.65it/s]
 99%|█████████▉| 368055/371472 [7:07:24<16:00,  3.56it/s]
 99%|█████████▉| 368056/371472 [7:07:25<15:32,  3.66it/s]
 99%|█████████▉| 368057/371472 [7:07:25<15:56,  3.57it/s]
 99%|█████████▉| 368058/371472 [7:07:25<16:04,  3.54it/s]
 99%|█████████▉| 368059/371472 [7:07:26<17:01,  3.34it/s]
 99%|█████████▉| 368060/371472 [7:07:26<17:06,  3.32it/s]
                                                         
{'loss': 2.7296, 'learning_rate': 1.0827102501669934e-07, 'epoch': 15.85}
+
 99%|█████████▉| 368060/371472 [7:07:26<17:06,  3.32it/s]
 99%|█████████▉| 368061/371472 [7:07:26<17:46,  3.20it/s]
 99%|█████████▉| 368062/371472 [7:07:26<18:21,  3.09it/s]
 99%|█████████▉| 368063/371472 [7:07:27<16:56,  3.36it/s]
 99%|█████████▉| 368064/371472 [7:07:27<16:26,  3.46it/s]
 99%|█████████▉| 368065/371472 [7:07:27<16:05,  3.53it/s]
 99%|█████████▉| 368066/371472 [7:07:28<16:22,  3.47it/s]
 99%|█████████▉| 368067/371472 [7:07:28<16:05,  3.53it/s]
 99%|█████████▉| 368068/371472 [7:07:28<16:33,  3.43it/s]
 99%|█████████▉| 368069/371472 [7:07:28<15:48,  3.59it/s]
 99%|█████████▉| 368070/371472 [7:07:29<15:26,  3.67it/s]
 99%|█████████▉| 368071/371472 [7:07:29<15:33,  3.64it/s]
 99%|█████████▉| 368072/371472 [7:07:29<16:35,  3.42it/s]
 99%|█████████▉| 368073/371472 [7:07:30<15:55,  3.56it/s]
 99%|█████████▉| 368074/371472 [7:07:30<15:30,  3.65it/s]
 99%|█████████▉| 368075/371472 [7:07:30<15:27,  3.66it/s]
 99%|█████████▉| 368076/371472 [7:07:30<15:21,  3.69it/s]
 99%|█████████▉| 368077/371472 [7:07:31<15:10,  3.73it/s]
 99%|█████████▉| 368078/371472 [7:07:31<15:15,  3.71it/s]
 99%|█████████▉| 368079/371472 [7:07:31<15:22,  3.68it/s]
 99%|█████████▉| 368080/371472 [7:07:31<15:10,  3.73it/s]
                                                         
{'loss': 2.5957, 'learning_rate': 1.082225430412205e-07, 'epoch': 15.85}
+
 99%|█████████▉| 368080/371472 [7:07:31<15:10,  3.73it/s]
 99%|█████████▉| 368081/371472 [7:07:32<15:14,  3.71it/s]
 99%|█████████▉| 368082/371472 [7:07:32<15:21,  3.68it/s]
 99%|█████████▉| 368083/371472 [7:07:32<15:56,  3.54it/s]
 99%|█████████▉| 368084/371472 [7:07:33<17:01,  3.32it/s]
 99%|█████████▉| 368085/371472 [7:07:33<16:19,  3.46it/s]
 99%|█████████▉| 368086/371472 [7:07:33<16:06,  3.50it/s]
 99%|█████████▉| 368087/371472 [7:07:33<16:52,  3.34it/s]
 99%|█████████▉| 368088/371472 [7:07:34<16:27,  3.43it/s]
 99%|█████████▉| 368089/371472 [7:07:34<15:53,  3.55it/s]
 99%|█████████▉| 368090/371472 [7:07:34<15:59,  3.53it/s]
 99%|█████████▉| 368091/371472 [7:07:35<16:55,  3.33it/s]
 99%|█████████▉| 368092/371472 [7:07:35<16:41,  3.38it/s]
 99%|█████████▉| 368093/371472 [7:07:35<16:48,  3.35it/s]
 99%|█████████▉| 368094/371472 [7:07:35<16:08,  3.49it/s]
 99%|█████████▉| 368095/371472 [7:07:36<17:00,  3.31it/s]
 99%|█████████▉| 368096/371472 [7:07:36<17:07,  3.29it/s]
 99%|█████████▉| 368097/371472 [7:07:36<16:29,  3.41it/s]
 99%|█████████▉| 368098/371472 [7:07:37<16:47,  3.35it/s]
 99%|█████████▉| 368099/371472 [7:07:37<17:31,  3.21it/s]
 99%|█████████▉| 368100/371472 [7:07:37<16:32,  3.40it/s]
                                                         
{'loss': 2.471, 'learning_rate': 1.0817406106574154e-07, 'epoch': 15.85}
+
 99%|█████████▉| 368100/371472 [7:07:37<16:32,  3.40it/s]
 99%|█████████▉| 368101/371472 [7:07:38<16:34,  3.39it/s]
 99%|█████████▉| 368102/371472 [7:07:38<17:03,  3.29it/s]
 99%|█████████▉| 368103/371472 [7:07:38<16:47,  3.34it/s]
 99%|█████████▉| 368104/371472 [7:07:38<16:12,  3.46it/s]
 99%|█████████▉| 368105/371472 [7:07:39<15:46,  3.56it/s]
 99%|█████████▉| 368106/371472 [7:07:39<16:22,  3.43it/s]
 99%|█████████▉| 368107/371472 [7:07:39<16:37,  3.37it/s]
 99%|█████████▉| 368108/371472 [7:07:40<18:09,  3.09it/s]
 99%|█████████▉| 368109/371472 [7:07:40<16:50,  3.33it/s]
 99%|█████████▉| 368110/371472 [7:07:40<15:44,  3.56it/s]
 99%|███��█████▉| 368111/371472 [7:07:41<15:37,  3.58it/s]
 99%|█████████▉| 368112/371472 [7:07:41<15:55,  3.52it/s]
 99%|█████████▉| 368113/371472 [7:07:41<15:45,  3.55it/s]
 99%|█████████▉| 368114/371472 [7:07:41<16:18,  3.43it/s]
 99%|█████████▉| 368115/371472 [7:07:42<15:40,  3.57it/s]
 99%|█████████▉| 368116/371472 [7:07:42<15:25,  3.63it/s]
 99%|█████████▉| 368117/371472 [7:07:42<15:38,  3.58it/s]
 99%|█████████▉| 368118/371472 [7:07:42<15:10,  3.68it/s]
 99%|█████████▉| 368119/371472 [7:07:43<15:02,  3.71it/s]
 99%|█████████▉| 368120/371472 [7:07:43<14:48,  3.77it/s]
                                                         
{'loss': 2.8156, 'learning_rate': 1.0812557909026271e-07, 'epoch': 15.86}
+
 99%|█████████▉| 368120/371472 [7:07:43<14:48,  3.77it/s]
 99%|█████████▉| 368121/371472 [7:07:43<14:59,  3.73it/s]
 99%|█████████▉| 368122/371472 [7:07:44<15:32,  3.59it/s]
 99%|█████████▉| 368123/371472 [7:07:44<15:09,  3.68it/s]
 99%|█████████▉| 368124/371472 [7:07:44<15:07,  3.69it/s]
 99%|█████████▉| 368125/371472 [7:07:44<15:40,  3.56it/s]
 99%|█████████▉| 368126/371472 [7:07:45<15:20,  3.63it/s]
 99%|█████████▉| 368127/371472 [7:07:45<15:09,  3.68it/s]
 99%|█████████▉| 368128/371472 [7:07:45<14:56,  3.73it/s]
 99%|█████████▉| 368129/371472 [7:07:45<15:26,  3.61it/s]
 99%|█████████▉| 368130/371472 [7:07:46<15:12,  3.66it/s]
 99%|█████████▉| 368131/371472 [7:07:46<15:42,  3.55it/s]
 99%|█████████▉| 368132/371472 [7:07:46<16:15,  3.42it/s]
 99%|█████████▉| 368133/371472 [7:07:47<15:39,  3.55it/s]
 99%|█████████▉| 368134/371472 [7:07:47<15:20,  3.63it/s]
 99%|█████████▉| 368135/371472 [7:07:47<14:47,  3.76it/s]
 99%|█████████▉| 368136/371472 [7:07:47<15:01,  3.70it/s]
 99%|█████████▉| 368137/371472 [7:07:48<14:45,  3.76it/s]
 99%|█████████▉| 368138/371472 [7:07:48<16:18,  3.41it/s]
 99%|█████████▉| 368139/371472 [7:07:48<15:54,  3.49it/s]
 99%|█████████▉| 368140/371472 [7:07:49<16:05,  3.45it/s]
                                                         
{'loss': 2.5233, 'learning_rate': 1.0807709711478377e-07, 'epoch': 15.86}
+
 99%|█████████▉| 368140/371472 [7:07:49<16:05,  3.45it/s]
 99%|█████████▉| 368141/371472 [7:07:49<16:05,  3.45it/s]
 99%|█████████▉| 368142/371472 [7:07:49<16:19,  3.40it/s]
 99%|█████████▉| 368143/371472 [7:07:49<16:48,  3.30it/s]
 99%|█████████▉| 368144/371472 [7:07:50<17:21,  3.20it/s]
 99%|█████████▉| 368145/371472 [7:07:50<16:23,  3.38it/s]
 99%|█████████▉| 368146/371472 [7:07:50<16:20,  3.39it/s]
 99%|█████████▉| 368147/371472 [7:07:51<15:52,  3.49it/s]
 99%|█████████▉| 368148/371472 [7:07:51<15:45,  3.52it/s]
 99%|█████████▉| 368149/371472 [7:07:51<15:08,  3.66it/s]
 99%|█████████▉| 368150/371472 [7:07:51<14:54,  3.71it/s]
 99%|█████████▉| 368151/371472 [7:07:52<15:08,  3.66it/s]
 99%|█████████▉| 368152/371472 [7:07:52<15:36,  3.55it/s]
 99%|█████████▉| 368153/371472 [7:07:52<16:10,  3.42it/s]
 99%|█████████▉| 368154/371472 [7:07:53<15:22,  3.60it/s]
 99%|█████████▉| 368155/371472 [7:07:53<14:49,  3.73it/s]
 99%|█████████▉| 368156/371472 [7:07:53<14:59,  3.69it/s]
 99%|█████████▉| 368157/371472 [7:07:53<14:54,  3.70it/s]
 99%|█████████▉| 368158/371472 [7:07:54<15:07,  3.65it/s]
 99%|█████████▉| 368159/371472 [7:07:54<14:50,  3.72it/s]
 99%|█████████▉| 368160/371472 [7:07:54<15:24,  3.58it/s]
                                                         
{'loss': 2.5476, 'learning_rate': 1.0802861513930483e-07, 'epoch': 15.86}
+
 99%|█████████▉| 368160/371472 [7:07:54<15:24,  3.58it/s]
 99%|█████████▉| 368161/371472 [7:07:54<14:56,  3.69it/s]
 99%|█████████▉| 368162/371472 [7:07:55<14:52,  3.71it/s]
 99%|█████████▉| 368163/371472 [7:07:55<15:19,  3.60it/s]
 99%|█████████▉| 368164/371472 [7:07:55<15:30,  3.55it/s]
 99%|█████████▉| 368165/371472 [7:07:56<15:13,  3.62it/s]
 99%|█████████▉| 368166/371472 [7:07:56<15:28,  3.56it/s]
 99%|█████████▉| 368167/371472 [7:07:56<15:57,  3.45it/s]
 99%|█████████▉| 368168/371472 [7:07:56<15:59,  3.44it/s]
 99%|█████████▉| 368169/371472 [7:07:57<15:32,  3.54it/s]
 99%|█████████▉| 368170/371472 [7:07:57<15:01,  3.66it/s]
 99%|█████████▉| 368171/371472 [7:07:57<15:37,  3.52it/s]
 99%|█████████▉| 368172/371472 [7:07:58<15:47,  3.48it/s]
 99%|█████████▉| 368173/371472 [7:07:58<15:13,  3.61it/s]
 99%|█████████▉| 368174/371472 [7:07:58<15:07,  3.63it/s]
 99%|█████████▉| 368175/371472 [7:07:58<15:37,  3.52it/s]
 99%|█████████▉| 368176/371472 [7:07:59<15:17,  3.59it/s]
 99%|█████████▉| 368177/371472 [7:07:59<15:21,  3.58it/s]
 99%|█████████▉| 368178/371472 [7:07:59<16:07,  3.41it/s]
 99%|█████████▉| 368179/371472 [7:08:00<15:21,  3.57it/s]
 99%|█████████▉| 368180/371472 [7:08:00<15:22,  3.57it/s]
                                                         
{'loss': 2.5079, 'learning_rate': 1.0798013316382598e-07, 'epoch': 15.86}
+
 99%|█████████▉| 368180/371472 [7:08:00<15:22,  3.57it/s]
 99%|█████████▉| 368181/371472 [7:08:00<15:13,  3.60it/s]
 99%|█████████▉| 368182/371472 [7:08:00<15:13,  3.60it/s]
 99%|█████████▉| 368183/371472 [7:08:01<15:33,  3.53it/s]
 99%|█████████▉| 368184/371472 [7:08:01<15:07,  3.62it/s]
 99%|█████████▉| 368185/371472 [7:08:01<16:32,  3.31it/s]
 99%|█████████▉| 368186/371472 [7:08:02<15:53,  3.45it/s]
 99%|█████████▉| 368187/371472 [7:08:02<15:40,  3.49it/s]
 99%|█████████▉| 368188/371472 [7:08:02<15:27,  3.54it/s]
 99%|█████████▉| 368189/371472 [7:08:02<16:05,  3.40it/s]
 99%|█████████▉| 368190/371472 [7:08:03<16:09,  3.39it/s]
 99%|█████████▉| 368191/371472 [7:08:03<15:39,  3.49it/s]
 99%|█████████▉| 368192/371472 [7:08:03<15:24,  3.55it/s]
 99%|█████████▉| 368193/371472 [7:08:04<15:37,  3.50it/s]
 99%|█████████▉| 368194/371472 [7:08:04<15:05,  3.62it/s]
 99%|█████████▉| 368195/371472 [7:08:04<14:32,  3.75it/s]
 99%|█████████▉| 368196/371472 [7:08:04<14:35,  3.74it/s]
 99%|█████████▉| 368197/371472 [7:08:05<14:15,  3.83it/s]
 99%|█████████▉| 368198/371472 [7:08:05<14:17,  3.82it/s]
 99%|█████████▉| 368199/371472 [7:08:05<14:13,  3.83it/s]
 99%|█████████▉| 368200/371472 [7:08:05<14:56,  3.65it/s]
                                                         
{'loss': 2.3992, 'learning_rate': 1.0793165118834703e-07, 'epoch': 15.86}
+
 99%|█████████▉| 368200/371472 [7:08:05<14:56,  3.65it/s]
 99%|█████████▉| 368201/371472 [7:08:06<14:42,  3.71it/s]
 99%|█████████▉| 368202/371472 [7:08:06<14:54,  3.66it/s]
 99%|█████████▉| 368203/371472 [7:08:06<15:39,  3.48it/s]
 99%|█████████▉| 368204/371472 [7:08:07<15:22,  3.54it/s]
 99%|█████████▉| 368205/371472 [7:08:07<14:57,  3.64it/s]
 99%|█████████▉| 368206/371472 [7:08:07<15:42,  3.47it/s]
 99%|█████████▉| 368207/371472 [7:08:07<16:02,  3.39it/s]
 99%|█████████▉| 368208/371472 [7:08:08<15:59,  3.40it/s]
 99%|█████████▉| 368209/371472 [7:08:08<15:13,  3.57it/s]
 99%|█████████▉| 368210/371472 [7:08:08<16:22,  3.32it/s]
 99%|█████████▉| 368211/371472 [7:08:09<16:22,  3.32it/s]
 99%|█████████▉| 368212/371472 [7:08:09<16:29,  3.29it/s]
 99%|█████████▉| 368213/371472 [7:08:09<15:52,  3.42it/s]
 99%|█████████▉| 368214/371472 [7:08:09<15:56,  3.41it/s]
 99%|█████████▉| 368215/371472 [7:08:10<20:35,  2.64it/s]
 99%|█████████▉| 368216/371472 [7:08:10<19:04,  2.84it/s]
 99%|█████████▉| 368217/371472 [7:08:11<17:19,  3.13it/s]
 99%|█████████▉| 368218/371472 [7:08:11<16:25,  3.30it/s]
 99%|█████████▉| 368219/371472 [7:08:11<15:26,  3.51it/s]
 99%|█████████▉| 368220/371472 [7:08:11<15:33,  3.48it/s]
                                                         
{'loss': 2.5286, 'learning_rate': 1.078831692128682e-07, 'epoch': 15.86}
+
 99%|█████████▉| 368220/371472 [7:08:11<15:33,  3.48it/s]
 99%|█████████▉| 368221/371472 [7:08:12<15:30,  3.49it/s]
 99%|█████████▉| 368222/371472 [7:08:12<15:08,  3.58it/s]
 99%|█████████▉| 368223/371472 [7:08:12<15:13,  3.56it/s]
 99%|█████████▉| 368224/371472 [7:08:13<15:54,  3.40it/s]
 99%|█████████▉| 368225/371472 [7:08:13<16:07,  3.36it/s]
 99%|███���█████▉| 368226/371472 [7:08:13<16:05,  3.36it/s]
 99%|█████████▉| 368227/371472 [7:08:14<17:16,  3.13it/s]
 99%|█████████▉| 368228/371472 [7:08:14<17:11,  3.14it/s]
 99%|█████████▉| 368229/371472 [7:08:14<16:14,  3.33it/s]
 99%|█████████▉| 368230/371472 [7:08:14<15:18,  3.53it/s]
 99%|█████████▉| 368231/371472 [7:08:15<15:32,  3.48it/s]
 99%|█████████▉| 368232/371472 [7:08:15<14:55,  3.62it/s]
 99%|█████████▉| 368233/371472 [7:08:15<14:51,  3.63it/s]
 99%|█████████▉| 368234/371472 [7:08:15<14:48,  3.64it/s]
 99%|█████████▉| 368235/371472 [7:08:16<14:54,  3.62it/s]
 99%|█████████▉| 368236/371472 [7:08:16<15:36,  3.45it/s]
 99%|█████████▉| 368237/371472 [7:08:16<15:25,  3.50it/s]
 99%|█████████▉| 368238/371472 [7:08:17<14:47,  3.65it/s]
 99%|█████████▉| 368239/371472 [7:08:17<15:30,  3.47it/s]
 99%|█████████▉| 368240/371472 [7:08:17<15:08,  3.56it/s]
                                                         
{'loss': 2.484, 'learning_rate': 1.0783468723738926e-07, 'epoch': 15.86}
+
 99%|█████████▉| 368240/371472 [7:08:17<15:08,  3.56it/s]
 99%|█████████▉| 368241/371472 [7:08:17<14:58,  3.59it/s]
 99%|█████████▉| 368242/371472 [7:08:18<14:43,  3.65it/s]
 99%|█████████▉| 368243/371472 [7:08:18<14:32,  3.70it/s]
 99%|█████████▉| 368244/371472 [7:08:18<14:51,  3.62it/s]
 99%|█████████▉| 368245/371472 [7:08:18<14:39,  3.67it/s]
 99%|█████████▉| 368246/371472 [7:08:19<14:52,  3.61it/s]
 99%|█████████▉| 368247/371472 [7:08:19<15:05,  3.56it/s]
 99%|█████████▉| 368248/371472 [7:08:19<16:07,  3.33it/s]
 99%|█████████▉| 368249/371472 [7:08:20<15:28,  3.47it/s]
 99%|█████████▉| 368250/371472 [7:08:20<15:08,  3.55it/s]
 99%|█████████▉| 368251/371472 [7:08:20<14:39,  3.66it/s]
 99%|█████████▉| 368252/371472 [7:08:20<15:07,  3.55it/s]
 99%|█████████▉| 368253/371472 [7:08:21<15:34,  3.44it/s]
 99%|█████████▉| 368254/371472 [7:08:21<15:23,  3.48it/s]
 99%|█████████▉| 368255/371472 [7:08:21<14:49,  3.62it/s]
 99%|█████████▉| 368256/371472 [7:08:22<14:55,  3.59it/s]
 99%|█████████▉| 368257/371472 [7:08:22<14:54,  3.59it/s]
 99%|█████████▉| 368258/371472 [7:08:22<17:17,  3.10it/s]
 99%|█████████▉| 368259/371472 [7:08:23<17:36,  3.04it/s]
 99%|█████████▉| 368260/371472 [7:08:23<16:53,  3.17it/s]
                                                         
{'loss': 2.5445, 'learning_rate': 1.077862052619104e-07, 'epoch': 15.86}
+
 99%|█████████▉| 368260/371472 [7:08:23<16:53,  3.17it/s]
 99%|█████████▉| 368261/371472 [7:08:23<16:33,  3.23it/s]
 99%|█████████▉| 368262/371472 [7:08:24<16:48,  3.18it/s]
 99%|█████████▉| 368263/371472 [7:08:24<16:32,  3.23it/s]
 99%|█████████▉| 368264/371472 [7:08:24<16:15,  3.29it/s]
 99%|█████████▉| 368265/371472 [7:08:24<15:43,  3.40it/s]
 99%|█████████▉| 368266/371472 [7:08:25<15:45,  3.39it/s]
 99%|█████████▉| 368267/371472 [7:08:25<15:36,  3.42it/s]
 99%|█████████▉| 368268/371472 [7:08:25<15:33,  3.43it/s]
 99%|█████████▉| 368269/371472 [7:08:26<15:19,  3.48it/s]
 99%|█████████▉| 368270/371472 [7:08:26<15:44,  3.39it/s]
 99%|█████████▉| 368271/371472 [7:08:26<15:36,  3.42it/s]
 99%|█████████▉| 368272/371472 [7:08:26<14:59,  3.56it/s]
 99%|█████████▉| 368273/371472 [7:08:27<15:05,  3.53it/s]
 99%|█████████▉| 368274/371472 [7:08:27<14:50,  3.59it/s]
 99%|█████████▉| 368275/371472 [7:08:27<14:21,  3.71it/s]
 99%|█████████▉| 368276/371472 [7:08:27<14:11,  3.75it/s]
 99%|█████████▉| 368277/371472 [7:08:28<15:46,  3.38it/s]
 99%|█████████▉| 368278/371472 [7:08:28<15:32,  3.43it/s]
 99%|█████████▉| 368279/371472 [7:08:28<15:27,  3.44it/s]
 99%|█████████▉| 368280/371472 [7:08:29<16:03,  3.31it/s]
                                                         
{'loss': 2.6825, 'learning_rate': 1.0773772328643147e-07, 'epoch': 15.86}
+
 99%|█████████▉| 368280/371472 [7:08:29<16:03,  3.31it/s]
 99%|█████████▉| 368281/371472 [7:08:29<15:31,  3.43it/s]
 99%|█████████▉| 368282/371472 [7:08:29<15:54,  3.34it/s]
 99%|█████████▉| 368283/371472 [7:08:30<16:14,  3.27it/s]
 99%|█████████▉| 368284/371472 [7:08:30<15:33,  3.41it/s]
 99%|█████████▉| 368285/371472 [7:08:30<15:04,  3.52it/s]
 99%|█████████▉| 368286/371472 [7:08:30<15:04,  3.52it/s]
 99%|█████████▉| 368287/371472 [7:08:31<15:32,  3.42it/s]
 99%|█████████▉| 368288/371472 [7:08:31<15:15,  3.48it/s]
 99%|█████████▉| 368289/371472 [7:08:31<14:56,  3.55it/s]
 99%|█████████▉| 368290/371472 [7:08:32<14:35,  3.63it/s]
 99%|█████████▉| 368291/371472 [7:08:32<14:51,  3.57it/s]
 99%|█████████▉| 368292/371472 [7:08:32<15:11,  3.49it/s]
 99%|█████████▉| 368293/371472 [7:08:33<16:05,  3.29it/s]
 99%|█████████▉| 368294/371472 [7:08:33<15:19,  3.46it/s]
 99%|█████████▉| 368295/371472 [7:08:33<15:42,  3.37it/s]
 99%|█████████▉| 368296/371472 [7:08:33<15:45,  3.36it/s]
 99%|█████████▉| 368297/371472 [7:08:34<15:31,  3.41it/s]
 99%|█████████▉| 368298/371472 [7:08:34<15:29,  3.42it/s]
 99%|█████████▉| 368299/371472 [7:08:34<14:58,  3.53it/s]
 99%|█████████▉| 368300/371472 [7:08:35<14:48,  3.57it/s]
                                                         
{'loss': 2.5596, 'learning_rate': 1.0768924131095263e-07, 'epoch': 15.86}
+
 99%|█████████▉| 368300/371472 [7:08:35<14:48,  3.57it/s]
 99%|█████████▉| 368301/371472 [7:08:35<15:14,  3.47it/s]
 99%|█████████▉| 368302/371472 [7:08:35<15:10,  3.48it/s]
 99%|█████████▉| 368303/371472 [7:08:35<15:00,  3.52it/s]
 99%|█████████▉| 368304/371472 [7:08:36<14:23,  3.67it/s]
 99%|█████████▉| 368305/371472 [7:08:36<13:57,  3.78it/s]
 99%|█████████▉| 368306/371472 [7:08:36<14:12,  3.71it/s]
 99%|█████████▉| 368307/371472 [7:08:36<14:01,  3.76it/s]
 99%|█████████▉| 368308/371472 [7:08:37<14:02,  3.76it/s]
 99%|█████████▉| 368309/371472 [7:08:37<13:59,  3.77it/s]
 99%|█████████▉| 368310/371472 [7:08:37<13:39,  3.86it/s]
 99%|█████████▉| 368311/371472 [7:08:37<13:42,  3.84it/s]
 99%|█████████▉| 368312/371472 [7:08:38<13:43,  3.84it/s]
 99%|█████████▉| 368313/371472 [7:08:38<13:44,  3.83it/s]
 99%|█████████▉| 368314/371472 [7:08:38<13:23,  3.93it/s]
 99%|█████████▉| 368315/371472 [7:08:38<13:27,  3.91it/s]
 99%|█████████▉| 368316/371472 [7:08:39<13:34,  3.87it/s]
 99%|█████████▉| 368317/371472 [7:08:39<14:38,  3.59it/s]
 99%|█████████▉| 368318/371472 [7:08:39<16:39,  3.16it/s]
 99%|█████████▉| 368319/371472 [7:08:40<15:42,  3.34it/s]
 99%|█████████▉| 368320/371472 [7:08:40<15:32,  3.38it/s]
                                                         
{'loss': 2.6674, 'learning_rate': 1.0764075933547369e-07, 'epoch': 15.86}
+
 99%|█████████▉| 368320/371472 [7:08:40<15:32,  3.38it/s]
 99%|█████████▉| 368321/371472 [7:08:40<16:16,  3.23it/s]
 99%|█████████▉| 368322/371472 [7:08:41<15:20,  3.42it/s]
 99%|█████████▉| 368323/371472 [7:08:41<15:30,  3.39it/s]
 99%|█████████▉| 368324/371472 [7:08:41<15:05,  3.48it/s]
 99%|█████████▉| 368325/371472 [7:08:42<15:42,  3.34it/s]
 99%|█████████▉| 368326/371472 [7:08:42<16:37,  3.15it/s]
 99%|█████████▉| 368327/371472 [7:08:42<15:48,  3.32it/s]
 99%|█████████▉| 368328/371472 [7:08:42<16:53,  3.10it/s]
 99%|█████████▉| 368329/371472 [7:08:43<16:01,  3.27it/s]
 99%|█████████▉| 368330/371472 [7:08:43<15:46,  3.32it/s]
 99%|█████████▉| 368331/371472 [7:08:43<15:16,  3.43it/s]
 99%|█████████▉| 368332/371472 [7:08:44<15:13,  3.44it/s]
 99%|█████████▉| 368333/371472 [7:08:44<14:44,  3.55it/s]
 99%|█████████▉| 368334/371472 [7:08:44<14:35,  3.58it/s]
 99%|█████████▉| 368335/371472 [7:08:44<14:52,  3.52it/s]
 99%|█████████▉| 368336/371472 [7:08:45<15:15,  3.43it/s]
 99%|█████████▉| 368337/371472 [7:08:45<16:10,  3.23it/s]
 99%|█████████▉| 368338/371472 [7:08:45<16:28,  3.17it/s]
 99%|█████████▉| 368339/371472 [7:08:46<15:36,  3.34it/s]
 99%|█████████▉| 368340/371472 [7:08:46<15:19,  3.41it/s]
                                                         
{'loss': 2.4682, 'learning_rate': 1.0759227735999484e-07, 'epoch': 15.87}
+
 99%|█████████▉| 368340/371472 [7:08:46<15:19,  3.41it/s]
 99%|███���█████▉| 368341/371472 [7:08:46<15:39,  3.33it/s]
 99%|█████████▉| 368342/371472 [7:08:47<15:38,  3.34it/s]
 99%|█████████▉| 368343/371472 [7:08:47<14:42,  3.55it/s]
 99%|█████████▉| 368344/371472 [7:08:47<14:06,  3.69it/s]
 99%|█████████▉| 368345/371472 [7:08:47<13:58,  3.73it/s]
 99%|█████████▉| 368346/371472 [7:08:48<15:17,  3.41it/s]
 99%|█████████▉| 368347/371472 [7:08:48<14:56,  3.48it/s]
 99%|█████████▉| 368348/371472 [7:08:48<15:02,  3.46it/s]
 99%|█████████▉| 368349/371472 [7:08:48<14:18,  3.64it/s]
 99%|█████████▉| 368350/371472 [7:08:49<14:02,  3.71it/s]
 99%|█████████▉| 368351/371472 [7:08:49<14:27,  3.60it/s]
 99%|█████████▉| 368352/371472 [7:08:49<13:56,  3.73it/s]
 99%|█████████▉| 368353/371472 [7:08:50<13:38,  3.81it/s]
 99%|█████████▉| 368354/371472 [7:08:50<13:18,  3.91it/s]
 99%|█████████▉| 368355/371472 [7:08:50<14:23,  3.61it/s]
 99%|█████████▉| 368356/371472 [7:08:50<14:26,  3.60it/s]
 99%|█████████▉| 368357/371472 [7:08:51<14:47,  3.51it/s]
 99%|█████████▉| 368358/371472 [7:08:51<15:08,  3.43it/s]
 99%|█████████▉| 368359/371472 [7:08:51<15:33,  3.33it/s]
 99%|█████████▉| 368360/371472 [7:08:52<15:10,  3.42it/s]
                                                         
{'loss': 2.7336, 'learning_rate': 1.075437953845159e-07, 'epoch': 15.87}
+
 99%|█████████▉| 368360/371472 [7:08:52<15:10,  3.42it/s]
 99%|█████████▉| 368361/371472 [7:08:52<14:44,  3.52it/s]
 99%|█████████▉| 368362/371472 [7:08:52<15:26,  3.36it/s]
 99%|█████████▉| 368363/371472 [7:08:52<15:02,  3.45it/s]
 99%|█████████▉| 368364/371472 [7:08:53<14:12,  3.64it/s]
 99%|█████████▉| 368365/371472 [7:08:53<14:39,  3.53it/s]
 99%|█████████▉| 368366/371472 [7:08:53<14:28,  3.58it/s]
 99%|█████████▉| 368367/371472 [7:08:54<15:16,  3.39it/s]
 99%|█████████▉| 368368/371472 [7:08:54<14:42,  3.52it/s]
 99%|█████████▉| 368369/371472 [7:08:54<14:48,  3.49it/s]
 99%|█████████▉| 368370/371472 [7:08:54<14:27,  3.58it/s]
 99%|█████████▉| 368371/371472 [7:08:55<14:24,  3.59it/s]
 99%|█████████▉| 368372/371472 [7:08:55<14:51,  3.48it/s]
 99%|█████████▉| 368373/371472 [7:08:55<14:51,  3.48it/s]
 99%|█████████▉| 368374/371472 [7:08:56<14:33,  3.55it/s]
 99%|█████████▉| 368375/371472 [7:08:56<15:07,  3.41it/s]
 99%|█████████▉| 368376/371472 [7:08:56<15:36,  3.31it/s]
 99%|█████████▉| 368377/371472 [7:08:57<15:52,  3.25it/s]
 99%|█████████▉| 368378/371472 [7:08:57<15:34,  3.31it/s]
 99%|█████████▉| 368379/371472 [7:08:57<15:53,  3.24it/s]
 99%|█████████▉| 368380/371472 [7:08:57<16:18,  3.16it/s]
                                                         
{'loss': 2.5975, 'learning_rate': 1.0749531340903706e-07, 'epoch': 15.87}
+
 99%|█████████▉| 368380/371472 [7:08:57<16:18,  3.16it/s]
 99%|█████████▉| 368381/371472 [7:08:58<16:12,  3.18it/s]
 99%|█████████▉| 368382/371472 [7:08:58<15:22,  3.35it/s]
 99%|█████████▉| 368383/371472 [7:08:58<14:43,  3.50it/s]
 99%|█████████▉| 368384/371472 [7:08:59<15:08,  3.40it/s]
 99%|█████████▉| 368385/371472 [7:08:59<16:13,  3.17it/s]
 99%|█████████▉| 368386/371472 [7:08:59<15:45,  3.26it/s]
 99%|█████████▉| 368387/371472 [7:09:00<15:17,  3.36it/s]
 99%|█████████▉| 368388/371472 [7:09:00<15:21,  3.35it/s]
 99%|█████████▉| 368389/371472 [7:09:00<15:13,  3.38it/s]
 99%|█████████▉| 368390/371472 [7:09:00<14:58,  3.43it/s]
 99%|█████████▉| 368391/371472 [7:09:01<14:31,  3.54it/s]
 99%|█████████▉| 368392/371472 [7:09:01<13:52,  3.70it/s]
 99%|█████████▉| 368393/371472 [7:09:01<13:59,  3.67it/s]
 99%|█████████▉| 368394/371472 [7:09:02<14:29,  3.54it/s]
 99%|█████████▉| 368395/371472 [7:09:02<14:41,  3.49it/s]
 99%|█████████▉| 368396/371472 [7:09:02<14:20,  3.58it/s]
 99%|█████████▉| 368397/371472 [7:09:02<14:26,  3.55it/s]
 99%|█████████▉| 368398/371472 [7:09:03<14:41,  3.49it/s]
 99%|█████████▉| 368399/371472 [7:09:03<14:50,  3.45it/s]
 99%|█████████▉| 368400/371472 [7:09:03<15:02,  3.41it/s]
                                                         
{'loss': 2.6667, 'learning_rate': 1.0744683143355812e-07, 'epoch': 15.87}
+
 99%|█████████▉| 368400/371472 [7:09:03<15:02,  3.41it/s]
 99%|█████████▉| 368401/371472 [7:09:04<15:38,  3.27it/s]
 99%|█████████▉| 368402/371472 [7:09:04<14:58,  3.42it/s]
 99%|█████████▉| 368403/371472 [7:09:04<14:33,  3.51it/s]
 99%|█████████▉| 368404/371472 [7:09:04<14:18,  3.57it/s]
 99%|█████████▉| 368405/371472 [7:09:05<15:10,  3.37it/s]
 99%|█████████▉| 368406/371472 [7:09:05<14:33,  3.51it/s]
 99%|█████████▉| 368407/371472 [7:09:05<14:10,  3.61it/s]
 99%|█████████▉| 368408/371472 [7:09:05<13:40,  3.73it/s]
 99%|█████████▉| 368409/371472 [7:09:06<13:23,  3.81it/s]
 99%|█████████▉| 368410/371472 [7:09:06<13:51,  3.68it/s]
 99%|█████████▉| 368411/371472 [7:09:06<16:05,  3.17it/s]
 99%|█████████▉| 368412/371472 [7:09:07<15:41,  3.25it/s]
 99%|█████████▉| 368413/371472 [7:09:07<15:25,  3.30it/s]
 99%|█████████▉| 368414/371472 [7:09:07<15:00,  3.40it/s]
 99%|█████████▉| 368415/371472 [7:09:08<15:08,  3.37it/s]
 99%|█████████▉| 368416/371472 [7:09:08<14:28,  3.52it/s]
 99%|█████████▉| 368417/371472 [7:09:08<14:12,  3.58it/s]
 99%|█████████▉| 368418/371472 [7:09:08<14:04,  3.62it/s]
 99%|█████████▉| 368419/371472 [7:09:09<14:44,  3.45it/s]
 99%|█████████▉| 368420/371472 [7:09:09<14:31,  3.50it/s]
                                                         
{'loss': 2.514, 'learning_rate': 1.0739834945807927e-07, 'epoch': 15.87}
+
 99%|█████████▉| 368420/371472 [7:09:09<14:31,  3.50it/s]
 99%|█████████▉| 368421/371472 [7:09:09<14:16,  3.56it/s]
 99%|█████████▉| 368422/371472 [7:09:10<14:06,  3.60it/s]
 99%|█████████▉| 368423/371472 [7:09:10<14:03,  3.61it/s]
 99%|█████████▉| 368424/371472 [7:09:10<14:09,  3.59it/s]
 99%|█████████▉| 368425/371472 [7:09:10<13:50,  3.67it/s]
 99%|█████████▉| 368426/371472 [7:09:11<13:15,  3.83it/s]
 99%|█████████▉| 368427/371472 [7:09:11<12:52,  3.94it/s]
 99%|█████████▉| 368428/371472 [7:09:11<13:27,  3.77it/s]
 99%|█████████▉| 368429/371472 [7:09:11<14:15,  3.56it/s]
 99%|█████████▉| 368430/371472 [7:09:12<15:12,  3.33it/s]
 99%|█████████▉| 368431/371472 [7:09:12<14:55,  3.39it/s]
 99%|█████████▉| 368432/371472 [7:09:12<14:34,  3.48it/s]
 99%|█████████▉| 368433/371472 [7:09:13<13:50,  3.66it/s]
 99%|█████████▉| 368434/371472 [7:09:13<13:18,  3.80it/s]
 99%|█████████▉| 368435/371472 [7:09:13<13:23,  3.78it/s]
 99%|█████████▉| 368436/371472 [7:09:13<13:08,  3.85it/s]
 99%|█████████▉| 368437/371472 [7:09:14<13:26,  3.76it/s]
 99%|█████████▉| 368438/371472 [7:09:14<13:28,  3.75it/s]
 99%|█████████▉| 368439/371472 [7:09:14<13:54,  3.64it/s]
 99%|█████████▉| 368440/371472 [7:09:14<13:37,  3.71it/s]
                                                         
{'loss': 2.6543, 'learning_rate': 1.0734986748260033e-07, 'epoch': 15.87}
+
 99%|█████████▉| 368440/371472 [7:09:14<13:37,  3.71it/s]
 99%|█████████▉| 368441/371472 [7:09:15<13:43,  3.68it/s]
 99%|█████████▉| 368442/371472 [7:09:15<13:22,  3.78it/s]
 99%|█████████▉| 368443/371472 [7:09:15<15:07,  3.34it/s]
 99%|█████████▉| 368444/371472 [7:09:16<15:15,  3.31it/s]
 99%|█████████▉| 368445/371472 [7:09:16<14:35,  3.46it/s]
 99%|█████████▉| 368446/371472 [7:09:16<14:56,  3.38it/s]
 99%|█████████▉| 368447/371472 [7:09:17<15:05,  3.34it/s]
 99%|█████████▉| 368448/371472 [7:09:17<14:29,  3.48it/s]
 99%|█████████▉| 368449/371472 [7:09:17<14:21,  3.51it/s]
 99%|█████████▉| 368450/371472 [7:09:17<13:58,  3.60it/s]
 99%|█████████▉| 368451/371472 [7:09:18<13:25,  3.75it/s]
 99%|█████████▉| 368452/371472 [7:09:18<13:20,  3.77it/s]
 99%|█████████▉| 368453/371472 [7:09:18<15:08,  3.32it/s]
 99%|█████████▉| 368454/371472 [7:09:18<15:01,  3.35it/s]
 99%|█████████▉| 368455/371472 [7:09:19<14:47,  3.40it/s]
 99%|█████████▉| 368456/371472 [7:09:19<14:22,  3.49it/s]
 99%|█████████▉| 368457/371472 [7:09:19<14:37,  3.44it/s]
 99%|█████████▉| 368458/371472 [7:09:20<14:08,  3.55it/s]
 99%|█████████▉| 368459/371472 [7:09:20<13:52,  3.62it/s]
 99%|█████████▉| 368460/371472 [7:09:20<14:13,  3.53it/s]
                                                         
{'loss': 2.5126, 'learning_rate': 1.073013855071215e-07, 'epoch': 15.87}
+
 99%|█████████▉| 368460/371472 [7:09:20<14:13,  3.53it/s]
 99%|█████████▉| 368461/371472 [7:09:20<13:54,  3.61it/s]
 99%|█████████▉| 368462/371472 [7:09:21<15:36,  3.21it/s]
 99%|█████████▉| 368463/371472 [7:09:21<15:02,  3.33it/s]
 99%|█████████▉| 368464/371472 [7:09:21<14:46,  3.39it/s]
 99%|█████████▉| 368465/371472 [7:09:22<14:10,  3.54it/s]
 99%|█████████▉| 368466/371472 [7:09:22<14:05,  3.55it/s]
 99%|█████████▉| 368467/371472 [7:09:22<14:07,  3.55it/s]
 99%|█████████▉| 368468/371472 [7:09:22<13:37,  3.67it/s]
 99%|█████████▉| 368469/371472 [7:09:23<13:22,  3.74it/s]
 99%|█████████▉| 368470/371472 [7:09:23<13:47,  3.63it/s]
 99%|█████████▉| 368471/371472 [7:09:23<13:31,  3.70it/s]
 99%|█████████▉| 368472/371472 [7:09:24<13:55,  3.59it/s]
 99%|█████████▉| 368473/371472 [7:09:24<15:22,  3.25it/s]
 99%|█████████▉| 368474/371472 [7:09:24<15:38,  3.19it/s]
 99%|█████████▉| 368475/371472 [7:09:25<15:17,  3.27it/s]
 99%|█████████▉| 368476/371472 [7:09:25<14:44,  3.39it/s]
 99%|█████████▉| 368477/371472 [7:09:25<14:36,  3.42it/s]
 99%|█████████▉| 368478/371472 [7:09:25<15:55,  3.13it/s]
 99%|█████████▉| 368479/371472 [7:09:26<14:49,  3.37it/s]
 99%|█████████▉| 368480/371472 [7:09:26<14:45,  3.38it/s]
                                                         
{'loss': 2.5822, 'learning_rate': 1.0725290353164255e-07, 'epoch': 15.87}
+
 99%|█████████▉| 368480/371472 [7:09:26<14:45,  3.38it/s]
 99%|█████████▉| 368481/371472 [7:09:26<14:52,  3.35it/s]
 99%|█████████▉| 368482/371472 [7:09:27<14:59,  3.32it/s]
 99%|█████████▉| 368483/371472 [7:09:27<14:21,  3.47it/s]
 99%|█████████▉| 368484/371472 [7:09:27<13:58,  3.56it/s]
 99%|█████████▉| 368485/371472 [7:09:27<13:51,  3.59it/s]
 99%|█████████▉| 368486/371472 [7:09:28<14:17,  3.48it/s]
 99%|█████████▉| 368487/371472 [7:09:28<13:55,  3.57it/s]
 99%|█████████▉| 368488/371472 [7:09:28<13:54,  3.58it/s]
 99%|█████████▉| 368489/371472 [7:09:29<14:44,  3.37it/s]
 99%|█████████▉| 368490/371472 [7:09:29<14:13,  3.49it/s]
 99%|█████████▉| 368491/371472 [7:09:29<14:24,  3.45it/s]
 99%|█████████▉| 368492/371472 [7:09:29<13:58,  3.55it/s]
 99%|█████████▉| 368493/371472 [7:09:30<13:36,  3.65it/s]
 99%|█████████▉| 368494/371472 [7:09:30<14:00,  3.54it/s]
 99%|█████████▉| 368495/371472 [7:09:30<13:25,  3.69it/s]
 99%|█████████▉| 368496/371472 [7:09:31<13:40,  3.63it/s]
 99%|█████████▉| 368497/371472 [7:09:31<14:24,  3.44it/s]
 99%|█████████▉| 368498/371472 [7:09:31<14:39,  3.38it/s]
 99%|█████████▉| 368499/371472 [7:09:31<14:34,  3.40it/s]
 99%|█████████▉| 368500/371472 [7:09:32<13:54,  3.56it/s]
                                                         
{'loss': 2.5354, 'learning_rate': 1.072044215561637e-07, 'epoch': 15.87}
+
 99%|█████████▉| 368500/371472 [7:09:32<13:54,  3.56it/s]
 99%|█████████▉| 368501/371472 [7:09:32<14:00,  3.53it/s]
 99%|█████████▉| 368502/371472 [7:09:32<14:07,  3.50it/s]
 99%|█████████▉| 368503/371472 [7:09:33<14:07,  3.50it/s]
 99%|█████████▉| 368504/371472 [7:09:33<13:55,  3.55it/s]
 99%|█████████▉| 368505/371472 [7:09:33<13:25,  3.68it/s]
 99%|█████████▉| 368506/371472 [7:09:33<13:05,  3.78it/s]
 99%|█████████▉| 368507/371472 [7:09:34<12:45,  3.87it/s]
 99%|█████████▉| 368508/371472 [7:09:34<14:08,  3.49it/s]
 99%|█████████▉| 368509/371472 [7:09:34<13:31,  3.65it/s]
 99%|█████████▉| 368510/371472 [7:09:34<13:43,  3.60it/s]
 99%|█████████▉| 368511/371472 [7:09:35<14:05,  3.50it/s]
 99%|█████████▉| 368512/371472 [7:09:35<14:01,  3.52it/s]
 99%|█████████▉| 368513/371472 [7:09:35<13:58,  3.53it/s]
 99%|█████████▉| 368514/371472 [7:09:36<13:46,  3.58it/s]
 99%|█████████▉| 368515/371472 [7:09:36<14:15,  3.45it/s]
 99%|█████████���| 368516/371472 [7:09:36<14:07,  3.49it/s]
 99%|█████████▉| 368517/371472 [7:09:36<13:41,  3.60it/s]
 99%|█████████▉| 368518/371472 [7:09:37<13:48,  3.56it/s]
 99%|█████████▉| 368519/371472 [7:09:37<14:02,  3.50it/s]
 99%|█████████▉| 368520/371472 [7:09:37<14:15,  3.45it/s]
                                                         
{'loss': 2.6308, 'learning_rate': 1.0715593958068476e-07, 'epoch': 15.87}
+
 99%|█████████▉| 368520/371472 [7:09:37<14:15,  3.45it/s]
 99%|█████████▉| 368521/371472 [7:09:38<14:23,  3.42it/s]
 99%|█████████▉| 368522/371472 [7:09:38<14:27,  3.40it/s]
 99%|█████████▉| 368523/371472 [7:09:38<15:22,  3.20it/s]
 99%|█████████▉| 368524/371472 [7:09:39<14:48,  3.32it/s]
 99%|█████████▉| 368525/371472 [7:09:39<14:02,  3.50it/s]
 99%|█████████▉| 368526/371472 [7:09:39<13:39,  3.59it/s]
 99%|█████████▉| 368527/371472 [7:09:39<13:23,  3.67it/s]
 99%|█████████▉| 368528/371472 [7:09:40<14:09,  3.46it/s]
 99%|█████████▉| 368529/371472 [7:09:40<14:26,  3.40it/s]
 99%|█████████▉| 368530/371472 [7:09:40<13:42,  3.58it/s]
 99%|█████████▉| 368531/371472 [7:09:41<14:20,  3.42it/s]
 99%|█████████▉| 368532/371472 [7:09:41<15:17,  3.20it/s]
 99%|█████████▉| 368533/371472 [7:09:41<15:30,  3.16it/s]
 99%|█████████▉| 368534/371472 [7:09:41<14:44,  3.32it/s]
 99%|█████████▉| 368535/371472 [7:09:42<14:07,  3.46it/s]
 99%|█████████▉| 368536/371472 [7:09:42<13:44,  3.56it/s]
 99%|█████████▉| 368537/371472 [7:09:42<15:08,  3.23it/s]
 99%|█████████▉| 368538/371472 [7:09:43<14:43,  3.32it/s]
 99%|█████████▉| 368539/371472 [7:09:43<13:50,  3.53it/s]
 99%|█████████▉| 368540/371472 [7:09:43<13:40,  3.57it/s]
                                                         
{'loss': 2.6684, 'learning_rate': 1.0710745760520593e-07, 'epoch': 15.87}
+
 99%|█████████▉| 368540/371472 [7:09:43<13:40,  3.57it/s]
 99%|█████████▉| 368541/371472 [7:09:43<13:23,  3.65it/s]
 99%|█████████▉| 368542/371472 [7:09:44<13:44,  3.55it/s]
 99%|█████████▉| 368543/371472 [7:09:44<14:25,  3.38it/s]
 99%|█████████▉| 368544/371472 [7:09:44<14:04,  3.47it/s]
 99%|█████████▉| 368545/371472 [7:09:45<14:46,  3.30it/s]
 99%|█████████▉| 368546/371472 [7:09:45<14:51,  3.28it/s]
 99%|█████████▉| 368547/371472 [7:09:45<14:47,  3.30it/s]
 99%|█████████▉| 368548/371472 [7:09:46<14:40,  3.32it/s]
 99%|█████████▉| 368549/371472 [7:09:46<13:53,  3.51it/s]
 99%|█████████▉| 368550/371472 [7:09:46<13:47,  3.53it/s]
 99%|█████████▉| 368551/371472 [7:09:46<13:58,  3.48it/s]
 99%|█████████▉| 368552/371472 [7:09:47<13:46,  3.53it/s]
 99%|█████████▉| 368553/371472 [7:09:47<14:03,  3.46it/s]
 99%|█████████▉| 368554/371472 [7:09:47<13:50,  3.51it/s]
 99%|█████████▉| 368555/371472 [7:09:48<13:57,  3.48it/s]
 99%|█████████▉| 368556/371472 [7:09:48<14:15,  3.41it/s]
 99%|█████████▉| 368557/371472 [7:09:48<14:50,  3.27it/s]
 99%|█████████▉| 368558/371472 [7:09:48<13:50,  3.51it/s]
 99%|█████████▉| 368559/371472 [7:09:49<13:28,  3.60it/s]
 99%|█████████▉| 368560/371472 [7:09:49<13:12,  3.67it/s]
                                                         
{'loss': 2.7047, 'learning_rate': 1.0705897562972698e-07, 'epoch': 15.87}
+
 99%|█████████▉| 368560/371472 [7:09:49<13:12,  3.67it/s]
 99%|█████████▉| 368561/371472 [7:09:49<13:48,  3.51it/s]
 99%|█████████▉| 368562/371472 [7:09:50<14:37,  3.32it/s]
 99%|█████████▉| 368563/371472 [7:09:50<14:09,  3.42it/s]
 99%|█████████▉| 368564/371472 [7:09:50<14:05,  3.44it/s]
 99%|█████████▉| 368565/371472 [7:09:50<13:59,  3.46it/s]
 99%|█████████▉| 368566/371472 [7:09:51<14:27,  3.35it/s]
 99%|█████████▉| 368567/371472 [7:09:51<13:53,  3.49it/s]
 99%|█████████▉| 368568/371472 [7:09:51<13:56,  3.47it/s]
 99%|█████████▉| 368569/371472 [7:09:52<13:44,  3.52it/s]
 99%|█████████▉| 368570/371472 [7:09:52<13:29,  3.58it/s]
 99%|█████████▉| 368571/371472 [7:09:52<13:51,  3.49it/s]
 99%|█████████▉| 368572/371472 [7:09:52<13:26,  3.60it/s]
 99%|█████████▉| 368573/371472 [7:09:53<13:08,  3.68it/s]
 99%|█████████▉| 368574/371472 [7:09:53<14:03,  3.44it/s]
 99%|█████████▉| 368575/371472 [7:09:53<14:14,  3.39it/s]
 99%|█████████▉| 368576/371472 [7:09:54<13:34,  3.55it/s]
 99%|█████████▉| 368577/371472 [7:09:54<14:08,  3.41it/s]
 99%|█████████▉| 368578/371472 [7:09:54<13:40,  3.53it/s]
 99%|█████████▉| 368579/371472 [7:09:54<13:45,  3.51it/s]
 99%|█████████▉| 368580/371472 [7:09:55<13:31,  3.56it/s]
                                                         
{'loss': 2.5319, 'learning_rate': 1.0701049365424813e-07, 'epoch': 15.88}
+
 99%|█████████▉| 368580/371472 [7:09:55<13:31,  3.56it/s]
 99%|█████████▉| 368581/371472 [7:09:55<13:22,  3.60it/s]
 99%|█████████▉| 368582/371472 [7:09:55<14:16,  3.37it/s]
 99%|█████████▉| 368583/371472 [7:09:56<13:44,  3.51it/s]
 99%|█████████▉| 368584/371472 [7:09:56<14:59,  3.21it/s]
 99%|█████████▉| 368585/371472 [7:09:56<14:38,  3.29it/s]
 99%|█████████▉| 368586/371472 [7:09:57<15:23,  3.13it/s]
 99%|█████████▉| 368587/371472 [7:09:57<14:59,  3.21it/s]
 99%|█████████▉| 368588/371472 [7:09:57<14:13,  3.38it/s]
 99%|█████████▉| 368589/371472 [7:09:57<13:57,  3.44it/s]
 99%|█████████▉| 368590/371472 [7:09:58<14:44,  3.26it/s]
 99%|█████████▉| 368591/371472 [7:09:58<13:53,  3.46it/s]
 99%|█████████▉| 368592/371472 [7:09:58<13:49,  3.47it/s]
 99%|█████████▉| 368593/371472 [7:09:59<13:37,  3.52it/s]
 99%|█████████▉| 368594/371472 [7:09:59<13:57,  3.43it/s]
 99%|█████████▉| 368595/371472 [7:09:59<14:48,  3.24it/s]
 99%|█████████▉| 368596/371472 [7:09:59<13:57,  3.43it/s]
 99%|█████████▉| 368597/371472 [7:10:00<13:36,  3.52it/s]
 99%|█████████▉| 368598/371472 [7:10:00<14:21,  3.33it/s]
 99%|█████████▉| 368599/371472 [7:10:00<14:01,  3.41it/s]
 99%|█████████▉| 368600/371472 [7:10:01<13:10,  3.63it/s]
                                                         
{'loss': 2.4935, 'learning_rate': 1.0696201167876919e-07, 'epoch': 15.88}
+
 99%|█████████▉| 368600/371472 [7:10:01<13:10,  3.63it/s]
 99%|█████████▉| 368601/371472 [7:10:01<13:13,  3.62it/s]
 99%|█████████▉| 368602/371472 [7:10:01<14:01,  3.41it/s]
 99%|█████████▉| 368603/371472 [7:10:01<14:08,  3.38it/s]
 99%|█████████▉| 368604/371472 [7:10:02<14:22,  3.32it/s]
 99%|█████████▉| 368605/371472 [7:10:02<13:48,  3.46it/s]
 99%|█████████▉| 368606/371472 [7:10:02<13:33,  3.52it/s]
 99%|█████████▉| 368607/371472 [7:10:03<13:32,  3.52it/s]
 99%|█████████▉| 368608/371472 [7:10:03<13:27,  3.55it/s]
 99%|█████████▉| 368609/371472 [7:10:03<13:10,  3.62it/s]
 99%|█████████▉| 368610/371472 [7:10:03<12:51,  3.71it/s]
 99%|█████████▉| 368611/371472 [7:10:04<13:07,  3.63it/s]
 99%|█████████▉| 368612/371472 [7:10:04<13:25,  3.55it/s]
 99%|█████████▉| 368613/371472 [7:10:04<13:02,  3.65it/s]
 99%|█████████▉| 368614/371472 [7:10:05<12:47,  3.72it/s]
 99%|█████████▉| 368615/371472 [7:10:05<12:28,  3.82it/s]
 99%|█████████▉| 368616/371472 [7:10:05<13:30,  3.53it/s]
 99%|█████████▉| 368617/371472 [7:10:05<12:57,  3.67it/s]
 99%|█████████▉| 368618/371472 [7:10:06<13:50,  3.44it/s]
 99%|█████████▉| 368619/371472 [7:10:06<15:37,  3.04it/s]
 99%|█████████▉| 368620/371472 [7:10:06<14:59,  3.17it/s]
                                                         
{'loss': 2.5373, 'learning_rate': 1.0691352970329036e-07, 'epoch': 15.88}
+
 99%|█████████▉| 368620/371472 [7:10:06<14:59,  3.17it/s]
 99%|█████████▉| 368621/371472 [7:10:07<14:08,  3.36it/s]
 99%|█████████▉| 368622/371472 [7:10:07<13:42,  3.46it/s]
 99%|█████████▉| 368623/371472 [7:10:07<13:32,  3.51it/s]
 99%|█████████▉| 368624/371472 [7:10:07<13:25,  3.54it/s]
 99%|█████████▉| 368625/371472 [7:10:08<12:58,  3.66it/s]
 99%|█████████▉| 368626/371472 [7:10:08<12:50,  3.70it/s]
 99%|█████████▉| 368627/371472 [7:10:08<13:03,  3.63it/s]
 99%|█████████▉| 368628/371472 [7:10:09<14:04,  3.37it/s]
 99%|█████████▉| 368629/371472 [7:10:09<13:43,  3.45it/s]
 99%|█████████▉| 368630/371472 [7:10:09<13:29,  3.51it/s]
 99%|████████��▉| 368631/371472 [7:10:09<13:04,  3.62it/s]
 99%|█████████▉| 368632/371472 [7:10:10<13:11,  3.59it/s]
 99%|█████████▉| 368633/371472 [7:10:10<13:16,  3.56it/s]
 99%|█████████▉| 368634/371472 [7:10:10<13:07,  3.60it/s]
 99%|█████████▉| 368635/371472 [7:10:10<12:40,  3.73it/s]
 99%|█████████▉| 368636/371472 [7:10:11<12:19,  3.83it/s]
 99%|█████████▉| 368637/371472 [7:10:11<12:47,  3.70it/s]
 99%|█████████▉| 368638/371472 [7:10:11<13:03,  3.62it/s]
 99%|█████████▉| 368639/371472 [7:10:12<12:58,  3.64it/s]
 99%|█████████▉| 368640/371472 [7:10:12<12:30,  3.77it/s]
                                                         
{'loss': 2.4797, 'learning_rate': 1.068650477278114e-07, 'epoch': 15.88}
+
 99%|█████████▉| 368640/371472 [7:10:12<12:30,  3.77it/s]
 99%|█████████▉| 368641/371472 [7:10:12<12:21,  3.82it/s]
 99%|█████████▉| 368642/371472 [7:10:12<12:50,  3.67it/s]
 99%|█████████▉| 368643/371472 [7:10:13<12:53,  3.66it/s]
 99%|█████████▉| 368644/371472 [7:10:13<12:31,  3.76it/s]
 99%|█████████▉| 368645/371472 [7:10:13<12:28,  3.77it/s]
 99%|█████████▉| 368646/371472 [7:10:13<12:36,  3.74it/s]
 99%|█████████▉| 368647/371472 [7:10:14<12:59,  3.62it/s]
 99%|█████████▉| 368648/371472 [7:10:14<12:57,  3.63it/s]
 99%|█████████▉| 368649/371472 [7:10:14<12:36,  3.73it/s]
 99%|█████████▉| 368650/371472 [7:10:15<12:15,  3.84it/s]
 99%|█████████▉| 368651/371472 [7:10:15<12:28,  3.77it/s]
 99%|█████████▉| 368652/371472 [7:10:15<12:27,  3.77it/s]
 99%|█████████▉| 368653/371472 [7:10:15<12:24,  3.79it/s]
 99%|█████████▉| 368654/371472 [7:10:16<12:23,  3.79it/s]
 99%|█████████▉| 368655/371472 [7:10:16<12:18,  3.82it/s]
 99%|█████████▉| 368656/371472 [7:10:16<12:15,  3.83it/s]
 99%|█████████▉| 368657/371472 [7:10:16<12:05,  3.88it/s]
 99%|█████████▉| 368658/371472 [7:10:17<13:00,  3.60it/s]
 99%|█████████▉| 368659/371472 [7:10:17<12:44,  3.68it/s]
 99%|█████████▉| 368660/371472 [7:10:17<12:56,  3.62it/s]
                                                         
{'loss': 2.7023, 'learning_rate': 1.0681656575233257e-07, 'epoch': 15.88}
+
 99%|█████████▉| 368660/371472 [7:10:17<12:56,  3.62it/s]
 99%|█████████▉| 368661/371472 [7:10:17<12:43,  3.68it/s]
 99%|█████████▉| 368662/371472 [7:10:18<13:18,  3.52it/s]
 99%|█████████▉| 368663/371472 [7:10:18<13:05,  3.58it/s]
 99%|█████████▉| 368664/371472 [7:10:18<13:59,  3.34it/s]
 99%|█████████▉| 368665/371472 [7:10:19<13:27,  3.48it/s]
 99%|█████████▉| 368666/371472 [7:10:19<14:17,  3.27it/s]
 99%|█████████▉| 368667/371472 [7:10:19<13:42,  3.41it/s]
 99%|█████████▉| 368668/371472 [7:10:20<12:56,  3.61it/s]
 99%|█████████▉| 368669/371472 [7:10:20<14:00,  3.34it/s]
 99%|█████████▉| 368670/371472 [7:10:20<13:45,  3.40it/s]
 99%|█████████▉| 368671/371472 [7:10:20<13:38,  3.42it/s]
 99%|█████████▉| 368672/371472 [7:10:21<13:13,  3.53it/s]
 99%|█████████▉| 368673/371472 [7:10:21<13:49,  3.38it/s]
 99%|█████████▉| 368674/371472 [7:10:21<13:32,  3.44it/s]
 99%|█████████▉| 368675/371472 [7:10:22<13:31,  3.45it/s]
 99%|█████████▉| 368676/371472 [7:10:22<13:00,  3.58it/s]
 99%|█████████▉| 368677/371472 [7:10:22<12:53,  3.61it/s]
 99%|█████████▉| 368678/371472 [7:10:22<13:02,  3.57it/s]
 99%|█████████▉| 368679/371472 [7:10:23<13:29,  3.45it/s]
 99%|█████████▉| 368680/371472 [7:10:23<13:06,  3.55it/s]
                                                         
{'loss': 2.559, 'learning_rate': 1.0676808377685362e-07, 'epoch': 15.88}
+
 99%|█████████▉| 368680/371472 [7:10:23<13:06,  3.55it/s]
 99%|█████████▉| 368681/371472 [7:10:23<14:14,  3.26it/s]
 99%|█████████▉| 368682/371472 [7:10:24<14:25,  3.23it/s]
 99%|█████████▉| 368683/371472 [7:10:24<14:32,  3.20it/s]
 99%|█████████▉| 368684/371472 [7:10:24<14:58,  3.10it/s]
 99%|█████████▉| 368685/371472 [7:10:25<14:17,  3.25it/s]
 99%|█████████▉| 368686/371472 [7:10:25<14:33,  3.19it/s]
 99%|█████████▉| 368687/371472 [7:10:25<13:38,  3.40it/s]
 99%|█████████▉| 368688/371472 [7:10:25<13:56,  3.33it/s]
 99%|█████████▉| 368689/371472 [7:10:26<14:39,  3.17it/s]
 99%|█████████▉| 368690/371472 [7:10:26<13:42,  3.38it/s]
 99%|█████████▉| 368691/371472 [7:10:26<14:36,  3.17it/s]
 99%|█████████▉| 368692/371472 [7:10:27<14:14,  3.25it/s]
 99%|█████████▉| 368693/371472 [7:10:27<13:32,  3.42it/s]
 99%|█████████▉| 368694/371472 [7:10:27<14:20,  3.23it/s]
 99%|█████████▉| 368695/371472 [7:10:28<13:37,  3.40it/s]
 99%|█████████▉| 368696/371472 [7:10:28<13:06,  3.53it/s]
 99%|█████████▉| 368697/371472 [7:10:28<12:56,  3.57it/s]
 99%|█████████▉| 368698/371472 [7:10:28<13:54,  3.33it/s]
 99%|█████████▉| 368699/371472 [7:10:29<14:09,  3.26it/s]
 99%|█████████▉| 368700/371472 [7:10:29<13:17,  3.48it/s]
                                                         
{'loss': 2.5607, 'learning_rate': 1.0671960180137468e-07, 'epoch': 15.88}
+
 99%|█████████▉| 368700/371472 [7:10:29<13:17,  3.48it/s]
 99%|█████████▉| 368701/371472 [7:10:29<13:00,  3.55it/s]
 99%|█████████▉| 368702/371472 [7:10:30<12:42,  3.63it/s]
 99%|█████████▉| 368703/371472 [7:10:30<12:20,  3.74it/s]
 99%|█████████▉| 368704/371472 [7:10:30<12:10,  3.79it/s]
 99%|█████████▉| 368705/371472 [7:10:30<12:20,  3.74it/s]
 99%|█████████▉| 368706/371472 [7:10:31<13:01,  3.54it/s]
 99%|█████████▉| 368707/371472 [7:10:31<13:07,  3.51it/s]
 99%|█████████▉| 368708/371472 [7:10:31<12:29,  3.69it/s]
 99%|█████████▉| 368709/371472 [7:10:32<13:37,  3.38it/s]
 99%|█████████▉| 368710/371472 [7:10:32<13:11,  3.49it/s]
 99%|█████████▉| 368711/371472 [7:10:32<13:33,  3.40it/s]
 99%|█████████▉| 368712/371472 [7:10:32<13:20,  3.45it/s]
 99%|█████████▉| 368713/371472 [7:10:33<13:18,  3.46it/s]
 99%|█████████▉| 368714/371472 [7:10:33<13:33,  3.39it/s]
 99%|█████████▉| 368715/371472 [7:10:33<13:43,  3.35it/s]
 99%|█████████▉| 368716/371472 [7:10:34<13:07,  3.50it/s]
 99%|█████████▉| 368717/371472 [7:10:34<13:28,  3.41it/s]
 99%|█████████▉| 368718/371472 [7:10:34<13:03,  3.51it/s]
 99%|█████████▉| 368719/371472 [7:10:35<14:13,  3.23it/s]
 99%|█████████▉| 368720/371472 [7:10:35<13:43,  3.34it/s]
                                                         
{'loss': 2.5345, 'learning_rate': 1.0667111982589583e-07, 'epoch': 15.88}
+
 99%|█████████▉| 368720/371472 [7:10:35<13:43,  3.34it/s]
 99%|█████████▉| 368721/371472 [7:10:35<14:03,  3.26it/s]
 99%|█████████▉| 368722/371472 [7:10:35<13:54,  3.29it/s]
 99%|█████████▉| 368723/371472 [7:10:36<14:07,  3.24it/s]
 99%|█████████▉| 368724/371472 [7:10:36<14:13,  3.22it/s]
 99%|█████████▉| 368725/371472 [7:10:36<14:03,  3.26it/s]
 99%|█████████▉| 368726/371472 [7:10:37<13:14,  3.46it/s]
 99%|█████████▉| 368727/371472 [7:10:37<13:39,  3.35it/s]
 99%|█████████▉| 368728/371472 [7:10:37<13:34,  3.37it/s]
 99%|█████████▉| 368729/371472 [7:10:37<13:06,  3.49it/s]
 99%|█████████▉| 368730/371472 [7:10:38<14:34,  3.13it/s]
 99%|█████████▉| 368731/371472 [7:10:38<14:29,  3.15it/s]
 99%|█████████▉| 368732/371472 [7:10:39<15:06,  3.02it/s]
 99%|█████████▉| 368733/371472 [7:10:39<14:57,  3.05it/s]
 99%|█████████▉| 368734/371472 [7:10:39<14:01,  3.26it/s]
 99%|█████████▉| 368735/371472 [7:10:39<13:06,  3.48it/s]
 99%|█████████▉| 368736/371472 [7:10:40<13:31,  3.37it/s]
 99%|█████████▉| 368737/371472 [7:10:40<13:28,  3.38it/s]
 99%|█████████▉| 368738/371472 [7:10:40<12:57,  3.52it/s]
 99%|█████████▉| 368739/371472 [7:10:41<12:57,  3.51it/s]
 99%|█████████▉| 368740/371472 [7:10:41<12:36,  3.61it/s]
                                                         
{'loss': 2.5494, 'learning_rate': 1.0662263785041689e-07, 'epoch': 15.88}
+
 99%|█████████▉| 368740/371472 [7:10:41<12:36,  3.61it/s]
 99%|█████████▉| 368741/371472 [7:10:41<12:17,  3.70it/s]
 99%|█████████▉| 368742/371472 [7:10:41<12:59,  3.50it/s]
 99%|█████████▉| 368743/371472 [7:10:42<12:38,  3.60it/s]
 99%|█████████▉| 368744/371472 [7:10:42<12:35,  3.61it/s]
 99%|█████████▉| 368745/371472 [7:10:42<12:59,  3.50it/s]
 99%|████████��▉| 368746/371472 [7:10:42<12:49,  3.54it/s]
 99%|█████████▉| 368747/371472 [7:10:43<12:59,  3.50it/s]
 99%|█████████▉| 368748/371472 [7:10:43<12:28,  3.64it/s]
 99%|█████████▉| 368749/371472 [7:10:43<12:17,  3.69it/s]
 99%|█████████▉| 368750/371472 [7:10:44<11:58,  3.79it/s]
 99%|█████████▉| 368751/371472 [7:10:44<11:42,  3.87it/s]
 99%|█████████▉| 368752/371472 [7:10:44<12:03,  3.76it/s]
 99%|█████████▉| 368753/371472 [7:10:44<11:40,  3.88it/s]
 99%|█████████▉| 368754/371472 [7:10:45<11:50,  3.82it/s]
 99%|█████████▉| 368755/371472 [7:10:45<11:52,  3.82it/s]
 99%|█████████▉| 368756/371472 [7:10:45<12:01,  3.77it/s]
 99%|█████████▉| 368757/371472 [7:10:45<11:55,  3.79it/s]
 99%|█████████▉| 368758/371472 [7:10:46<12:46,  3.54it/s]
 99%|█████████▉| 368759/371472 [7:10:46<12:40,  3.57it/s]
 99%|█████████▉| 368760/371472 [7:10:46<12:40,  3.56it/s]
                                                         
{'loss': 2.5712, 'learning_rate': 1.0657415587493806e-07, 'epoch': 15.88}
+
 99%|█████████▉| 368760/371472 [7:10:46<12:40,  3.56it/s]
 99%|█████████▉| 368761/371472 [7:10:47<12:54,  3.50it/s]
 99%|█████████▉| 368762/371472 [7:10:47<12:21,  3.65it/s]
 99%|█████████▉| 368763/371472 [7:10:47<13:18,  3.39it/s]
 99%|█████████▉| 368764/371472 [7:10:47<13:18,  3.39it/s]
 99%|█████████▉| 368765/371472 [7:10:48<13:10,  3.42it/s]
 99%|█████████▉| 368766/371472 [7:10:48<12:36,  3.58it/s]
 99%|█████████▉| 368767/371472 [7:10:48<12:30,  3.60it/s]
 99%|█████████▉| 368768/371472 [7:10:49<12:54,  3.49it/s]
 99%|█████████▉| 368769/371472 [7:10:49<12:26,  3.62it/s]
 99%|█████████▉| 368770/371472 [7:10:49<12:20,  3.65it/s]
 99%|█████████▉| 368771/371472 [7:10:49<12:11,  3.69it/s]
 99%|█████████▉| 368772/371472 [7:10:50<12:02,  3.73it/s]
 99%|█████████▉| 368773/371472 [7:10:50<12:07,  3.71it/s]
 99%|█████████▉| 368774/371472 [7:10:50<11:59,  3.75it/s]
 99%|█████████▉| 368775/371472 [7:10:50<12:34,  3.57it/s]
 99%|█████████▉| 368776/371472 [7:10:51<12:39,  3.55it/s]
 99%|█████████▉| 368777/371472 [7:10:51<14:08,  3.18it/s]
 99%|█████████▉| 368778/371472 [7:10:51<14:50,  3.02it/s]
 99%|█████████▉| 368779/371472 [7:10:52<14:00,  3.20it/s]
 99%|█████████▉| 368780/371472 [7:10:52<14:16,  3.14it/s]
                                                         
{'loss': 2.5638, 'learning_rate': 1.0652567389945911e-07, 'epoch': 15.88}
+
 99%|█████████▉| 368780/371472 [7:10:52<14:16,  3.14it/s]
 99%|█████████▉| 368781/371472 [7:10:52<13:34,  3.30it/s]
 99%|█████████▉| 368782/371472 [7:10:53<13:10,  3.40it/s]
 99%|█████████▉| 368783/371472 [7:10:53<12:26,  3.60it/s]
 99%|█████████▉| 368784/371472 [7:10:53<12:05,  3.71it/s]
 99%|█████████▉| 368785/371472 [7:10:53<13:21,  3.35it/s]
 99%|█████████▉| 368786/371472 [7:10:54<13:05,  3.42it/s]
 99%|█████████▉| 368787/371472 [7:10:54<12:50,  3.48it/s]
 99%|█████████▉| 368788/371472 [7:10:54<13:50,  3.23it/s]
 99%|█████████▉| 368789/371472 [7:10:55<13:35,  3.29it/s]
 99%|█████████▉| 368790/371472 [7:10:55<12:58,  3.45it/s]
 99%|█████████▉| 368791/371472 [7:10:55<12:47,  3.49it/s]
 99%|█████████▉| 368792/371472 [7:10:55<12:34,  3.55it/s]
 99%|█████████▉| 368793/371472 [7:10:56<13:01,  3.43it/s]
 99%|█████████▉| 368794/371472 [7:10:56<13:11,  3.38it/s]
 99%|█████████▉| 368795/371472 [7:10:56<12:53,  3.46it/s]
 99%|█████████▉| 368796/371472 [7:10:57<14:09,  3.15it/s]
 99%|█████████▉| 368797/371472 [7:10:57<13:47,  3.23it/s]
 99%|█████████▉| 368798/371472 [7:10:57<13:42,  3.25it/s]
 99%|█████████▉| 368799/371472 [7:10:58<14:15,  3.13it/s]
 99%|█████████▉| 368800/371472 [7:10:58<14:01,  3.17it/s]
                                                         
{'loss': 2.5413, 'learning_rate': 1.0647719192398026e-07, 'epoch': 15.88}
+
 99%|█████████▉| 368800/371472 [7:10:58<14:01,  3.17it/s]
 99%|█████████▉| 368801/371472 [7:10:58<14:46,  3.01it/s]
 99%|█████████▉| 368802/371472 [7:10:59<14:11,  3.14it/s]
 99%|█████████▉| 368803/371472 [7:10:59<13:24,  3.32it/s]
 99%|█████████▉| 368804/371472 [7:10:59<13:35,  3.27it/s]
 99%|█████████▉| 368805/371472 [7:11:00<13:16,  3.35it/s]
 99%|█████████▉| 368806/371472 [7:11:00<13:07,  3.39it/s]
 99%|█████████▉| 368807/371472 [7:11:00<12:26,  3.57it/s]
 99%|█████████▉| 368808/371472 [7:11:00<12:05,  3.67it/s]
 99%|█████████▉| 368809/371472 [7:11:01<11:49,  3.75it/s]
 99%|█████████▉| 368810/371472 [7:11:01<12:04,  3.68it/s]
 99%|█████████▉| 368811/371472 [7:11:01<12:11,  3.64it/s]
 99%|█████████▉| 368812/371472 [7:11:01<12:00,  3.69it/s]
 99%|█████████▉| 368813/371472 [7:11:02<12:07,  3.66it/s]
 99%|█████████▉| 368814/371472 [7:11:02<11:49,  3.74it/s]
 99%|█████████▉| 368815/371472 [7:11:02<11:31,  3.84it/s]
 99%|█████████▉| 368816/371472 [7:11:02<12:09,  3.64it/s]
 99%|█████████▉| 368817/371472 [7:11:03<11:55,  3.71it/s]
 99%|█████████▉| 368818/371472 [7:11:03<12:03,  3.67it/s]
 99%|█████████▉| 368819/371472 [7:11:03<12:00,  3.68it/s]
 99%|█████████▉| 368820/371472 [7:11:04<11:43,  3.77it/s]
                                                         
{'loss': 2.5894, 'learning_rate': 1.0642870994850132e-07, 'epoch': 15.89}
+
 99%|█████████▉| 368820/371472 [7:11:04<11:43,  3.77it/s]
 99%|█████████▉| 368821/371472 [7:11:04<11:29,  3.84it/s]
 99%|█████████▉| 368822/371472 [7:11:04<11:31,  3.83it/s]
 99%|█████████▉| 368823/371472 [7:11:04<11:43,  3.76it/s]
 99%|█████████▉| 368824/371472 [7:11:05<11:34,  3.81it/s]
 99%|█████████▉| 368825/371472 [7:11:05<11:45,  3.75it/s]
 99%|█████████▉| 368826/371472 [7:11:05<11:41,  3.77it/s]
 99%|█████████▉| 368827/371472 [7:11:05<12:30,  3.52it/s]
 99%|█████████▉| 368828/371472 [7:11:06<12:12,  3.61it/s]
 99%|█████████▉| 368829/371472 [7:11:06<11:47,  3.73it/s]
 99%|█████████▉| 368830/371472 [7:11:06<11:49,  3.72it/s]
 99%|█████████▉| 368831/371472 [7:11:06<11:48,  3.73it/s]
 99%|█████████▉| 368832/371472 [7:11:07<12:13,  3.60it/s]
 99%|█████████▉| 368833/371472 [7:11:07<12:16,  3.59it/s]
 99%|█████████▉| 368834/371472 [7:11:07<12:59,  3.38it/s]
 99%|█████████▉| 368835/371472 [7:11:08<12:42,  3.46it/s]
 99%|█████████▉| 368836/371472 [7:11:08<13:01,  3.37it/s]
 99%|█████████▉| 368837/371472 [7:11:08<12:40,  3.47it/s]
 99%|█████████▉| 368838/371472 [7:11:09<12:14,  3.59it/s]
 99%|█████████▉| 368839/371472 [7:11:09<12:03,  3.64it/s]
 99%|█████████▉| 368840/371472 [7:11:09<12:35,  3.48it/s]
                                                         
{'loss': 2.7263, 'learning_rate': 1.0638022797302249e-07, 'epoch': 15.89}
+
 99%|█████████▉| 368840/371472 [7:11:09<12:35,  3.48it/s]
 99%|█████████▉| 368841/371472 [7:11:09<12:51,  3.41it/s]
 99%|█████████▉| 368842/371472 [7:11:10<12:36,  3.48it/s]
 99%|█████████▉| 368843/371472 [7:11:10<13:11,  3.32it/s]
 99%|█████████▉| 368844/371472 [7:11:10<12:30,  3.50it/s]
 99%|█████████▉| 368845/371472 [7:11:11<12:29,  3.51it/s]
 99%|█████████▉| 368846/371472 [7:11:11<13:02,  3.36it/s]
 99%|█████████▉| 368847/371472 [7:11:11<12:55,  3.38it/s]
 99%|█████████▉| 368848/371472 [7:11:11<12:39,  3.45it/s]
 99%|█████████▉| 368849/371472 [7:11:12<12:18,  3.55it/s]
 99%|█████████▉| 368850/371472 [7:11:12<12:15,  3.57it/s]
 99%|█████████▉| 368851/371472 [7:11:12<12:43,  3.43it/s]
 99%|█████████▉| 368852/371472 [7:11:13<12:07,  3.60it/s]
 99%|█████████▉| 368853/371472 [7:11:13<11:57,  3.65it/s]
 99%|█████████▉| 368854/371472 [7:11:13<11:58,  3.64it/s]
 99%|█████████▉| 368855/371472 [7:11:13<11:55,  3.66it/s]
 99%|█████████▉| 368856/371472 [7:11:14<11:36,  3.76it/s]
 99%|█████████▉| 368857/371472 [7:11:14<11:40,  3.73it/s]
 99%|█████████▉| 368858/371472 [7:11:14<11:41,  3.73it/s]
 99%|█████████▉| 368859/371472 [7:11:14<12:09,  3.58it/s]
 99%|█████████▉| 368860/371472 [7:11:15<11:51,  3.67it/s]
                                                         
{'loss': 2.5099, 'learning_rate': 1.0633174599754355e-07, 'epoch': 15.89}
+
 99%|█████████▉| 368860/371472 [7:11:15<11:51,  3.67it/s]
 99%|█████████▉| 368861/371472 [7:11:15<12:03,  3.61it/s]
 99%|█████████▉| 368862/371472 [7:11:15<12:00,  3.62it/s]
 99%|█████████▉| 368863/371472 [7:11:16<11:55,  3.65it/s]
 99%|█████████▉| 368864/371472 [7:11:16<12:09,  3.57it/s]
 99%|█████████▉| 368865/371472 [7:11:16<12:28,  3.48it/s]
 99%|█████████▉| 368866/371472 [7:11:16<12:47,  3.40it/s]
 99%|█████████▉| 368867/371472 [7:11:17<12:42,  3.42it/s]
 99%|█████████▉| 368868/371472 [7:11:17<12:09,  3.57it/s]
 99%|█████████▉| 368869/371472 [7:11:17<11:48,  3.67it/s]
 99%|█████████▉| 368870/371472 [7:11:18<12:06,  3.58it/s]
 99%|█████████▉| 368871/371472 [7:11:18<11:48,  3.67it/s]
 99%|█████████▉| 368872/371472 [7:11:18<11:39,  3.72it/s]
 99%|█████████▉| 368873/371472 [7:11:18<12:03,  3.59it/s]
 99%|█████████▉| 368874/371472 [7:11:19<13:23,  3.23it/s]
 99%|█████████▉| 368875/371472 [7:11:19<12:30,  3.46it/s]
 99%|█████████▉| 368876/371472 [7:11:19<12:32,  3.45it/s]
 99%|█████████▉| 368877/371472 [7:11:20<12:56,  3.34it/s]
 99%|█████████▉| 368878/371472 [7:11:20<13:29,  3.20it/s]
 99%|█████████▉| 368879/371472 [7:11:20<13:11,  3.27it/s]
 99%|█████████▉| 368880/371472 [7:11:20<12:51,  3.36it/s]
                                                         
{'loss': 2.5344, 'learning_rate': 1.062832640220647e-07, 'epoch': 15.89}
+
 99%|█████████▉| 368880/371472 [7:11:20<12:51,  3.36it/s]
 99%|█████████▉| 368881/371472 [7:11:21<12:50,  3.36it/s]
 99%|█████████▉| 368882/371472 [7:11:21<13:00,  3.32it/s]
 99%|█████████▉| 368883/371472 [7:11:21<13:00,  3.32it/s]
 99%|█████████▉| 368884/371472 [7:11:22<12:24,  3.47it/s]
 99%|█████████▉| 368885/371472 [7:11:22<12:51,  3.35it/s]
 99%|█████████▉| 368886/371472 [7:11:22<12:16,  3.51it/s]
 99%|█████████▉| 368887/371472 [7:11:23<12:23,  3.48it/s]
 99%|█████████▉| 368888/371472 [7:11:23<12:06,  3.56it/s]
 99%|█████████▉| 368889/371472 [7:11:23<11:52,  3.63it/s]
 99%|█████████▉| 368890/371472 [7:11:23<11:49,  3.64it/s]
 99%|█████████▉| 368891/371472 [7:11:24<11:42,  3.68it/s]
 99%|█████████▉| 368892/371472 [7:11:24<11:53,  3.62it/s]
 99%|█████████▉| 368893/371472 [7:11:24<11:49,  3.64it/s]
 99%|█████████▉| 368894/371472 [7:11:24<12:16,  3.50it/s]
 99%|█████████▉| 368895/371472 [7:11:25<12:02,  3.57it/s]
 99%|█████████▉| 368896/371472 [7:11:25<11:33,  3.71it/s]
 99%|█████████▉| 368897/371472 [7:11:25<12:25,  3.45it/s]
 99%|█████████▉| 368898/371472 [7:11:26<13:15,  3.24it/s]
 99%|█████████▉| 368899/371472 [7:11:26<12:29,  3.43it/s]
 99%|█████████▉| 368900/371472 [7:11:26<12:49,  3.34it/s]
                                                         
{'loss': 2.5552, 'learning_rate': 1.0623478204658575e-07, 'epoch': 15.89}
+
 99%|█████████▉| 368900/371472 [7:11:26<12:49,  3.34it/s]
 99%|█████████▉| 368901/371472 [7:11:26<12:13,  3.50it/s]
 99%|█████████▉| 368902/371472 [7:11:27<11:33,  3.70it/s]
 99%|█████████▉| 368903/371472 [7:11:27<11:29,  3.72it/s]
 99%|█████████▉| 368904/371472 [7:11:27<11:39,  3.67it/s]
 99%|█████████▉| 368905/371472 [7:11:28<11:43,  3.65it/s]
 99%|█████████▉| 368906/371472 [7:11:28<11:59,  3.57it/s]
 99%|█████████▉| 368907/371472 [7:11:28<11:47,  3.63it/s]
 99%|█████████▉| 368908/371472 [7:11:28<12:05,  3.53it/s]
 99%|█████████▉| 368909/371472 [7:11:29<12:15,  3.49it/s]
 99%|█████████▉| 368910/371472 [7:11:29<12:44,  3.35it/s]
 99%|█████████▉| 368911/371472 [7:11:29<12:13,  3.49it/s]
 99%|█████████▉| 368912/371472 [7:11:30<12:20,  3.46it/s]
 99%|█████████▉| 368913/371472 [7:11:30<12:40,  3.36it/s]
 99%|█████████▉| 368914/371472 [7:11:30<12:09,  3.51it/s]
 99%|█████████▉| 368915/371472 [7:11:30<11:49,  3.60it/s]
 99%|█████████▉| 368916/371472 [7:11:31<11:49,  3.60it/s]
 99%|█████████▉| 368917/371472 [7:11:31<11:46,  3.62it/s]
 99%|█████████▉| 368918/371472 [7:11:31<11:32,  3.69it/s]
 99%|█████████▉| 368919/371472 [7:11:31<11:26,  3.72it/s]
 99%|█████████▉| 368920/371472 [7:11:32<11:46,  3.61it/s]
                                                         
{'loss': 2.5136, 'learning_rate': 1.0618630007110692e-07, 'epoch': 15.89}
+
 99%|█████████▉| 368920/371472 [7:11:32<11:46,  3.61it/s]
 99%|█████████▉| 368921/371472 [7:11:32<12:36,  3.37it/s]
 99%|█████████▉| 368922/371472 [7:11:32<12:00,  3.54it/s]
 99%|█████████▉| 368923/371472 [7:11:33<12:17,  3.45it/s]
 99%|█████████▉| 368924/371472 [7:11:33<12:45,  3.33it/s]
 99%|█████████▉| 368925/371472 [7:11:33<13:10,  3.22it/s]
 99%|█████████▉| 368926/371472 [7:11:34<12:23,  3.42it/s]
 99%|█████████▉| 368927/371472 [7:11:34<12:04,  3.51it/s]
 99%|█████████▉| 368928/371472 [7:11:34<11:49,  3.58it/s]
 99%|█████████▉| 368929/371472 [7:11:34<11:20,  3.74it/s]
 99%|█████████▉| 368930/371472 [7:11:35<11:49,  3.59it/s]
 99%|█████████▉| 368931/371472 [7:11:35<11:34,  3.66it/s]
 99%|█████████▉| 368932/371472 [7:11:35<11:52,  3.57it/s]
 99%|█████████▉| 368933/371472 [7:11:35<11:47,  3.59it/s]
 99%|█████████▉| 368934/371472 [7:11:36<11:58,  3.53it/s]
 99%|█████████▉| 368935/371472 [7:11:36<11:38,  3.63it/s]
 99%|█████████▉| 368936/371472 [7:11:36<11:29,  3.68it/s]
 99%|█████████▉| 368937/371472 [7:11:37<12:04,  3.50it/s]
 99%|█████████▉| 368938/371472 [7:11:37<11:39,  3.62it/s]
 99%|█████████▉| 368939/371472 [7:11:37<11:19,  3.73it/s]
 99%|█████████▉| 368940/371472 [7:11:37<11:09,  3.78it/s]
                                                         
{'loss': 2.4577, 'learning_rate': 1.0613781809562798e-07, 'epoch': 15.89}
+
 99%|█████████▉| 368940/371472 [7:11:37<11:09,  3.78it/s]
 99%|█████████▉| 368941/371472 [7:11:38<10:58,  3.84it/s]
 99%|█████████▉| 368942/371472 [7:11:38<11:08,  3.79it/s]
 99%|█████████▉| 368943/371472 [7:11:38<11:05,  3.80it/s]
 99%|█████████▉| 368944/371472 [7:11:38<10:50,  3.89it/s]
 99%|█████████▉| 368945/371472 [7:11:39<11:40,  3.60it/s]
 99%|█████████▉| 368946/371472 [7:11:39<12:06,  3.48it/s]
 99%|█████████▉| 368947/371472 [7:11:39<12:04,  3.49it/s]
 99%|█████████▉| 368948/371472 [7:11:40<12:14,  3.44it/s]
 99%|█████████▉| 368949/371472 [7:11:40<11:53,  3.53it/s]
 99%|█████████▉| 368950/371472 [7:11:40<11:21,  3.70it/s]
 99%|█████████▉| 368951/371472 [7:11:40<11:31,  3.65it/s]
 99%|█████████▉| 368952/371472 [7:11:41<12:59,  3.23it/s]
 99%|█████████▉| 368953/371472 [7:11:41<12:44,  3.29it/s]
 99%|█████████▉| 368954/371472 [7:11:41<12:14,  3.43it/s]
 99%|█████████▉| 368955/371472 [7:11:42<12:10,  3.45it/s]
 99%|█████████▉| 368956/371472 [7:11:42<11:53,  3.53it/s]
 99%|█████████▉| 368957/371472 [7:11:42<12:08,  3.45it/s]
 99%|█████████▉| 368958/371472 [7:11:43<11:55,  3.51it/s]
 99%|█████████▉| 368959/371472 [7:11:43<12:27,  3.36it/s]
 99%|█████████▉| 368960/371472 [7:11:43<12:11,  3.43it/s]
                                                         
{'loss': 2.3523, 'learning_rate': 1.0608933612014913e-07, 'epoch': 15.89}
+
 99%|█████████▉| 368960/371472 [7:11:43<12:11,  3.43it/s]
 99%|█████████▉| 368961/371472 [7:11:43<12:47,  3.27it/s]
 99%|█████████▉| 368962/371472 [7:11:44<12:31,  3.34it/s]
 99%|█████████▉| 368963/371472 [7:11:44<12:32,  3.33it/s]
 99%|█████████▉| 368964/371472 [7:11:44<12:21,  3.38it/s]
 99%|█████████▉| 368965/371472 [7:11:45<11:56,  3.50it/s]
 99%|█████████▉| 368966/371472 [7:11:45<12:35,  3.32it/s]
 99%|█████████▉| 368967/371472 [7:11:45<12:05,  3.45it/s]
 99%|█████████▉| 368968/371472 [7:11:45<11:35,  3.60it/s]
 99%|█████████▉| 368969/371472 [7:11:46<11:16,  3.70it/s]
 99%|█████████▉| 368970/371472 [7:11:46<11:14,  3.71it/s]
 99%|█████████▉| 368971/371472 [7:11:46<12:02,  3.46it/s]
 99%|█████████▉| 368972/371472 [7:11:47<12:02,  3.46it/s]
 99%|█████████▉| 368973/371472 [7:11:47<12:26,  3.35it/s]
 99%|█████████▉| 368974/371472 [7:11:47<11:58,  3.48it/s]
 99%|█████████▉| 368975/371472 [7:11:47<11:26,  3.64it/s]
 99%|█████████▉| 368976/371472 [7:11:48<11:34,  3.59it/s]
 99%|█████████▉| 368977/371472 [7:11:48<11:35,  3.59it/s]
 99%|█████████▉| 368978/371472 [7:11:48<11:28,  3.62it/s]
 99%|���████████▉| 368979/371472 [7:11:49<11:16,  3.69it/s]
 99%|█████████▉| 368980/371472 [7:11:49<11:11,  3.71it/s]
                                                         
{'loss': 2.7452, 'learning_rate': 1.0604085414467019e-07, 'epoch': 15.89}
+
 99%|█████████▉| 368980/371472 [7:11:49<11:11,  3.71it/s]
 99%|█████████▉| 368981/371472 [7:11:49<12:17,  3.38it/s]
 99%|█████████▉| 368982/371472 [7:11:49<12:05,  3.43it/s]
 99%|█████████▉| 368983/371472 [7:11:50<12:03,  3.44it/s]
 99%|█████████▉| 368984/371472 [7:11:50<12:53,  3.22it/s]
 99%|█████████▉| 368985/371472 [7:11:50<12:04,  3.43it/s]
 99%|█████████▉| 368986/371472 [7:11:51<11:41,  3.55it/s]
 99%|█████████▉| 368987/371472 [7:11:51<11:45,  3.52it/s]
 99%|█████████▉| 368988/371472 [7:11:51<12:04,  3.43it/s]
 99%|█████████▉| 368989/371472 [7:11:51<11:40,  3.55it/s]
 99%|█████████▉| 368990/371472 [7:11:52<11:15,  3.67it/s]
 99%|█████████▉| 368991/371472 [7:11:52<12:40,  3.26it/s]
 99%|█████████▉| 368992/371472 [7:11:52<12:32,  3.29it/s]
 99%|█████████▉| 368993/371472 [7:11:53<12:32,  3.30it/s]
 99%|█████████▉| 368994/371472 [7:11:53<12:22,  3.34it/s]
 99%|█████████▉| 368995/371472 [7:11:53<12:12,  3.38it/s]
 99%|█████████▉| 368996/371472 [7:11:54<12:39,  3.26it/s]
 99%|█████████▉| 368997/371472 [7:11:54<12:24,  3.32it/s]
 99%|█████████▉| 368998/371472 [7:11:54<12:04,  3.42it/s]
 99%|█████████▉| 368999/371472 [7:11:54<11:41,  3.53it/s]
 99%|█████████▉| 369000/371472 [7:11:55<11:47,  3.49it/s]
                                                         
{'loss': 2.559, 'learning_rate': 1.0599237216919135e-07, 'epoch': 15.89}
+
 99%|█████████▉| 369000/371472 [7:11:55<11:47,  3.49it/s]
 99%|█████████▉| 369001/371472 [7:11:55<11:48,  3.49it/s]
 99%|█████████▉| 369002/371472 [7:11:55<11:28,  3.59it/s]
 99%|█████████▉| 369003/371472 [7:11:55<11:13,  3.67it/s]
 99%|█████████▉| 369004/371472 [7:11:56<12:37,  3.26it/s]
 99%|█████████▉| 369005/371472 [7:11:56<12:30,  3.29it/s]
 99%|█████████▉| 369006/371472 [7:11:56<11:53,  3.46it/s]
 99%|█████████▉| 369007/371472 [7:11:57<12:08,  3.38it/s]
 99%|█████████▉| 369008/371472 [7:11:57<12:18,  3.34it/s]
 99%|█████████▉| 369009/371472 [7:11:57<11:44,  3.50it/s]
 99%|█████████▉| 369010/371472 [7:11:58<12:30,  3.28it/s]
 99%|█████████▉| 369011/371472 [7:11:58<12:27,  3.29it/s]
 99%|█████████▉| 369012/371472 [7:11:58<12:32,  3.27it/s]
 99%|█████████▉| 369013/371472 [7:11:59<12:01,  3.41it/s]
 99%|█████████▉| 369014/371472 [7:11:59<11:52,  3.45it/s]
 99%|█████████▉| 369015/371472 [7:11:59<11:27,  3.57it/s]
 99%|█████████▉| 369016/371472 [7:11:59<11:20,  3.61it/s]
 99%|█████████▉| 369017/371472 [7:12:00<11:09,  3.67it/s]
 99%|█████████▉| 369018/371472 [7:12:00<11:04,  3.70it/s]
 99%|█████████▉| 369019/371472 [7:12:00<11:16,  3.62it/s]
 99%|█████████▉| 369020/371472 [7:12:00<11:15,  3.63it/s]
                                                         
{'loss': 2.497, 'learning_rate': 1.0594389019371241e-07, 'epoch': 15.89}
+
 99%|█████████▉| 369020/371472 [7:12:00<11:15,  3.63it/s]
 99%|█████████▉| 369021/371472 [7:12:01<11:24,  3.58it/s]
 99%|█████████▉| 369022/371472 [7:12:01<11:10,  3.66it/s]
 99%|█████████▉| 369023/371472 [7:12:01<12:01,  3.39it/s]
 99%|█████████▉| 369024/371472 [7:12:02<12:20,  3.31it/s]
 99%|█████████▉| 369025/371472 [7:12:02<12:27,  3.28it/s]
 99%|█████████▉| 369026/371472 [7:12:02<13:04,  3.12it/s]
 99%|█████████▉| 369027/371472 [7:12:03<12:47,  3.19it/s]
 99%|█████████▉| 369028/371472 [7:12:03<12:13,  3.33it/s]
 99%|█████████▉| 369029/371472 [7:12:03<12:10,  3.34it/s]
 99%|█████████▉| 369030/371472 [7:12:03<11:47,  3.45it/s]
 99%|█████████▉| 369031/371472 [7:12:04<11:21,  3.58it/s]
 99%|█████████▉| 369032/371472 [7:12:04<10:56,  3.72it/s]
 99%|█████████▉| 369033/371472 [7:12:04<10:46,  3.77it/s]
 99%|█████████▉| 369034/371472 [7:12:04<11:18,  3.59it/s]
 99%|█████████▉| 369035/371472 [7:12:05<10:57,  3.71it/s]
 99%|█████████▉| 369036/371472 [7:12:05<10:55,  3.72it/s]
 99%|█████████▉| 369037/371472 [7:12:05<10:45,  3.77it/s]
 99%|█████████▉| 369038/371472 [7:12:06<11:16,  3.60it/s]
 99%|█████████▉| 369039/371472 [7:12:06<11:06,  3.65it/s]
 99%|█████████▉| 369040/371472 [7:12:06<11:04,  3.66it/s]
                                                         
{'loss': 2.5806, 'learning_rate': 1.0589540821823356e-07, 'epoch': 15.9}
+
 99%|█████████▉| 369040/371472 [7:12:06<11:04,  3.66it/s]
 99%|█████████▉| 369041/371472 [7:12:06<11:20,  3.57it/s]
 99%|█████████▉| 369042/371472 [7:12:07<10:57,  3.70it/s]
 99%|█████████▉| 369043/371472 [7:12:07<10:53,  3.72it/s]
 99%|█████████▉| 369044/371472 [7:12:07<11:18,  3.58it/s]
 99%|█████████▉| 369045/371472 [7:12:08<11:48,  3.43it/s]
 99%|█████████▉| 369046/371472 [7:12:08<11:32,  3.51it/s]
 99%|█████████▉| 369047/371472 [7:12:08<11:00,  3.67it/s]
 99%|█████████▉| 369048/371472 [7:12:08<10:40,  3.79it/s]
 99%|█████████▉| 369049/371472 [7:12:09<12:05,  3.34it/s]
 99%|█████████▉| 369050/371472 [7:12:09<11:51,  3.41it/s]
 99%|█████████▉| 369051/371472 [7:12:09<11:13,  3.60it/s]
 99%|█████████▉| 369052/371472 [7:12:09<11:09,  3.62it/s]
 99%|█████████▉| 369053/371472 [7:12:10<11:03,  3.65it/s]
 99%|█████████▉| 369054/371472 [7:12:10<10:53,  3.70it/s]
 99%|█████████▉| 369055/371472 [7:12:10<10:55,  3.69it/s]
 99%|█████████▉| 369056/371472 [7:12:11<11:06,  3.62it/s]
 99%|█████████▉| 369057/371472 [7:12:11<11:37,  3.46it/s]
 99%|█████████▉| 369058/371472 [7:12:11<11:07,  3.62it/s]
 99%|█████████▉| 369059/371472 [7:12:11<11:36,  3.47it/s]
 99%|█████████▉| 369060/371472 [7:12:12<12:03,  3.33it/s]
                                                         
{'loss': 2.5489, 'learning_rate': 1.058469262427546e-07, 'epoch': 15.9}
+
 99%|█████████▉| 369060/371472 [7:12:12<12:03,  3.33it/s]
 99%|█████████▉| 369061/371472 [7:12:12<11:58,  3.36it/s]
 99%|█████████▉| 369062/371472 [7:12:12<11:43,  3.42it/s]
 99%|█████████▉| 369063/371472 [7:12:13<11:57,  3.36it/s]
 99%|█████████▉| 369064/371472 [7:12:13<11:53,  3.38it/s]
 99%|█████████▉| 369065/371472 [7:12:13<11:54,  3.37it/s]
 99%|█████████▉| 369066/371472 [7:12:14<11:16,  3.56it/s]
 99%|█████████▉| 369067/371472 [7:12:14<10:42,  3.74it/s]
 99%|█████████▉| 369068/371472 [7:12:14<11:43,  3.42it/s]
 99%|█████████▉| 369069/371472 [7:12:14<11:28,  3.49it/s]
 99%|█████████▉| 369070/371472 [7:12:15<12:41,  3.15it/s]
 99%|█████████▉| 369071/371472 [7:12:15<12:06,  3.31it/s]
 99%|█████████▉| 369072/371472 [7:12:15<12:10,  3.28it/s]
 99%|█████████▉| 369073/371472 [7:12:16<11:35,  3.45it/s]
 99%|█████████▉| 369074/371472 [7:12:16<12:10,  3.28it/s]
 99%|█████████▉| 369075/371472 [7:12:16<12:10,  3.28it/s]
 99%|█████████▉| 369076/371472 [7:12:17<11:50,  3.37it/s]
 99%|█████████▉| 369077/371472 [7:12:17<11:39,  3.42it/s]
 99%|█████████▉| 369078/371472 [7:12:17<11:53,  3.36it/s]
 99%|█████████▉| 369079/371472 [7:12:17<11:58,  3.33it/s]
 99%|█████████▉| 369080/371472 [7:12:18<11:56,  3.34it/s]
                                                         
{'loss': 2.6366, 'learning_rate': 1.0579844426727578e-07, 'epoch': 15.9}
+
 99%|█████████▉| 369080/371472 [7:12:18<11:56,  3.34it/s]
 99%|█████████▉| 369081/371472 [7:12:18<12:55,  3.08it/s]
 99%|█████████▉| 369082/371472 [7:12:18<12:34,  3.17it/s]
 99%|█████████▉| 369083/371472 [7:12:19<11:50,  3.36it/s]
 99%|█████████▉| 369084/371472 [7:12:19<11:32,  3.45it/s]
 99%|█████████▉| 369085/371472 [7:12:19<11:14,  3.54it/s]
 99%|█████████▉| 369086/371472 [7:12:19<10:54,  3.65it/s]
 99%|█████████▉| 369087/371472 [7:12:20<11:26,  3.47it/s]
 99%|█████████▉| 369088/371472 [7:12:20<11:03,  3.59it/s]
 99%|█████████▉| 369089/371472 [7:12:20<10:50,  3.66it/s]
 99%|█████████▉| 369090/371472 [7:12:21<10:32,  3.77it/s]
 99%|█████████▉| 369091/371472 [7:12:21<10:51,  3.66it/s]
 99%|█████████▉| 369092/371472 [7:12:21<11:01,  3.60it/s]
 99%|█████████▉| 369093/371472 [7:12:21<10:56,  3.62it/s]
 99%|█��███████▉| 369094/371472 [7:12:22<11:18,  3.50it/s]
 99%|█████████▉| 369095/371472 [7:12:22<11:06,  3.56it/s]
 99%|█████████▉| 369096/371472 [7:12:22<11:15,  3.52it/s]
 99%|█████████▉| 369097/371472 [7:12:22<10:42,  3.70it/s]
 99%|█████████▉| 369098/371472 [7:12:23<11:01,  3.59it/s]
 99%|█████████▉| 369099/371472 [7:12:23<10:35,  3.74it/s]
 99%|█████████▉| 369100/371472 [7:12:23<10:57,  3.61it/s]
                                                         
{'loss': 2.6813, 'learning_rate': 1.0574996229179684e-07, 'epoch': 15.9}
+
 99%|█████████▉| 369100/371472 [7:12:23<10:57,  3.61it/s]
 99%|█████████▉| 369101/371472 [7:12:24<11:25,  3.46it/s]
 99%|█████████▉| 369102/371472 [7:12:24<11:27,  3.45it/s]
 99%|█████████▉| 369103/371472 [7:12:24<11:13,  3.52it/s]
 99%|█████████▉| 369104/371472 [7:12:24<10:51,  3.64it/s]
 99%|█████████▉| 369105/371472 [7:12:25<11:00,  3.59it/s]
 99%|█████████▉| 369106/371472 [7:12:25<11:37,  3.39it/s]
 99%|█████████▉| 369107/371472 [7:12:25<11:44,  3.36it/s]
 99%|█████████▉| 369108/371472 [7:12:26<11:01,  3.58it/s]
 99%|█████████▉| 369109/371472 [7:12:26<10:58,  3.59it/s]
 99%|█████████▉| 369110/371472 [7:12:26<11:14,  3.50it/s]
 99%|█████████▉| 369111/371472 [7:12:26<11:25,  3.44it/s]
 99%|█████████▉| 369112/371472 [7:12:27<11:33,  3.40it/s]
 99%|█████████▉| 369113/371472 [7:12:27<11:23,  3.45it/s]
 99%|█████████▉| 369114/371472 [7:12:27<11:21,  3.46it/s]
 99%|█████████▉| 369115/371472 [7:12:28<12:32,  3.13it/s]
 99%|█████████▉| 369116/371472 [7:12:28<11:44,  3.35it/s]
 99%|█████████▉| 369117/371472 [7:12:28<11:00,  3.57it/s]
 99%|█████████▉| 369118/371472 [7:12:28<10:38,  3.69it/s]
 99%|█████████▉| 369119/371472 [7:12:29<10:51,  3.61it/s]
 99%|█████████▉| 369120/371472 [7:12:29<11:13,  3.49it/s]
                                                         
{'loss': 2.533, 'learning_rate': 1.0570148031631799e-07, 'epoch': 15.9}
+
 99%|█████████▉| 369120/371472 [7:12:29<11:13,  3.49it/s]
 99%|█████████▉| 369121/371472 [7:12:29<11:02,  3.55it/s]
 99%|█████████▉| 369122/371472 [7:12:30<10:52,  3.60it/s]
 99%|█████████▉| 369123/371472 [7:12:30<10:42,  3.66it/s]
 99%|█████████▉| 369124/371472 [7:12:30<10:59,  3.56it/s]
 99%|█████████▉| 369125/371472 [7:12:30<10:38,  3.68it/s]
 99%|█████████▉| 369126/371472 [7:12:31<10:20,  3.78it/s]
 99%|█████████▉| 369127/371472 [7:12:31<10:51,  3.60it/s]
 99%|█████████▉| 369128/371472 [7:12:31<10:46,  3.63it/s]
 99%|█████████▉| 369129/371472 [7:12:32<11:03,  3.53it/s]
 99%|█████████▉| 369130/371472 [7:12:32<10:38,  3.67it/s]
 99%|█████████▉| 369131/371472 [7:12:32<10:32,  3.70it/s]
 99%|█████████▉| 369132/371472 [7:12:32<11:21,  3.44it/s]
 99%|█████████▉| 369133/371472 [7:12:33<10:52,  3.58it/s]
 99%|█████████▉| 369134/371472 [7:12:33<10:18,  3.78it/s]
 99%|█████████▉| 369135/371472 [7:12:33<11:20,  3.43it/s]
 99%|█████████▉| 369136/371472 [7:12:34<11:30,  3.38it/s]
 99%|█████████▉| 369137/371472 [7:12:34<10:58,  3.54it/s]
 99%|█████████▉| 369138/371472 [7:12:34<11:27,  3.40it/s]
 99%|█████████▉| 369139/371472 [7:12:34<11:29,  3.38it/s]
 99%|█████████▉| 369140/371472 [7:12:35<11:03,  3.51it/s]
                                                         
{'loss': 2.5997, 'learning_rate': 1.0565299834083905e-07, 'epoch': 15.9}
+
 99%|█████████▉| 369140/371472 [7:12:35<11:03,  3.51it/s]
 99%|█████████▉| 369141/371472 [7:12:35<11:09,  3.48it/s]
 99%|█████████▉| 369142/371472 [7:12:35<11:17,  3.44it/s]
 99%|█████████▉| 369143/371472 [7:12:36<10:36,  3.66it/s]
 99%|█████████▉| 369144/371472 [7:12:36<10:43,  3.62it/s]
 99%|█████████▉| 369145/371472 [7:12:36<10:31,  3.69it/s]
 99%|█████████▉| 369146/371472 [7:12:36<10:28,  3.70it/s]
 99%|█████████▉| 369147/371472 [7:12:37<10:57,  3.54it/s]
 99%|█████████▉| 369148/371472 [7:12:37<10:34,  3.66it/s]
 99%|█████████▉| 369149/371472 [7:12:37<11:31,  3.36it/s]
 99%|█████████▉| 369150/371472 [7:12:38<11:58,  3.23it/s]
 99%|█████████▉| 369151/371472 [7:12:38<11:14,  3.44it/s]
 99%|█████████▉| 369152/371472 [7:12:38<11:05,  3.49it/s]
 99%|█████████▉| 369153/371472 [7:12:38<11:02,  3.50it/s]
 99%|█████████▉| 369154/371472 [7:12:39<12:44,  3.03it/s]
 99%|█████████▉| 369155/371472 [7:12:39<12:42,  3.04it/s]
 99%|█████████▉| 369156/371472 [7:12:39<11:56,  3.23it/s]
 99%|█████████▉| 369157/371472 [7:12:40<11:43,  3.29it/s]
 99%|█████████▉| 369158/371472 [7:12:40<11:50,  3.26it/s]
 99%|█████████▉| 369159/371472 [7:12:40<11:42,  3.29it/s]
 99%|█████████▉| 369160/371472 [7:12:41<11:00,  3.50it/s]
                                                         
{'loss': 2.4482, 'learning_rate': 1.0560451636536021e-07, 'epoch': 15.9}
+
 99%|█████████▉| 369160/371472 [7:12:41<11:00,  3.50it/s]
 99%|█████████▉| 369161/371472 [7:12:41<10:32,  3.65it/s]
 99%|█████████▉| 369162/371472 [7:12:41<10:37,  3.62it/s]
 99%|█████████▉| 369163/371472 [7:12:41<10:44,  3.58it/s]
 99%|█████████▉| 369164/371472 [7:12:42<10:39,  3.61it/s]
 99%|█████████▉| 369165/371472 [7:12:42<10:52,  3.53it/s]
 99%|█████████▉| 369166/371472 [7:12:42<10:55,  3.52it/s]
 99%|█████████▉| 369167/371472 [7:12:43<11:22,  3.38it/s]
 99%|█████████▉| 369168/371472 [7:12:43<10:39,  3.60it/s]
 99%|█████████▉| 369169/371472 [7:12:43<10:37,  3.61it/s]
 99%|█████████▉| 369170/371472 [7:12:43<10:40,  3.60it/s]
 99%|█████████▉| 369171/371472 [7:12:44<10:48,  3.55it/s]
 99%|█████████▉| 369172/371472 [7:12:44<10:30,  3.65it/s]
 99%|█████████▉| 369173/371472 [7:12:44<10:54,  3.51it/s]
 99%|█████████▉| 369174/371472 [7:12:44<10:33,  3.63it/s]
 99%|█████████▉| 369175/371472 [7:12:45<10:21,  3.70it/s]
 99%|█████████▉| 369176/371472 [7:12:45<10:12,  3.75it/s]
 99%|█████████▉| 369177/371472 [7:12:45<09:59,  3.83it/s]
 99%|█████████▉| 369178/371472 [7:12:46<10:43,  3.56it/s]
 99%|█████████▉| 369179/371472 [7:12:46<10:35,  3.61it/s]
 99%|█████████▉| 369180/371472 [7:12:46<10:51,  3.52it/s]
                                                         
{'loss': 2.5604, 'learning_rate': 1.0555603438988127e-07, 'epoch': 15.9}
+
 99%|█████████▉| 369180/371472 [7:12:46<10:51,  3.52it/s]
 99%|█████████▉| 369181/371472 [7:12:47<12:06,  3.16it/s]
 99%|█████████▉| 369182/371472 [7:12:47<12:19,  3.10it/s]
 99%|█████████▉| 369183/371472 [7:12:47<11:26,  3.33it/s]
 99%|█████████▉| 369184/371472 [7:12:47<11:04,  3.44it/s]
 99%|█████████▉| 369185/371472 [7:12:48<11:11,  3.41it/s]
 99%|█████████▉| 369186/371472 [7:12:48<11:25,  3.34it/s]
 99%|█████████▉| 369187/371472 [7:12:48<10:53,  3.50it/s]
 99%|█████████▉| 369188/371472 [7:12:49<11:13,  3.39it/s]
 99%|█████████▉| 369189/371472 [7:12:49<12:06,  3.14it/s]
 99%|█████████▉| 369190/371472 [7:12:49<11:24,  3.33it/s]
 99%|█████████▉| 369191/371472 [7:12:49<11:15,  3.38it/s]
 99%|█████████▉| 369192/371472 [7:12:50<11:03,  3.44it/s]
 99%|█████████▉| 369193/371472 [7:12:50<10:56,  3.47it/s]
 99%|█████████▉| 369194/371472 [7:12:50<10:38,  3.57it/s]
 99%|█████████▉| 369195/371472 [7:12:51<10:35,  3.59it/s]
 99%|█████████▉| 369196/371472 [7:12:51<10:21,  3.66it/s]
 99%|█████████▉| 369197/371472 [7:12:51<10:08,  3.74it/s]
 99%|█████████▉| 369198/371472 [7:12:51<10:38,  3.56it/s]
 99%|█████████▉| 369199/371472 [7:12:52<10:25,  3.63it/s]
 99%|█████████▉| 369200/371472 [7:12:52<10:46,  3.51it/s]
                                                         
{'loss': 2.5789, 'learning_rate': 1.0550755241440242e-07, 'epoch': 15.9}
+
 99%|█████████▉| 369200/371472 [7:12:52<10:46,  3.51it/s]
 99%|█████████▉| 369201/371472 [7:12:52<11:03,  3.42it/s]
 99%|█████████▉| 369202/371472 [7:12:53<10:53,  3.48it/s]
 99%|█████████▉| 369203/371472 [7:12:53<10:26,  3.62it/s]
 99%|█████████▉| 369204/371472 [7:12:53<10:25,  3.62it/s]
 99%|█████████▉| 369205/371472 [7:12:53<10:20,  3.65it/s]
 99%|█████████▉| 369206/371472 [7:12:54<10:39,  3.55it/s]
 99%|█████████▉| 369207/371472 [7:12:54<10:26,  3.61it/s]
 99%|█████████▉| 369208/371472 [7:12:54<10:14,  3.68it/s]
 99%|███���█████▉| 369209/371472 [7:12:54<10:27,  3.61it/s]
 99%|█████████▉| 369210/371472 [7:12:55<10:48,  3.49it/s]
 99%|█████████▉| 369211/371472 [7:12:55<11:35,  3.25it/s]
 99%|█████████▉| 369212/371472 [7:12:55<12:05,  3.11it/s]
 99%|█████████▉| 369213/371472 [7:12:56<11:20,  3.32it/s]
 99%|█████████▉| 369214/371472 [7:12:56<11:30,  3.27it/s]
 99%|█████████▉| 369215/371472 [7:12:56<11:18,  3.33it/s]
 99%|█████████▉| 369216/371472 [7:12:57<11:22,  3.30it/s]
 99%|█████████▉| 369217/371472 [7:12:57<12:32,  3.00it/s]
 99%|█████████▉| 369218/371472 [7:12:57<11:59,  3.13it/s]
 99%|█████████▉| 369219/371472 [7:12:58<11:45,  3.19it/s]
 99%|█████████▉| 369220/371472 [7:12:58<11:08,  3.37it/s]
                                                         
{'loss': 2.6682, 'learning_rate': 1.0545907043892348e-07, 'epoch': 15.9}
+
 99%|█████████▉| 369220/371472 [7:12:58<11:08,  3.37it/s]
 99%|█████████▉| 369221/371472 [7:12:58<14:20,  2.62it/s]
 99%|█████████▉| 369222/371472 [7:12:59<12:53,  2.91it/s]
 99%|█████████▉| 369223/371472 [7:12:59<12:38,  2.96it/s]
 99%|█████████▉| 369224/371472 [7:12:59<11:38,  3.22it/s]
 99%|█████████▉| 369225/371472 [7:13:00<11:42,  3.20it/s]
 99%|█████████▉| 369226/371472 [7:13:00<11:01,  3.40it/s]
 99%|█████████▉| 369227/371472 [7:13:00<10:46,  3.47it/s]
 99%|█████████▉| 369228/371472 [7:13:00<10:26,  3.58it/s]
 99%|█████████▉| 369229/371472 [7:13:01<10:22,  3.60it/s]
 99%|█████████▉| 369230/371472 [7:13:01<10:15,  3.64it/s]
 99%|█████████▉| 369231/371472 [7:13:01<10:38,  3.51it/s]
 99%|█████████▉| 369232/371472 [7:13:02<10:33,  3.54it/s]
 99%|█████████▉| 369233/371472 [7:13:02<10:36,  3.52it/s]
 99%|█████████▉| 369234/371472 [7:13:02<10:31,  3.54it/s]
 99%|█████████▉| 369235/371472 [7:13:02<11:08,  3.35it/s]
 99%|█████████▉| 369236/371472 [7:13:03<10:55,  3.41it/s]
 99%|█████████▉| 369237/371472 [7:13:03<10:24,  3.58it/s]
 99%|█████████▉| 369238/371472 [7:13:03<10:12,  3.65it/s]
 99%|█████████▉| 369239/371472 [7:13:03<10:12,  3.65it/s]
 99%|█████████▉| 369240/371472 [7:13:04<10:41,  3.48it/s]
                                                         
{'loss': 2.453, 'learning_rate': 1.0541058846344454e-07, 'epoch': 15.9}
+
 99%|█████████▉| 369240/371472 [7:13:04<10:41,  3.48it/s]
 99%|█████████▉| 369241/371472 [7:13:04<11:31,  3.22it/s]
 99%|█████████▉| 369242/371472 [7:13:04<11:04,  3.36it/s]
 99%|█████████▉| 369243/371472 [7:13:05<11:09,  3.33it/s]
 99%|█████████▉| 369244/371472 [7:13:05<11:14,  3.30it/s]
 99%|█████████▉| 369245/371472 [7:13:05<10:31,  3.52it/s]
 99%|█████████▉| 369246/371472 [7:13:06<10:27,  3.55it/s]
 99%|█████████▉| 369247/371472 [7:13:06<10:18,  3.60it/s]
 99%|█████████▉| 369248/371472 [7:13:06<10:32,  3.52it/s]
 99%|█████████▉| 369249/371472 [7:13:06<11:01,  3.36it/s]
 99%|█████████▉| 369250/371472 [7:13:07<10:44,  3.45it/s]
 99%|█████████▉| 369251/371472 [7:13:07<10:20,  3.58it/s]
 99%|█████████▉| 369252/371472 [7:13:07<10:04,  3.67it/s]
 99%|█████████▉| 369253/371472 [7:13:07<09:41,  3.81it/s]
 99%|█████████▉| 369254/371472 [7:13:08<10:17,  3.59it/s]
 99%|█████████▉| 369255/371472 [7:13:08<10:25,  3.54it/s]
 99%|█████████▉| 369256/371472 [7:13:08<10:19,  3.58it/s]
 99%|█████████▉| 369257/371472 [7:13:09<10:09,  3.64it/s]
 99%|█████████▉| 369258/371472 [7:13:09<10:54,  3.38it/s]
 99%|█████████▉| 369259/371472 [7:13:09<10:49,  3.41it/s]
 99%|█████████▉| 369260/371472 [7:13:10<10:21,  3.56it/s]
                                                         
{'loss': 2.4837, 'learning_rate': 1.053621064879657e-07, 'epoch': 15.9}
+
 99%|█████████▉| 369260/371472 [7:13:10<10:21,  3.56it/s]
 99%|█████████▉| 369261/371472 [7:13:10<10:56,  3.37it/s]
 99%|█████████▉| 369262/371472 [7:13:10<10:57,  3.36it/s]
 99%|█████████▉| 369263/371472 [7:13:10<11:10,  3.30it/s]
 99%|█████████▉| 369264/371472 [7:13:11<10:38,  3.46it/s]
 99%|█████████▉| 369265/371472 [7:13:11<10:21,  3.55it/s]
 99%|█████████▉| 369266/371472 [7:13:11<10:47,  3.41it/s]
 99%|█████████▉| 369267/371472 [7:13:12<10:15,  3.58it/s]
 99%|█████████▉| 369268/371472 [7:13:12<10:52,  3.38it/s]
 99%|█████████▉| 369269/371472 [7:13:12<10:22,  3.54it/s]
 99%|█████████▉| 369270/371472 [7:13:12<09:56,  3.69it/s]
 99%|█████████▉| 369271/371472 [7:13:13<10:05,  3.63it/s]
 99%|█████████▉| 369272/371472 [7:13:13<10:42,  3.43it/s]
 99%|█████████▉| 369273/371472 [7:13:13<10:55,  3.36it/s]
 99%|█████████▉| 369274/371472 [7:13:14<10:33,  3.47it/s]
 99%|█████████▉| 369275/371472 [7:13:14<10:55,  3.35it/s]
 99%|█████████▉| 369276/371472 [7:13:14<10:28,  3.50it/s]
 99%|█████████▉| 369277/371472 [7:13:14<10:15,  3.56it/s]
 99%|█████████▉| 369278/371472 [7:13:15<11:23,  3.21it/s]
 99%|█████████▉| 369279/371472 [7:13:15<11:23,  3.21it/s]
 99%|█████████▉| 369280/371472 [7:13:15<10:48,  3.38it/s]
                                                         
{'loss': 2.5853, 'learning_rate': 1.0531362451248675e-07, 'epoch': 15.91}
+
 99%|█████████▉| 369280/371472 [7:13:15<10:48,  3.38it/s]
 99%|█████████▉| 369281/371472 [7:13:16<11:06,  3.29it/s]
 99%|█████████▉| 369282/371472 [7:13:16<10:32,  3.46it/s]
 99%|█████████▉| 369283/371472 [7:13:16<10:28,  3.48it/s]
 99%|█████████▉| 369284/371472 [7:13:17<10:29,  3.48it/s]
 99%|█████████▉| 369285/371472 [7:13:17<10:28,  3.48it/s]
 99%|█████████▉| 369286/371472 [7:13:17<10:14,  3.55it/s]
 99%|█████████▉| 369287/371472 [7:13:17<09:52,  3.69it/s]
 99%|█████████▉| 369288/371472 [7:13:18<11:07,  3.27it/s]
 99%|█████████▉| 369289/371472 [7:13:18<10:45,  3.38it/s]
 99%|█████████▉| 369290/371472 [7:13:18<10:11,  3.57it/s]
 99%|█████████▉| 369291/371472 [7:13:19<11:08,  3.26it/s]
 99%|█████████▉| 369292/371472 [7:13:19<10:55,  3.33it/s]
 99%|█████████▉| 369293/371472 [7:13:19<10:28,  3.47it/s]
 99%|█████████▉| 369294/371472 [7:13:19<10:43,  3.38it/s]
 99%|█████████▉| 369295/371472 [7:13:20<10:28,  3.47it/s]
 99%|█████████▉| 369296/371472 [7:13:20<10:26,  3.47it/s]
 99%|█████████▉| 369297/371472 [7:13:20<10:19,  3.51it/s]
 99%|█████████▉| 369298/371472 [7:13:21<10:21,  3.50it/s]
 99%|█████████▉| 369299/371472 [7:13:21<10:02,  3.61it/s]
 99%|█████████▉| 369300/371472 [7:13:21<09:51,  3.67it/s]
                                                         
{'loss': 2.6337, 'learning_rate': 1.0526514253700791e-07, 'epoch': 15.91}
+
 99%|█████████▉| 369300/371472 [7:13:21<09:51,  3.67it/s]
 99%|█████████▉| 369301/371472 [7:13:21<09:40,  3.74it/s]
 99%|█████████▉| 369302/371472 [7:13:22<09:35,  3.77it/s]
 99%|█████████▉| 369303/371472 [7:13:22<09:45,  3.70it/s]
 99%|█████████▉| 369304/371472 [7:13:22<09:31,  3.79it/s]
 99%|█████████▉| 369305/371472 [7:13:22<09:43,  3.71it/s]
 99%|█████████▉| 369306/371472 [7:13:23<09:43,  3.71it/s]
 99%|█████████▉| 369307/371472 [7:13:23<09:42,  3.72it/s]
 99%|█████████▉| 369308/371472 [7:13:23<10:53,  3.31it/s]
 99%|█████████▉| 369309/371472 [7:13:24<11:14,  3.21it/s]
 99%|█████████▉| 369310/371472 [7:13:24<10:32,  3.42it/s]
 99%|█████████▉| 369311/371472 [7:13:24<10:11,  3.53it/s]
 99%|█████████▉| 369312/371472 [7:13:24<10:02,  3.58it/s]
 99%|█████████▉| 369313/371472 [7:13:25<10:03,  3.58it/s]
 99%|█████████▉| 369314/371472 [7:13:25<10:06,  3.56it/s]
 99%|█████████▉| 369315/371472 [7:13:25<10:36,  3.39it/s]
 99%|█████████▉| 369316/371472 [7:13:26<11:07,  3.23it/s]
 99%|█████████▉| 369317/371472 [7:13:26<11:17,  3.18it/s]
 99%|█████████▉| 369318/371472 [7:13:27<14:04,  2.55it/s]
 99%|█████████▉| 369319/371472 [7:13:27<12:41,  2.83it/s]
 99%|█████████▉| 369320/371472 [7:13:27<11:48,  3.04it/s]
                                                         
{'loss': 2.6394, 'learning_rate': 1.0521666056152897e-07, 'epoch': 15.91}
+
 99%|█████████▉| 369320/371472 [7:13:27<11:48,  3.04it/s]
 99%|█████████▉| 369321/371472 [7:13:27<11:16,  3.18it/s]
 99%|█████████▉| 369322/371472 [7:13:28<11:12,  3.20it/s]
 99%|█████████▉| 369323/371472 [7:13:28<11:02,  3.24it/s]
 99%|████���████▉| 369324/371472 [7:13:28<10:43,  3.34it/s]
 99%|█████████▉| 369325/371472 [7:13:29<10:30,  3.41it/s]
 99%|█████████▉| 369326/371472 [7:13:29<10:37,  3.37it/s]
 99%|█████████▉| 369327/371472 [7:13:29<10:15,  3.48it/s]
 99%|█████████▉| 369328/371472 [7:13:29<10:07,  3.53it/s]
 99%|█████████▉| 369329/371472 [7:13:30<09:54,  3.61it/s]
 99%|█████████▉| 369330/371472 [7:13:30<10:11,  3.50it/s]
 99%|█████████▉| 369331/371472 [7:13:30<09:46,  3.65it/s]
 99%|█████████▉| 369332/371472 [7:13:31<09:48,  3.64it/s]
 99%|█████████▉| 369333/371472 [7:13:31<09:37,  3.70it/s]
 99%|█████████▉| 369334/371472 [7:13:31<09:37,  3.70it/s]
 99%|█████████▉| 369335/371472 [7:13:31<09:26,  3.77it/s]
 99%|█████████▉| 369336/371472 [7:13:32<10:09,  3.51it/s]
 99%|█████████▉| 369337/371472 [7:13:32<09:52,  3.60it/s]
 99%|█████████▉| 369338/371472 [7:13:32<10:33,  3.37it/s]
 99%|█████████▉| 369339/371472 [7:13:33<10:41,  3.33it/s]
 99%|█████████▉| 369340/371472 [7:13:33<10:13,  3.48it/s]
                                                         
{'loss': 2.3991, 'learning_rate': 1.0516817858605012e-07, 'epoch': 15.91}
+
 99%|█████████▉| 369340/371472 [7:13:33<10:13,  3.48it/s]
 99%|█████████▉| 369341/371472 [7:13:33<10:07,  3.50it/s]
 99%|█████████▉| 369342/371472 [7:13:33<09:38,  3.68it/s]
 99%|█████████▉| 369343/371472 [7:13:34<09:47,  3.62it/s]
 99%|█████████▉| 369344/371472 [7:13:34<09:57,  3.56it/s]
 99%|█████████▉| 369345/371472 [7:13:34<09:34,  3.70it/s]
 99%|█████████▉| 369346/371472 [7:13:34<10:15,  3.45it/s]
 99%|█████████▉| 369347/371472 [7:13:35<10:41,  3.31it/s]
 99%|█████████▉| 369348/371472 [7:13:35<10:17,  3.44it/s]
 99%|█████████▉| 369349/371472 [7:13:35<10:07,  3.49it/s]
 99%|█████████▉| 369350/371472 [7:13:36<09:48,  3.60it/s]
 99%|█████████▉| 369351/371472 [7:13:36<10:12,  3.47it/s]
 99%|█████████▉| 369352/371472 [7:13:36<10:13,  3.45it/s]
 99%|█████████▉| 369353/371472 [7:13:37<10:43,  3.29it/s]
 99%|█████████▉| 369354/371472 [7:13:37<10:13,  3.45it/s]
 99%|█████████▉| 369355/371472 [7:13:37<10:05,  3.50it/s]
 99%|█████████▉| 369356/371472 [7:13:37<09:57,  3.54it/s]
 99%|█████████▉| 369357/371472 [7:13:38<09:46,  3.60it/s]
 99%|█████████▉| 369358/371472 [7:13:38<09:41,  3.63it/s]
 99%|█████████▉| 369359/371472 [7:13:38<09:55,  3.55it/s]
 99%|█████████▉| 369360/371472 [7:13:38<09:41,  3.63it/s]
                                                         
{'loss': 2.6143, 'learning_rate': 1.0511969661057118e-07, 'epoch': 15.91}
+
 99%|█████████▉| 369360/371472 [7:13:38<09:41,  3.63it/s]
 99%|█████████▉| 369361/371472 [7:13:39<09:35,  3.66it/s]
 99%|█████████▉| 369362/371472 [7:13:39<09:43,  3.61it/s]
 99%|█████████▉| 369363/371472 [7:13:39<09:29,  3.70it/s]
 99%|█████████▉| 369364/371472 [7:13:40<09:34,  3.67it/s]
 99%|█████████▉| 369365/371472 [7:13:40<09:49,  3.58it/s]
 99%|█████████▉| 369366/371472 [7:13:40<09:33,  3.67it/s]
 99%|█████████▉| 369367/371472 [7:13:40<10:00,  3.50it/s]
 99%|█████████▉| 369368/371472 [7:13:41<09:55,  3.53it/s]
 99%|█████████▉| 369369/371472 [7:13:41<09:34,  3.66it/s]
 99%|█████████▉| 369370/371472 [7:13:41<09:45,  3.59it/s]
 99%|█████████▉| 369371/371472 [7:13:42<10:09,  3.45it/s]
 99%|█████████▉| 369372/371472 [7:13:42<10:07,  3.45it/s]
 99%|█████████▉| 369373/371472 [7:13:42<10:16,  3.40it/s]
 99%|█████████▉| 369374/371472 [7:13:42<09:52,  3.54it/s]
 99%|█████████▉| 369375/371472 [7:13:43<10:07,  3.45it/s]
 99%|█████████▉| 369376/371472 [7:13:43<09:52,  3.54it/s]
 99%|█████████▉| 369377/371472 [7:13:43<09:39,  3.62it/s]
 99%|█████████▉| 369378/371472 [7:13:44<10:07,  3.45it/s]
 99%|█████████▉| 369379/371472 [7:13:44<10:33,  3.30it/s]
 99%|█████████▉| 369380/371472 [7:13:44<10:24,  3.35it/s]
                                                         
{'loss': 2.6166, 'learning_rate': 1.0507121463509234e-07, 'epoch': 15.91}
+
 99%|█████████▉| 369380/371472 [7:13:44<10:24,  3.35it/s]
 99%|█████████▉| 369381/371472 [7:13:44<09:57,  3.50it/s]
 99%|█████████▉| 369382/371472 [7:13:45<09:52,  3.53it/s]
 99%|█████████▉| 369383/371472 [7:13:45<09:37,  3.62it/s]
 99%|█████████▉| 369384/371472 [7:13:45<09:34,  3.63it/s]
 99%|█████████▉| 369385/371472 [7:13:45<09:16,  3.75it/s]
 99%|█████████▉| 369386/371472 [7:13:46<09:05,  3.83it/s]
 99%|█████████▉| 369387/371472 [7:13:46<09:33,  3.64it/s]
 99%|█████████▉| 369388/371472 [7:13:46<09:56,  3.49it/s]
 99%|█████████▉| 369389/371472 [7:13:47<09:43,  3.57it/s]
 99%|█████████▉| 369390/371472 [7:13:47<10:04,  3.45it/s]
 99%|█████████▉| 369391/371472 [7:13:47<09:44,  3.56it/s]
 99%|█████████▉| 369392/371472 [7:13:47<09:46,  3.55it/s]
 99%|█████████▉| 369393/371472 [7:13:48<09:36,  3.61it/s]
 99%|█████████▉| 369394/371472 [7:13:48<10:04,  3.44it/s]
 99%|█████████▉| 369395/371472 [7:13:48<09:55,  3.49it/s]
 99%|█████████▉| 369396/371472 [7:13:49<09:46,  3.54it/s]
 99%|█████████▉| 369397/371472 [7:13:49<09:45,  3.55it/s]
 99%|█████████▉| 369398/371472 [7:13:49<09:35,  3.61it/s]
 99%|█████████▉| 369399/371472 [7:13:49<09:28,  3.64it/s]
 99%|█████████▉| 369400/371472 [7:13:50<09:38,  3.58it/s]
                                                         
{'loss': 2.6362, 'learning_rate': 1.050227326596134e-07, 'epoch': 15.91}
+
 99%|█████████▉| 369400/371472 [7:13:50<09:38,  3.58it/s]
 99%|█████████▉| 369401/371472 [7:13:50<09:22,  3.68it/s]
 99%|█████████▉| 369402/371472 [7:13:50<09:24,  3.67it/s]
 99%|█████████▉| 369403/371472 [7:13:51<09:32,  3.62it/s]
 99%|█████████▉| 369404/371472 [7:13:51<09:18,  3.70it/s]
 99%|█████████▉| 369405/371472 [7:13:51<08:56,  3.85it/s]
 99%|█████████▉| 369406/371472 [7:13:51<10:20,  3.33it/s]
 99%|█████████▉| 369407/371472 [7:13:52<10:06,  3.41it/s]
 99%|█████████▉| 369408/371472 [7:13:52<09:32,  3.60it/s]
 99%|█████████▉| 369409/371472 [7:13:52<09:24,  3.66it/s]
 99%|█████████▉| 369410/371472 [7:13:52<09:18,  3.69it/s]
 99%|█████████▉| 369411/371472 [7:13:53<09:06,  3.77it/s]
 99%|█████████▉| 369412/371472 [7:13:53<09:06,  3.77it/s]
 99%|█████████▉| 369413/371472 [7:13:53<09:05,  3.78it/s]
 99%|█████████▉| 369414/371472 [7:13:53<08:56,  3.83it/s]
 99%|█████████▉| 369415/371472 [7:13:54<09:33,  3.59it/s]
 99%|█████████▉| 369416/371472 [7:13:54<09:54,  3.46it/s]
 99%|█████████▉| 369417/371472 [7:13:54<09:49,  3.49it/s]
 99%|█████████▉| 369418/371472 [7:13:55<09:52,  3.47it/s]
 99%|█████████▉| 369419/371472 [7:13:55<10:36,  3.23it/s]
 99%|█████████▉| 369420/371472 [7:13:55<10:08,  3.37it/s]
                                                         
{'loss': 2.644, 'learning_rate': 1.0497425068413455e-07, 'epoch': 15.91}
+
 99%|█████████▉| 369420/371472 [7:13:55<10:08,  3.37it/s]
 99%|█████████▉| 369421/371472 [7:13:56<10:41,  3.20it/s]
 99%|█████████▉| 369422/371472 [7:13:56<10:15,  3.33it/s]
 99%|█████████▉| 369423/371472 [7:13:56<09:46,  3.49it/s]
 99%|█████████▉| 369424/371472 [7:13:56<09:50,  3.47it/s]
 99%|█████████▉| 369425/371472 [7:13:57<09:21,  3.65it/s]
 99%|█████████▉| 369426/371472 [7:13:57<09:41,  3.52it/s]
 99%|█████████▉| 369427/371472 [7:13:57<09:52,  3.45it/s]
 99%|█████████▉| 369428/371472 [7:13:58<10:20,  3.29it/s]
 99%|█████████▉| 369429/371472 [7:13:58<09:38,  3.53it/s]
 99%|█████████▉| 369430/371472 [7:13:58<09:25,  3.61it/s]
 99%|█████████▉| 369431/371472 [7:13:58<09:31,  3.57it/s]
 99%|█████████▉| 369432/371472 [7:13:59<09:37,  3.53it/s]
 99%|█████████▉| 369433/371472 [7:13:59<09:32,  3.56it/s]
 99%|█████████▉| 369434/371472 [7:13:59<09:11,  3.70it/s]
 99%|█████████▉| 369435/371472 [7:14:00<09:16,  3.66it/s]
 99%|█████████▉| 369436/371472 [7:14:00<09:05,  3.73it/s]
 99%|█████████▉| 369437/371472 [7:14:00<09:06,  3.72it/s]
 99%|█████████▉| 369438/371472 [7:14:00<09:03,  3.75it/s]
 99%|█████████▉| 369439/371472 [7:14:01<09:00,  3.76it/s]
 99%|█████████▉| 369440/371472 [7:14:01<08:58,  3.77it/s]
                                                         
{'loss': 2.7068, 'learning_rate': 1.0492576870865562e-07, 'epoch': 15.91}
+
 99%|█████████▉| 369440/371472 [7:14:01<08:58,  3.77it/s]
 99%|█████████▉| 369441/371472 [7:14:01<09:11,  3.68it/s]
 99%|█████████▉| 369442/371472 [7:14:01<09:53,  3.42it/s]
 99%|█████████▉| 369443/371472 [7:14:02<10:07,  3.34it/s]
 99%|█████████▉| 369444/371472 [7:14:02<09:48,  3.45it/s]
 99%|█████████▉| 369445/371472 [7:14:02<09:47,  3.45it/s]
 99%|█████████▉| 369446/371472 [7:14:03<09:45,  3.46it/s]
 99%|█████████▉| 369447/371472 [7:14:03<09:43,  3.47it/s]
 99%|█████████▉| 369448/371472 [7:14:03<09:37,  3.51it/s]
 99%|█████████▉| 369449/371472 [7:14:03<09:38,  3.50it/s]
 99%|█████████▉| 369450/371472 [7:14:04<09:46,  3.45it/s]
 99%|█████████▉| 369451/371472 [7:14:04<09:26,  3.57it/s]
 99%|█████████▉| 369452/371472 [7:14:04<09:26,  3.56it/s]
 99%|█████████▉| 369453/371472 [7:14:05<09:59,  3.37it/s]
 99%|█████████▉| 369454/371472 [7:14:05<09:34,  3.51it/s]
 99%|█████████▉| 369455/371472 [7:14:05<09:07,  3.68it/s]
 99%|█████████▉| 369456/371472 [7:14:05<08:56,  3.76it/s]
 99%|█████████▉| 369457/371472 [7:14:06<08:54,  3.77it/s]
 99%|█████████▉| 369458/371472 [7:14:06<09:31,  3.52it/s]
 99%|█████████▉| 369459/371472 [7:14:06<09:34,  3.50it/s]
 99%|█████████▉| 369460/371472 [7:14:07<09:25,  3.56it/s]
                                                         
{'loss': 2.6062, 'learning_rate': 1.0487728673317678e-07, 'epoch': 15.91}
+
 99%|█████████▉| 369460/371472 [7:14:07<09:25,  3.56it/s]
 99%|█████████▉| 369461/371472 [7:14:07<09:13,  3.63it/s]
 99%|█████████▉| 369462/371472 [7:14:07<09:10,  3.65it/s]
 99%|█████████▉| 369463/371472 [7:14:07<09:01,  3.71it/s]
 99%|█████████▉| 369464/371472 [7:14:08<09:02,  3.70it/s]
 99%|█████████▉| 369465/371472 [7:14:08<09:01,  3.71it/s]
 99%|█████████▉| 369466/371472 [7:14:08<09:00,  3.71it/s]
 99%|█████████▉| 369467/371472 [7:14:09<09:37,  3.47it/s]
 99%|█████████▉| 369468/371472 [7:14:09<09:19,  3.58it/s]
 99%|█████████▉| 369469/371472 [7:14:09<09:40,  3.45it/s]
 99%|█████████▉| 369470/371472 [7:14:09<09:07,  3.66it/s]
 99%|█████████▉| 369471/371472 [7:14:10<10:40,  3.12it/s]
 99%|█████████▉| 369472/371472 [7:14:10<10:10,  3.28it/s]
 99%|█████████▉| 369473/371472 [7:14:10<09:54,  3.36it/s]
 99%|█████████▉| 369474/371472 [7:14:11<10:08,  3.28it/s]
 99%|█████████▉| 369475/371472 [7:14:11<09:46,  3.40it/s]
 99%|█████████▉| 369476/371472 [7:14:11<09:38,  3.45it/s]
 99%|█████████▉| 369477/371472 [7:14:11<09:24,  3.53it/s]
 99%|█████████▉| 369478/371472 [7:14:12<09:29,  3.50it/s]
 99%|█████████▉| 369479/371472 [7:14:12<09:30,  3.49it/s]
 99%|█████████▉| 369480/371472 [7:14:12<09:18,  3.56it/s]
                                                         
{'loss': 2.8025, 'learning_rate': 1.0482880475769783e-07, 'epoch': 15.91}
+
 99%|█████████▉| 369480/371472 [7:14:12<09:18,  3.56it/s]
 99%|█████████▉| 369481/371472 [7:14:13<09:11,  3.61it/s]
 99%|█████████▉| 369482/371472 [7:14:13<09:45,  3.40it/s]
 99%|█████████▉| 369483/371472 [7:14:13<09:17,  3.57it/s]
 99%|█████████▉| 369484/371472 [7:14:13<09:20,  3.54it/s]
 99%|█████████▉| 369485/371472 [7:14:14<09:09,  3.62it/s]
 99%|█████████▉| 369486/371472 [7:14:14<08:52,  3.73it/s]
 99%|█████████▉| 369487/371472 [7:14:14<08:57,  3.69it/s]
 99%|█████████▉| 369488/371472 [7:14:15<09:28,  3.49it/s]
 99%|█████████▉| 369489/371472 [7:14:15<09:08,  3.62it/s]
 99%|█████████▉| 369490/371472 [7:14:15<09:05,  3.64it/s]
 99%|█████████▉| 369491/371472 [7:14:15<09:27,  3.49it/s]
 99%|█████████▉| 369492/371472 [7:14:16<09:57,  3.32it/s]
 99%|█████████▉| 369493/371472 [7:14:16<09:36,  3.43it/s]
 99%|█████████▉| 369494/371472 [7:14:16<09:31,  3.46it/s]
 99%|█████████▉| 369495/371472 [7:14:17<09:17,  3.54it/s]
 99%|█████████▉| 369496/371472 [7:14:17<09:33,  3.45it/s]
 99%|█████████▉| 369497/371472 [7:14:17<10:00,  3.29it/s]
 99%|█████████▉| 369498/371472 [7:14:17<09:49,  3.35it/s]
 99%|█████████��| 369499/371472 [7:14:18<09:40,  3.40it/s]
 99%|█████████▉| 369500/371472 [7:14:18<09:58,  3.30it/s]
                                                         
{'loss': 2.4199, 'learning_rate': 1.0478032278221898e-07, 'epoch': 15.92}
+
 99%|█████████▉| 369500/371472 [7:14:18<09:58,  3.30it/s]
 99%|█████████▉| 369501/371472 [7:14:18<09:38,  3.41it/s]
 99%|█████████▉| 369502/371472 [7:14:19<09:24,  3.49it/s]
 99%|█████████▉| 369503/371472 [7:14:19<09:32,  3.44it/s]
 99%|█████████▉| 369504/371472 [7:14:19<09:17,  3.53it/s]
 99%|█████████▉| 369505/371472 [7:14:19<09:10,  3.58it/s]
 99%|█████████▉| 369506/371472 [7:14:20<09:03,  3.61it/s]
 99%|█████████▉| 369507/371472 [7:14:20<08:46,  3.74it/s]
 99%|█████████▉| 369508/371472 [7:14:20<08:30,  3.84it/s]
 99%|█████████▉| 369509/371472 [7:14:20<08:32,  3.83it/s]
 99%|█████████▉| 369510/371472 [7:14:21<08:43,  3.75it/s]
 99%|█████████▉| 369511/371472 [7:14:21<08:43,  3.74it/s]
 99%|█████████▉| 369512/371472 [7:14:21<09:01,  3.62it/s]
 99%|█████████▉| 369513/371472 [7:14:22<09:11,  3.55it/s]
 99%|█████████▉| 369514/371472 [7:14:22<09:01,  3.62it/s]
 99%|█████████▉| 369515/371472 [7:14:22<08:37,  3.78it/s]
 99%|█████████▉| 369516/371472 [7:14:22<08:43,  3.74it/s]
 99%|█████████▉| 369517/371472 [7:14:23<08:33,  3.81it/s]
 99%|█████████▉| 369518/371472 [7:14:23<09:02,  3.60it/s]
 99%|█████████▉| 369519/371472 [7:14:23<08:58,  3.63it/s]
 99%|█████████▉| 369520/371472 [7:14:24<09:42,  3.35it/s]
                                                         
{'loss': 2.5001, 'learning_rate': 1.0473184080674004e-07, 'epoch': 15.92}
+
 99%|█████████▉| 369520/371472 [7:14:24<09:42,  3.35it/s]
 99%|█████████▉| 369521/371472 [7:14:24<09:31,  3.41it/s]
 99%|█████████▉| 369522/371472 [7:14:24<09:54,  3.28it/s]
 99%|█████████▉| 369523/371472 [7:14:24<09:51,  3.29it/s]
 99%|█████████▉| 369524/371472 [7:14:25<09:30,  3.41it/s]
 99%|█████████▉| 369525/371472 [7:14:25<09:05,  3.57it/s]
 99%|█████████▉| 369526/371472 [7:14:25<08:58,  3.61it/s]
 99%|█████████▉| 369527/371472 [7:14:26<09:01,  3.59it/s]
 99%|█████████▉| 369528/371472 [7:14:26<08:50,  3.66it/s]
 99%|█████████▉| 369529/371472 [7:14:26<09:05,  3.56it/s]
 99%|█████████▉| 369530/371472 [7:14:26<09:37,  3.36it/s]
 99%|█████████▉| 369531/371472 [7:14:27<09:15,  3.49it/s]
 99%|█████████▉| 369532/371472 [7:14:27<09:06,  3.55it/s]
 99%|█████████▉| 369533/371472 [7:14:27<09:07,  3.54it/s]
 99%|█████████▉| 369534/371472 [7:14:28<09:53,  3.27it/s]
 99%|█████████▉| 369535/371472 [7:14:28<09:33,  3.38it/s]
 99%|█████████▉| 369536/371472 [7:14:28<09:05,  3.55it/s]
 99%|█████████▉| 369537/371472 [7:14:28<09:37,  3.35it/s]
 99%|█████████▉| 369538/371472 [7:14:29<09:14,  3.49it/s]
 99%|█████████▉| 369539/371472 [7:14:29<09:07,  3.53it/s]
 99%|█████████▉| 369540/371472 [7:14:29<09:03,  3.55it/s]
                                                         
{'loss': 2.6826, 'learning_rate': 1.0468335883126121e-07, 'epoch': 15.92}
+
 99%|█████████▉| 369540/371472 [7:14:29<09:03,  3.55it/s]
 99%|█████████▉| 369541/371472 [7:14:30<09:28,  3.39it/s]
 99%|█████████▉| 369542/371472 [7:14:30<09:33,  3.37it/s]
 99%|█████████▉| 369543/371472 [7:14:30<09:28,  3.39it/s]
 99%|█████████▉| 369544/371472 [7:14:30<09:07,  3.52it/s]
 99%|█████████▉| 369545/371472 [7:14:31<08:43,  3.68it/s]
 99%|█████████▉| 369546/371472 [7:14:31<08:47,  3.65it/s]
 99%|█████████▉| 369547/371472 [7:14:31<08:30,  3.77it/s]
 99%|█████████▉| 369548/371472 [7:14:31<08:29,  3.78it/s]
 99%|█████████▉| 369549/371472 [7:14:32<08:21,  3.84it/s]
 99%|█████████▉| 369550/371472 [7:14:32<08:41,  3.69it/s]
 99%|█████████▉| 369551/371472 [7:14:32<08:30,  3.77it/s]
 99%|█████████▉| 369552/371472 [7:14:33<08:40,  3.69it/s]
 99%|█████████▉| 369553/371472 [7:14:33<08:31,  3.75it/s]
 99%|█████████▉| 369554/371472 [7:14:33<08:44,  3.65it/s]
 99%|█████████▉| 369555/371472 [7:14:33<09:15,  3.45it/s]
 99%|█████████▉| 369556/371472 [7:14:34<08:54,  3.59it/s]
 99%|█████████▉| 369557/371472 [7:14:34<08:32,  3.74it/s]
 99%|█████████▉| 369558/371472 [7:14:34<08:48,  3.62it/s]
 99%|█████████▉| 369559/371472 [7:14:35<09:24,  3.39it/s]
 99%|█████████▉| 369560/371472 [7:14:35<09:07,  3.49it/s]
                                                         
{'loss': 2.6841, 'learning_rate': 1.0463487685578227e-07, 'epoch': 15.92}
+
 99%|█████████▉| 369560/371472 [7:14:35<09:07,  3.49it/s]
 99%|█████████▉| 369561/371472 [7:14:35<09:02,  3.52it/s]
 99%|█████████▉| 369562/371472 [7:14:35<08:45,  3.64it/s]
 99%|█████████▉| 369563/371472 [7:14:36<08:22,  3.80it/s]
 99%|█████████▉| 369564/371472 [7:14:36<08:21,  3.80it/s]
 99%|█████████▉| 369565/371472 [7:14:36<08:39,  3.67it/s]
 99%|█████████▉| 369566/371472 [7:14:37<09:12,  3.45it/s]
 99%|█████████▉| 369567/371472 [7:14:37<09:20,  3.40it/s]
 99%|█████████▉| 369568/371472 [7:14:37<09:39,  3.29it/s]
 99%|█████████▉| 369569/371472 [7:14:37<09:45,  3.25it/s]
 99%|█████████▉| 369570/371472 [7:14:38<09:08,  3.47it/s]
 99%|█████████▉| 369571/371472 [7:14:38<09:20,  3.39it/s]
 99%|█████████▉| 369572/371472 [7:14:38<09:22,  3.38it/s]
 99%|█████████▉| 369573/371472 [7:14:39<09:27,  3.35it/s]
 99%|█████████▉| 369574/371472 [7:14:39<09:13,  3.43it/s]
 99%|█████████▉| 369575/371472 [7:14:39<08:55,  3.54it/s]
 99%|█████████▉| 369576/371472 [7:14:39<09:23,  3.36it/s]
 99%|█████████▉| 369577/371472 [7:14:40<09:01,  3.50it/s]
 99%|█████████▉| 369578/371472 [7:14:40<08:56,  3.53it/s]
 99%|█████████▉| 369579/371472 [7:14:40<08:54,  3.54it/s]
 99%|█████████▉| 369580/371472 [7:14:41<09:03,  3.48it/s]
                                                         
{'loss': 2.6618, 'learning_rate': 1.0458639488030342e-07, 'epoch': 15.92}
+
 99%|█████████▉| 369580/371472 [7:14:41<09:03,  3.48it/s]
 99%|█████████▉| 369581/371472 [7:14:41<09:10,  3.44it/s]
 99%|█████████▉| 369582/371472 [7:14:41<08:41,  3.62it/s]
 99%|█████████▉| 369583/371472 [7:14:41<08:35,  3.66it/s]
 99%|█████████▉| 369584/371472 [7:14:42<08:27,  3.72it/s]
 99%|█████████▉| 369585/371472 [7:14:42<08:39,  3.64it/s]
 99%|█████████▉| 369586/371472 [7:14:42<08:53,  3.54it/s]
 99%|█████████▉| 369587/371472 [7:14:43<08:52,  3.54it/s]
 99%|█████████▉| 369588/371472 [7:14:43<08:44,  3.59it/s]
 99%|█████████▉| 369589/371472 [7:14:43<08:30,  3.69it/s]
 99%|█████████▉| 369590/371472 [7:14:43<08:08,  3.85it/s]
 99%|█████████▉| 369591/371472 [7:14:44<08:41,  3.61it/s]
 99%|█████████▉| 369592/371472 [7:14:44<08:13,  3.81it/s]
 99%|█████████▉| 369593/371472 [7:14:44<08:10,  3.83it/s]
 99%|█████████▉| 369594/371472 [7:14:44<08:15,  3.79it/s]
 99%|█████████▉| 369595/371472 [7:14:45<08:43,  3.59it/s]
 99%|█████████▉| 369596/371472 [7:14:45<08:56,  3.49it/s]
 99%|█████████▉| 369597/371472 [7:14:45<08:44,  3.58it/s]
 99%|█████████▉| 369598/371472 [7:14:46<08:38,  3.61it/s]
 99%|█████████▉| 369599/371472 [7:14:46<08:39,  3.60it/s]
 99%|█████████▉| 369600/371472 [7:14:46<08:38,  3.61it/s]
                                                         
{'loss': 2.4794, 'learning_rate': 1.0453791290482447e-07, 'epoch': 15.92}
+
 99%|█████████▉| 369600/371472 [7:14:46<08:38,  3.61it/s]
 99%|█████████▉| 369601/371472 [7:14:46<08:29,  3.67it/s]
 99%|█████████▉| 369602/371472 [7:14:47<08:29,  3.67it/s]
 99%|█████████▉| 369603/371472 [7:14:47<08:32,  3.65it/s]
 99%|█████████▉| 369604/371472 [7:14:47<08:36,  3.62it/s]
 99%|█████████▉| 369605/371472 [7:14:47<09:09,  3.40it/s]
 99%|█████████▉| 369606/371472 [7:14:48<08:50,  3.52it/s]
 99%|█████████▉| 369607/371472 [7:14:48<09:23,  3.31it/s]
 99%|█████████▉| 369608/371472 [7:14:48<09:20,  3.33it/s]
 99%|█████████▉| 369609/371472 [7:14:49<09:35,  3.24it/s]
 99%|█████████▉| 369610/371472 [7:14:49<09:20,  3.32it/s]
 99%|█████████▉| 369611/371472 [7:14:49<09:22,  3.31it/s]
 99%|█████████▉| 369612/371472 [7:14:50<08:57,  3.46it/s]
 99%|█████████▉| 369613/371472 [7:14:50<08:59,  3.45it/s]
 99%|█████████▉| 369614/371472 [7:14:50<08:46,  3.53it/s]
100%|█████████▉| 369615/371472 [7:14:50<08:39,  3.57it/s]
100%|█████████▉| 369616/371472 [7:14:51<09:22,  3.30it/s]
100%|█████████▉| 369617/371472 [7:14:51<08:58,  3.45it/s]
100%|█████████▉| 369618/371472 [7:14:51<09:28,  3.26it/s]
100%|█████████▉| 369619/371472 [7:14:52<09:05,  3.39it/s]
100%|█████████▉| 369620/371472 [7:14:52<10:02,  3.07it/s]
                                                         
{'loss': 2.5479, 'learning_rate': 1.0448943092934565e-07, 'epoch': 15.92}
+
100%|█████████▉| 369620/371472 [7:14:52<10:02,  3.07it/s]
100%|█████████▉| 369621/371472 [7:14:52<09:39,  3.20it/s]
100%|█████████▉| 369622/371472 [7:14:53<09:05,  3.39it/s]
100%|█████████▉| 369623/371472 [7:14:53<09:04,  3.40it/s]
100%|█████████▉| 369624/371472 [7:14:53<08:41,  3.55it/s]
100%|█████████▉| 369625/371472 [7:14:53<08:48,  3.49it/s]
100%|█████████▉| 369626/371472 [7:14:54<08:48,  3.49it/s]
100%|█████████▉| 369627/371472 [7:14:54<08:34,  3.59it/s]
100%|█████████▉| 369628/371472 [7:14:54<09:15,  3.32it/s]
100%|█████████▉| 369629/371472 [7:14:55<08:53,  3.45it/s]
100%|█████████▉| 369630/371472 [7:14:55<08:34,  3.58it/s]
100%|█████████▉| 369631/371472 [7:14:55<09:18,  3.29it/s]
100%|█████████▉| 369632/371472 [7:14:55<09:05,  3.38it/s]
100%|█████████▉| 369633/371472 [7:14:56<09:13,  3.32it/s]
100%|█████████▉| 369634/371472 [7:14:56<09:02,  3.39it/s]
100%|█████████▉| 369635/371472 [7:14:56<08:48,  3.48it/s]
100%|█████████▉| 369636/371472 [7:14:57<08:38,  3.54it/s]
100%|█████████▉| 369637/371472 [7:14:57<08:38,  3.54it/s]
100%|█████████▉| 369638/371472 [7:14:57<08:30,  3.59it/s]
100%|█████████▉| 369639/371472 [7:14:57<08:30,  3.59it/s]
100%|█████████▉| 369640/371472 [7:14:58<08:43,  3.50it/s]
                                                         
{'loss': 2.5421, 'learning_rate': 1.044409489538667e-07, 'epoch': 15.92}
+
100%|█████████▉| 369640/371472 [7:14:58<08:43,  3.50it/s]
100%|█████████▉| 369641/371472 [7:14:58<08:32,  3.57it/s]
100%|█████████▉| 369642/371472 [7:14:58<08:32,  3.57it/s]
100%|█████████▉| 369643/371472 [7:14:59<08:20,  3.66it/s]
100%|█████████▉| 369644/371472 [7:14:59<08:41,  3.51it/s]
100%|█████████▉| 369645/371472 [7:14:59<08:28,  3.59it/s]
100%|█████████▉| 369646/371472 [7:14:59<08:04,  3.77it/s]
100%|█████████▉| 369647/371472 [7:15:00<08:04,  3.76it/s]
100%|█████████▉| 369648/371472 [7:15:00<07:49,  3.89it/s]
100%|█████████▉| 369649/371472 [7:15:00<08:15,  3.68it/s]
100%|█████████▉| 369650/371472 [7:15:01<09:01,  3.36it/s]
100%|█████████▉| 369651/371472 [7:15:01<08:39,  3.50it/s]
100%|█████████▉| 369652/371472 [7:15:01<08:37,  3.52it/s]
100%|█████████▉| 369653/371472 [7:15:01<08:29,  3.57it/s]
100%|█████████▉| 369654/371472 [7:15:02<08:34,  3.54it/s]
100%|█████████▉| 369655/371472 [7:15:02<08:12,  3.69it/s]
100%|█████████▉| 369656/371472 [7:15:02<08:07,  3.73it/s]
100%|█████████▉| 369657/371472 [7:15:02<08:06,  3.73it/s]
100%|█████████▉| 369658/371472 [7:15:03<08:04,  3.75it/s]
100%|█████████▉| 369659/371472 [7:15:03<07:47,  3.88it/s]
100%|█████████▉| 369660/371472 [7:15:03<07:58,  3.78it/s]
                                                         
{'loss': 2.6306, 'learning_rate': 1.0439246697838785e-07, 'epoch': 15.92}
+
100%|█████████▉| 369660/371472 [7:15:03<07:58,  3.78it/s]
100%|█████████▉| 369661/371472 [7:15:03<08:03,  3.75it/s]
100%|█████████▉| 369662/371472 [7:15:04<08:06,  3.72it/s]
100%|█████████▉| 369663/371472 [7:15:04<08:24,  3.59it/s]
100%|█████████▉| 369664/371472 [7:15:04<08:16,  3.64it/s]
100%|█████████▉| 369665/371472 [7:15:05<08:55,  3.37it/s]
100%|█████████▉| 369666/371472 [7:15:05<08:47,  3.42it/s]
100%|█████████▉| 369667/371472 [7:15:05<08:29,  3.54it/s]
100%|█████████▉| 369668/371472 [7:15:06<09:14,  3.25it/s]
100%|█████████▉| 369669/371472 [7:15:06<09:14,  3.25it/s]
100%|█████████▉| 369670/371472 [7:15:06<08:56,  3.36it/s]
100%|█████████▉| 369671/371472 [7:15:06<08:46,  3.42it/s]
100%|█████████▉| 369672/371472 [7:15:07<08:15,  3.63it/s]
100%|█████████▉| 369673/371472 [7:15:07<09:00,  3.33it/s]
100%|█████████▉| 369674/371472 [7:15:07<08:37,  3.48it/s]
100%|█████████▉| 369675/371472 [7:15:08<08:54,  3.36it/s]
100%|█████████▉| 369676/371472 [7:15:08<09:05,  3.30it/s]
100%|█████████▉| 369677/371472 [7:15:08<08:40,  3.45it/s]
100%|█████████▉| 369678/371472 [7:15:08<08:23,  3.56it/s]
100%|█████████▉| 369679/371472 [7:15:09<08:10,  3.66it/s]
100%|█████████▉| 369680/371472 [7:15:09<08:08,  3.67it/s]
                                                         
{'loss': 2.5146, 'learning_rate': 1.043439850029089e-07, 'epoch': 15.92}
+
100%|█████████▉| 369680/371472 [7:15:09<08:08,  3.67it/s]
100%|█████████▉| 369681/371472 [7:15:09<08:02,  3.71it/s]
100%|█████████▉| 369682/371472 [7:15:10<08:37,  3.46it/s]
100%|█████████▉| 369683/371472 [7:15:10<08:36,  3.47it/s]
100%|█████████▉| 369684/371472 [7:15:10<08:14,  3.62it/s]
100%|█████████▉| 369685/371472 [7:15:10<08:37,  3.46it/s]
100%|█████████▉| 369686/371472 [7:15:11<08:31,  3.49it/s]
100%|█████████▉| 369687/371472 [7:15:11<08:39,  3.44it/s]
100%|█████████▉| 369688/371472 [7:15:11<08:59,  3.30it/s]
100%|█████████▉| 369689/371472 [7:15:12<08:38,  3.44it/s]
100%|█████████▉| 369690/371472 [7:15:12<08:35,  3.45it/s]
100%|█████████▉| 369691/371472 [7:15:12<08:21,  3.55it/s]
100%|█████████▉| 369692/371472 [7:15:12<08:05,  3.66it/s]
100%|█████████▉| 369693/371472 [7:15:13<08:13,  3.60it/s]
100%|█████████▉| 369694/371472 [7:15:13<08:27,  3.51it/s]
100%|█████████▉| 369695/371472 [7:15:13<08:41,  3.41it/s]
100%|█████████▉| 369696/371472 [7:15:14<08:18,  3.56it/s]
100%|█████████▉| 369697/371472 [7:15:14<08:18,  3.56it/s]
100%|█████████▉| 369698/371472 [7:15:14<08:27,  3.49it/s]
100%|█████████▉| 369699/371472 [7:15:14<08:24,  3.51it/s]
100%|█████████▉| 369700/371472 [7:15:15<08:11,  3.61it/s]
                                                         
{'loss': 2.7233, 'learning_rate': 1.0429550302743007e-07, 'epoch': 15.92}
+
100%|█████████▉| 369700/371472 [7:15:15<08:11,  3.61it/s]
100%|█████████▉| 369701/371472 [7:15:15<08:20,  3.54it/s]
100%|█████████▉| 369702/371472 [7:15:15<08:47,  3.35it/s]
100%|█████████▉| 369703/371472 [7:15:16<08:32,  3.45it/s]
100%|█████████▉| 369704/371472 [7:15:16<08:12,  3.59it/s]
100%|█████████▉| 369705/371472 [7:15:16<08:13,  3.58it/s]
100%|█████████▉| 369706/371472 [7:15:16<08:08,  3.62it/s]
100%|█████████▉| 369707/371472 [7:15:17<08:05,  3.63it/s]
100%|█████████▉| 369708/371472 [7:15:17<08:03,  3.65it/s]
100%|█████████▉| 369709/371472 [7:15:17<07:54,  3.72it/s]
100%|█████████▉| 369710/371472 [7:15:17<08:01,  3.66it/s]
100%|█████████▉| 369711/371472 [7:15:18<08:03,  3.64it/s]
100%|█████████▉| 369712/371472 [7:15:18<08:03,  3.64it/s]
100%|█████████▉| 369713/371472 [7:15:18<07:54,  3.71it/s]
100%|█████████▉| 369714/371472 [7:15:19<08:03,  3.64it/s]
100%|█████████▉| 369715/371472 [7:15:19<08:38,  3.39it/s]
100%|█████████▉| 369716/371472 [7:15:19<08:22,  3.50it/s]
100%|█████████▉| 369717/371472 [7:15:19<08:18,  3.52it/s]
100%|█████████▉| 369718/371472 [7:15:20<08:28,  3.45it/s]
100%|█████████▉| 369719/371472 [7:15:20<08:38,  3.38it/s]
100%|█████████▉| 369720/371472 [7:15:20<08:35,  3.40it/s]
                                                         
{'loss': 2.497, 'learning_rate': 1.0424702105195113e-07, 'epoch': 15.92}
+
100%|█████████▉| 369720/371472 [7:15:20<08:35,  3.40it/s]
100%|█████████▉| 369721/371472 [7:15:21<08:27,  3.45it/s]
100%|█████████▉| 369722/371472 [7:15:21<08:14,  3.54it/s]
100%|█████████▉| 369723/371472 [7:15:21<08:15,  3.53it/s]
100%|█████████▉| 369724/371472 [7:15:21<08:12,  3.55it/s]
100%|█████████▉| 369725/371472 [7:15:22<08:04,  3.61it/s]
100%|█████████▉| 369726/371472 [7:15:22<07:59,  3.64it/s]
100%|█████████▉| 369727/371472 [7:15:22<08:00,  3.63it/s]
100%|█████████▉| 369728/371472 [7:15:22<07:45,  3.74it/s]
100%|█████████���| 369729/371472 [7:15:23<07:35,  3.83it/s]
100%|█████████▉| 369730/371472 [7:15:23<07:25,  3.91it/s]
100%|█████████▉| 369731/371472 [7:15:23<07:37,  3.81it/s]
100%|█████████▉| 369732/371472 [7:15:24<07:40,  3.78it/s]
100%|█████████▉| 369733/371472 [7:15:24<07:46,  3.72it/s]
100%|█████████▉| 369734/371472 [7:15:24<07:47,  3.72it/s]
100%|█████████▉| 369735/371472 [7:15:24<08:06,  3.57it/s]
100%|█████████▉| 369736/371472 [7:15:25<08:04,  3.58it/s]
100%|█████████▉| 369737/371472 [7:15:25<08:45,  3.30it/s]
100%|█████████▉| 369738/371472 [7:15:25<08:42,  3.32it/s]
100%|█████████▉| 369739/371472 [7:15:26<08:38,  3.34it/s]
100%|█████████▉| 369740/371472 [7:15:26<08:52,  3.25it/s]
                                                         
{'loss': 2.4305, 'learning_rate': 1.0419853907647228e-07, 'epoch': 15.93}
+
100%|█████████▉| 369740/371472 [7:15:26<08:52,  3.25it/s]
100%|█████████▉| 369741/371472 [7:15:26<09:25,  3.06it/s]
100%|█████████▉| 369742/371472 [7:15:27<09:09,  3.15it/s]
100%|█████████▉| 369743/371472 [7:15:27<08:49,  3.26it/s]
100%|█████████▉| 369744/371472 [7:15:27<08:26,  3.41it/s]
100%|█████████▉| 369745/371472 [7:15:27<08:13,  3.50it/s]
100%|█████████▉| 369746/371472 [7:15:28<08:07,  3.54it/s]
100%|█████████▉| 369747/371472 [7:15:28<08:25,  3.41it/s]
100%|█████████▉| 369748/371472 [7:15:28<08:09,  3.52it/s]
100%|█████████▉| 369749/371472 [7:15:29<08:14,  3.49it/s]
100%|█████████▉| 369750/371472 [7:15:29<08:06,  3.54it/s]
100%|█████████▉| 369751/371472 [7:15:29<07:49,  3.67it/s]
100%|█████████▉| 369752/371472 [7:15:29<08:11,  3.50it/s]
100%|█████████▉| 369753/371472 [7:15:30<08:11,  3.49it/s]
100%|█████████▉| 369754/371472 [7:15:30<08:23,  3.41it/s]
100%|█████████▉| 369755/371472 [7:15:30<08:11,  3.49it/s]
100%|█████████▉| 369756/371472 [7:15:31<08:11,  3.49it/s]
100%|█████████▉| 369757/371472 [7:15:31<08:05,  3.53it/s]
100%|█████████▉| 369758/371472 [7:15:31<08:31,  3.35it/s]
100%|█████████▉| 369759/371472 [7:15:31<08:37,  3.31it/s]
100%|█████████▉| 369760/371472 [7:15:32<09:11,  3.10it/s]
                                                         
{'loss': 2.5312, 'learning_rate': 1.0415005710099334e-07, 'epoch': 15.93}
+
100%|█████████▉| 369760/371472 [7:15:32<09:11,  3.10it/s]
100%|█████████▉| 369761/371472 [7:15:32<09:02,  3.15it/s]
100%|█████████▉| 369762/371472 [7:15:32<08:34,  3.32it/s]
100%|█████████▉| 369763/371472 [7:15:33<08:14,  3.46it/s]
100%|█████████▉| 369764/371472 [7:15:33<07:52,  3.61it/s]
100%|█████████▉| 369765/371472 [7:15:33<07:36,  3.74it/s]
100%|█████████▉| 369766/371472 [7:15:33<07:26,  3.82it/s]
100%|█████████▉| 369767/371472 [7:15:34<07:56,  3.58it/s]
100%|█████████▉| 369768/371472 [7:15:34<07:55,  3.58it/s]
100%|█████████▉| 369769/371472 [7:15:34<07:47,  3.64it/s]
100%|█████████▉| 369770/371472 [7:15:35<07:59,  3.55it/s]
100%|█████████▉| 369771/371472 [7:15:35<08:17,  3.42it/s]
100%|█████████▉| 369772/371472 [7:15:35<08:21,  3.39it/s]
100%|█████████▉| 369773/371472 [7:15:35<07:57,  3.55it/s]
100%|█████████▉| 369774/371472 [7:15:36<07:52,  3.59it/s]
100%|█████████▉| 369775/371472 [7:15:36<07:55,  3.57it/s]
100%|█████████▉| 369776/371472 [7:15:36<08:15,  3.42it/s]
100%|█████████▉| 369777/371472 [7:15:37<08:07,  3.48it/s]
100%|█████████▉| 369778/371472 [7:15:37<07:44,  3.65it/s]
100%|█████████▉| 369779/371472 [7:15:37<07:38,  3.69it/s]
100%|█████████▉| 369780/371472 [7:15:37<07:28,  3.77it/s]
                                                         
{'loss': 2.6119, 'learning_rate': 1.041015751255144e-07, 'epoch': 15.93}
+
100%|█████████▉| 369780/371472 [7:15:37<07:28,  3.77it/s]
100%|█████████▉| 369781/371472 [7:15:38<07:20,  3.84it/s]
100%|█████████▉| 369782/371472 [7:15:38<07:13,  3.90it/s]
100%|█████████▉| 369783/371472 [7:15:38<07:09,  3.93it/s]
100%|█████████▉| 369784/371472 [7:15:38<07:02,  3.99it/s]
100%|█████████▉| 369785/371472 [7:15:39<07:15,  3.87it/s]
100%|█████████▉| 369786/371472 [7:15:39<08:23,  3.35it/s]
100%|█████████▉| 369787/371472 [7:15:39<07:56,  3.54it/s]
100%|█████████▉| 369788/371472 [7:15:40<08:36,  3.26it/s]
100%|█████████▉| 369789/371472 [7:15:40<08:13,  3.41it/s]
100%|█████████▉| 369790/371472 [7:15:40<08:08,  3.44it/s]
100%|█████████▉| 369791/371472 [7:15:41<08:38,  3.24it/s]
100%|█████████▉| 369792/371472 [7:15:41<08:03,  3.47it/s]
100%|█████████▉| 369793/371472 [7:15:41<07:56,  3.53it/s]
100%|█████████▉| 369794/371472 [7:15:41<07:49,  3.58it/s]
100%|█████████▉| 369795/371472 [7:15:42<07:39,  3.65it/s]
100%|█████████▉| 369796/371472 [7:15:42<07:23,  3.78it/s]
100%|█████████▉| 369797/371472 [7:15:42<07:25,  3.76it/s]
100%|█████████▉| 369798/371472 [7:15:42<08:06,  3.44it/s]
100%|█████████▉| 369799/371472 [7:15:43<08:15,  3.38it/s]
100%|█████████▉| 369800/371472 [7:15:43<07:56,  3.51it/s]
                                                         
{'loss': 2.5815, 'learning_rate': 1.0405309315003556e-07, 'epoch': 15.93}
+
100%|█████████▉| 369800/371472 [7:15:43<07:56,  3.51it/s]
100%|█████████▉| 369801/371472 [7:15:43<07:56,  3.51it/s]
100%|█████████▉| 369802/371472 [7:15:44<08:06,  3.43it/s]
100%|█████████▉| 369803/371472 [7:15:44<08:37,  3.23it/s]
100%|█████████▉| 369804/371472 [7:15:44<08:31,  3.26it/s]
100%|█████████▉| 369805/371472 [7:15:44<08:08,  3.41it/s]
100%|█████████▉| 369806/371472 [7:15:45<07:58,  3.48it/s]
100%|█████████▉| 369807/371472 [7:15:45<08:02,  3.45it/s]
100%|█████████▉| 369808/371472 [7:15:45<07:45,  3.57it/s]
100%|█████████▉| 369809/371472 [7:15:46<07:42,  3.59it/s]
100%|█████████▉| 369810/371472 [7:15:46<07:43,  3.58it/s]
100%|█████████▉| 369811/371472 [7:15:46<07:39,  3.61it/s]
100%|█████████▉| 369812/371472 [7:15:46<07:48,  3.54it/s]
100%|█████████▉| 369813/371472 [7:15:47<08:07,  3.40it/s]
100%|█████████▉| 369814/371472 [7:15:47<07:45,  3.56it/s]
100%|█████████▉| 369815/371472 [7:15:47<07:41,  3.59it/s]
100%|█████████▉| 369816/371472 [7:15:48<08:11,  3.37it/s]
100%|█████████▉| 369817/371472 [7:15:48<08:10,  3.38it/s]
100%|█████████▉| 369818/371472 [7:15:48<07:57,  3.46it/s]
100%|█████████▉| 369819/371472 [7:15:48<08:04,  3.41it/s]
100%|█████████▉| 369820/371472 [7:15:49<07:41,  3.58it/s]
                                                         
{'loss': 2.6251, 'learning_rate': 1.040046111745566e-07, 'epoch': 15.93}
+
100%|█████████▉| 369820/371472 [7:15:49<07:41,  3.58it/s]
100%|█████████▉| 369821/371472 [7:15:49<08:05,  3.40it/s]
100%|█████████▉| 369822/371472 [7:15:49<07:48,  3.52it/s]
100%|█████████▉| 369823/371472 [7:15:50<07:58,  3.45it/s]
100%|█████████▉| 369824/371472 [7:15:50<09:11,  2.99it/s]
100%|█████████▉| 369825/371472 [7:15:50<08:35,  3.19it/s]
100%|█████████▉| 369826/371472 [7:15:51<08:11,  3.35it/s]
100%|█████████▉| 369827/371472 [7:15:51<07:57,  3.44it/s]
100%|█████████▉| 369828/371472 [7:15:51<07:41,  3.56it/s]
100%|█████████▉| 369829/371472 [7:15:51<07:41,  3.56it/s]
100%|█████████▉| 369830/371472 [7:15:52<08:09,  3.35it/s]
100%|█████████▉| 369831/371472 [7:15:52<08:09,  3.35it/s]
100%|█████████▉| 369832/371472 [7:15:52<08:07,  3.37it/s]
100%|█████████▉| 369833/371472 [7:15:53<07:46,  3.51it/s]
100%|█████████▉| 369834/371472 [7:15:53<07:38,  3.57it/s]
100%|█████████▉| 369835/371472 [7:15:53<07:25,  3.67it/s]
100%|█████████▉| 369836/371472 [7:15:53<07:36,  3.58it/s]
100%|█████████▉| 369837/371472 [7:15:54<08:09,  3.34it/s]
100%|█████████▉| 369838/371472 [7:15:54<07:45,  3.51it/s]
100%|█████████▉| 369839/371472 [7:15:54<07:42,  3.53it/s]
100%|█████████▉| 369840/371472 [7:15:55<07:27,  3.65it/s]
                                                         
{'loss': 2.6105, 'learning_rate': 1.0395612919907777e-07, 'epoch': 15.93}
+
100%|█████████▉| 369840/371472 [7:15:55<07:27,  3.65it/s]
100%|█████████▉| 369841/371472 [7:15:55<07:38,  3.56it/s]
100%|█████████▉| 369842/371472 [7:15:55<07:41,  3.53it/s]
100%|█████████▉| 369843/371472 [7:15:55<07:29,  3.63it/s]
100%|█████████���| 369844/371472 [7:15:56<07:35,  3.58it/s]
100%|█████████▉| 369845/371472 [7:15:56<07:42,  3.52it/s]
100%|█████████▉| 369846/371472 [7:15:56<07:47,  3.47it/s]
100%|█████████▉| 369847/371472 [7:15:57<07:46,  3.49it/s]
100%|█████████▉| 369848/371472 [7:15:57<07:39,  3.53it/s]
100%|█████████▉| 369849/371472 [7:15:57<07:17,  3.71it/s]
100%|█████████▉| 369850/371472 [7:15:57<07:48,  3.46it/s]
100%|█████████▉| 369851/371472 [7:15:58<07:36,  3.55it/s]
100%|█████████▉| 369852/371472 [7:15:58<07:34,  3.56it/s]
100%|█████████▉| 369853/371472 [7:15:58<07:57,  3.39it/s]
100%|█████████▉| 369854/371472 [7:15:59<07:38,  3.53it/s]
100%|█████████▉| 369855/371472 [7:15:59<07:30,  3.59it/s]
100%|█████████▉| 369856/371472 [7:15:59<07:30,  3.58it/s]
100%|█████████▉| 369857/371472 [7:15:59<07:23,  3.64it/s]
100%|█████████▉| 369858/371472 [7:16:00<07:29,  3.59it/s]
100%|█████████▉| 369859/371472 [7:16:00<07:15,  3.71it/s]
100%|█████████▉| 369860/371472 [7:16:00<07:49,  3.44it/s]
                                                         
{'loss': 2.556, 'learning_rate': 1.0390764722359883e-07, 'epoch': 15.93}
+
100%|█████████▉| 369860/371472 [7:16:00<07:49,  3.44it/s]
100%|█████████▉| 369861/371472 [7:16:01<07:56,  3.38it/s]
100%|█████████▉| 369862/371472 [7:16:01<07:40,  3.49it/s]
100%|█████████▉| 369863/371472 [7:16:01<07:18,  3.67it/s]
100%|█████████▉| 369864/371472 [7:16:01<08:26,  3.18it/s]
100%|█████████▉| 369865/371472 [7:16:02<08:07,  3.30it/s]
100%|█████████▉| 369866/371472 [7:16:02<07:51,  3.40it/s]
100%|█████████▉| 369867/371472 [7:16:02<07:38,  3.50it/s]
100%|█████████▉| 369868/371472 [7:16:03<07:35,  3.52it/s]
100%|█████████▉| 369869/371472 [7:16:03<07:26,  3.59it/s]
100%|█████████▉| 369870/371472 [7:16:03<07:15,  3.68it/s]
100%|█████████▉| 369871/371472 [7:16:03<07:13,  3.69it/s]
100%|█████████▉| 369872/371472 [7:16:04<07:28,  3.57it/s]
100%|█████████▉| 369873/371472 [7:16:04<07:14,  3.68it/s]
100%|█████████▉| 369874/371472 [7:16:04<08:02,  3.31it/s]
100%|█████████▉| 369875/371472 [7:16:05<07:52,  3.38it/s]
100%|█████████▉| 369876/371472 [7:16:05<07:54,  3.37it/s]
100%|█████████▉| 369877/371472 [7:16:05<08:12,  3.24it/s]
100%|█████████▉| 369878/371472 [7:16:05<07:53,  3.36it/s]
100%|█████████▉| 369879/371472 [7:16:06<07:23,  3.59it/s]
100%|█████████▉| 369880/371472 [7:16:06<07:16,  3.65it/s]
                                                         
{'loss': 2.6763, 'learning_rate': 1.0385916524811998e-07, 'epoch': 15.93}
+
100%|█████████▉| 369880/371472 [7:16:06<07:16,  3.65it/s]
100%|█████████▉| 369881/371472 [7:16:06<07:25,  3.57it/s]
100%|█████████▉| 369882/371472 [7:16:06<07:07,  3.72it/s]
100%|█████████▉| 369883/371472 [7:16:07<07:40,  3.45it/s]
100%|█████████▉| 369884/371472 [7:16:07<07:27,  3.55it/s]
100%|█████████▉| 369885/371472 [7:16:07<08:05,  3.27it/s]
100%|█████████▉| 369886/371472 [7:16:08<07:38,  3.46it/s]
100%|█████████▉| 369887/371472 [7:16:08<07:29,  3.52it/s]
100%|█████████▉| 369888/371472 [7:16:08<07:17,  3.62it/s]
100%|█████████▉| 369889/371472 [7:16:08<07:09,  3.68it/s]
100%|█████████▉| 369890/371472 [7:16:09<07:42,  3.42it/s]
100%|█████████▉| 369891/371472 [7:16:09<07:32,  3.49it/s]
100%|█████████▉| 369892/371472 [7:16:09<07:19,  3.60it/s]
100%|█████████▉| 369893/371472 [7:16:10<07:34,  3.47it/s]
100%|█████████▉| 369894/371472 [7:16:10<07:22,  3.57it/s]
100%|█████████▉| 369895/371472 [7:16:10<07:44,  3.39it/s]
100%|█████████▉| 369896/371472 [7:16:11<07:26,  3.53it/s]
100%|█████████▉| 369897/371472 [7:16:11<07:26,  3.53it/s]
100%|█████████▉| 369898/371472 [7:16:11<07:20,  3.57it/s]
100%|█████████▉| 369899/371472 [7:16:11<07:20,  3.57it/s]
100%|█████████▉| 369900/371472 [7:16:12<07:08,  3.67it/s]
                                                         
{'loss': 2.7512, 'learning_rate': 1.0381068327264104e-07, 'epoch': 15.93}
+
100%|█████████▉| 369900/371472 [7:16:12<07:08,  3.67it/s]
100%|█████████▉| 369901/371472 [7:16:12<06:55,  3.78it/s]
100%|█████████▉| 369902/371472 [7:16:12<06:55,  3.78it/s]
100%|█████████▉| 369903/371472 [7:16:12<06:57,  3.76it/s]
100%|█████████▉| 369904/371472 [7:16:13<07:03,  3.70it/s]
100%|█████████▉| 369905/371472 [7:16:13<07:03,  3.70it/s]
100%|█████████▉| 369906/371472 [7:16:13<07:03,  3.70it/s]
100%|█████████▉| 369907/371472 [7:16:13<06:50,  3.81it/s]
100%|█████████▉| 369908/371472 [7:16:14<07:11,  3.62it/s]
100%|█████████▉| 369909/371472 [7:16:14<06:55,  3.76it/s]
100%|█████████▉| 369910/371472 [7:16:14<06:53,  3.78it/s]
100%|█████████▉| 369911/371472 [7:16:15<06:55,  3.76it/s]
100%|█████████▉| 369912/371472 [7:16:15<06:52,  3.78it/s]
100%|█████████▉| 369913/371472 [7:16:15<06:59,  3.72it/s]
100%|█████████▉| 369914/371472 [7:16:15<07:12,  3.60it/s]
100%|█████████▉| 369915/371472 [7:16:16<07:13,  3.59it/s]
100%|█████████▉| 369916/371472 [7:16:16<06:57,  3.73it/s]
100%|█████████▉| 369917/371472 [7:16:16<07:17,  3.56it/s]
100%|█████████▉| 369918/371472 [7:16:16<07:08,  3.63it/s]
100%|█████████▉| 369919/371472 [7:16:17<07:08,  3.62it/s]
100%|█████████▉| 369920/371472 [7:16:17<07:36,  3.40it/s]
                                                         
{'loss': 2.5968, 'learning_rate': 1.037622012971622e-07, 'epoch': 15.93}
+
100%|█████████▉| 369920/371472 [7:16:17<07:36,  3.40it/s]
100%|█████████▉| 369921/371472 [7:16:17<08:12,  3.15it/s]
100%|█████████▉| 369922/371472 [7:16:18<08:03,  3.21it/s]
100%|█████████▉| 369923/371472 [7:16:18<07:32,  3.43it/s]
100%|█████████▉| 369924/371472 [7:16:18<07:11,  3.58it/s]
100%|█████████▉| 369925/371472 [7:16:19<07:14,  3.56it/s]
100%|█████████▉| 369926/371472 [7:16:19<07:14,  3.55it/s]
100%|█████████▉| 369927/371472 [7:16:19<06:56,  3.71it/s]
100%|█████████▉| 369928/371472 [7:16:19<07:22,  3.49it/s]
100%|█████████▉| 369929/371472 [7:16:20<07:28,  3.44it/s]
100%|█████████▉| 369930/371472 [7:16:20<07:56,  3.23it/s]
100%|█████████▉| 369931/371472 [7:16:20<07:52,  3.26it/s]
100%|█████████▉| 369932/371472 [7:16:21<07:44,  3.32it/s]
100%|█████████▉| 369933/371472 [7:16:21<07:34,  3.39it/s]
100%|█████████▉| 369934/371472 [7:16:21<07:20,  3.49it/s]
100%|█████████▉| 369935/371472 [7:16:21<07:21,  3.48it/s]
100%|█████████▉| 369936/371472 [7:16:22<07:16,  3.52it/s]
100%|█████████▉| 369937/371472 [7:16:22<07:09,  3.57it/s]
100%|█████████▉| 369938/371472 [7:16:22<07:10,  3.56it/s]
100%|█████████▉| 369939/371472 [7:16:23<08:17,  3.08it/s]
100%|█████████▉| 369940/371472 [7:16:23<08:00,  3.19it/s]
                                                         
{'loss': 2.6362, 'learning_rate': 1.0371371932168326e-07, 'epoch': 15.93}
+
100%|█████████▉| 369940/371472 [7:16:23<08:00,  3.19it/s]
100%|█████████▉| 369941/371472 [7:16:23<08:00,  3.18it/s]
100%|█████████▉| 369942/371472 [7:16:24<07:38,  3.34it/s]
100%|█████████▉| 369943/371472 [7:16:24<07:10,  3.55it/s]
100%|█████████▉| 369944/371472 [7:16:24<07:03,  3.61it/s]
100%|█████████▉| 369945/371472 [7:16:24<06:51,  3.71it/s]
100%|█████████▉| 369946/371472 [7:16:25<06:39,  3.82it/s]
100%|█████████▉| 369947/371472 [7:16:25<06:54,  3.68it/s]
100%|█████████▉| 369948/371472 [7:16:25<07:00,  3.62it/s]
100%|█████████▉| 369949/371472 [7:16:25<07:13,  3.52it/s]
100%|█████████▉| 369950/371472 [7:16:26<07:03,  3.60it/s]
100%|█████████▉| 369951/371472 [7:16:26<06:51,  3.70it/s]
100%|█████████▉| 369952/371472 [7:16:26<06:54,  3.67it/s]
100%|█████████▉| 369953/371472 [7:16:27<07:01,  3.61it/s]
100%|█████████▉| 369954/371472 [7:16:27<06:56,  3.65it/s]
100%|█████████▉| 369955/371472 [7:16:27<07:22,  3.43it/s]
100%|█████████▉| 369956/371472 [7:16:28<07:59,  3.16it/s]
100%|█████████▉| 369957/371472 [7:16:28<07:46,  3.25it/s]
100%|█████████▉| 369958/371472 [7:16:28<07:35,  3.32it/s]
100%|█████████▉| 369959/371472 [7:16:28<07:18,  3.45it/s]
100%|█████████▉| 369960/371472 [7:16:29<07:13,  3.49it/s]
                                                         
{'loss': 2.6194, 'learning_rate': 1.0366523734620441e-07, 'epoch': 15.93}
+
100%|█████████▉| 369960/371472 [7:16:29<07:13,  3.49it/s]
100%|█████████▉| 369961/371472 [7:16:29<07:18,  3.44it/s]
100%|█████████▉| 369962/371472 [7:16:29<07:27,  3.37it/s]
100%|█████████▉| 369963/371472 [7:16:30<07:12,  3.49it/s]
100%|█████████▉| 369964/371472 [7:16:30<07:03,  3.57it/s]
100%|█████████▉| 369965/371472 [7:16:30<07:01,  3.58it/s]
100%|█████████▉| 369966/371472 [7:16:30<06:55,  3.62it/s]
100%|█████████▉| 369967/371472 [7:16:31<07:02,  3.57it/s]
100%|█████████▉| 369968/371472 [7:16:31<06:56,  3.61it/s]
100%|█████████▉| 369969/371472 [7:16:31<06:54,  3.62it/s]
100%|█████████▉| 369970/371472 [7:16:31<06:57,  3.60it/s]
100%|█████████▉| 369971/371472 [7:16:32<06:50,  3.65it/s]
100%|█████████▉| 369972/371472 [7:16:32<06:45,  3.70it/s]
100%|█████████▉| 369973/371472 [7:16:32<06:58,  3.58it/s]
100%|█████████▉| 369974/371472 [7:16:33<06:57,  3.59it/s]
100%|█████████▉| 369975/371472 [7:16:33<06:54,  3.61it/s]
100%|█████████▉| 369976/371472 [7:16:33<06:42,  3.72it/s]
100%|█████████▉| 369977/371472 [7:16:33<07:31,  3.31it/s]
100%|█████████▉| 369978/371472 [7:16:34<07:08,  3.49it/s]
100%|█████████▉| 369979/371472 [7:16:34<06:57,  3.57it/s]
100%|█████████▉| 369980/371472 [7:16:34<07:20,  3.39it/s]
                                                         
{'loss': 2.6984, 'learning_rate': 1.0361675537072547e-07, 'epoch': 15.94}
+
100%|█████████▉| 369980/371472 [7:16:34<07:20,  3.39it/s]
100%|█████████▉| 369981/371472 [7:16:35<07:25,  3.35it/s]
100%|█████████▉| 369982/371472 [7:16:35<07:27,  3.33it/s]
100%|█████████▉| 369983/371472 [7:16:35<07:17,  3.40it/s]
100%|█████████▉| 369984/371472 [7:16:35<07:29,  3.31it/s]
100%|█████████▉| 369985/371472 [7:16:36<07:18,  3.39it/s]
100%|█████████▉| 369986/371472 [7:16:36<08:10,  3.03it/s]
100%|█████████▉| 369987/371472 [7:16:36<07:56,  3.12it/s]
100%|█████████▉| 369988/371472 [7:16:37<07:28,  3.31it/s]
100%|█████████▉| 369989/371472 [7:16:37<07:19,  3.38it/s]
100%|█████████▉| 369990/371472 [7:16:37<07:05,  3.48it/s]
100%|█████████▉| 369991/371472 [7:16:38<07:10,  3.44it/s]
100%|█████████▉| 369992/371472 [7:16:38<07:10,  3.44it/s]
100%|█████████▉| 369993/371472 [7:16:38<07:06,  3.47it/s]
100%|█████████▉| 369994/371472 [7:16:38<07:16,  3.39it/s]
100%|█████████▉| 369995/371472 [7:16:39<07:12,  3.41it/s]
100%|█████████▉| 369996/371472 [7:16:39<07:03,  3.49it/s]
100%|█████████▉| 369997/371472 [7:16:39<07:04,  3.47it/s]
100%|█████████▉| 369998/371472 [7:16:40<06:46,  3.62it/s]
100%|█████████▉| 369999/371472 [7:16:40<06:31,  3.76it/s]
100%|█████████▉| 370000/371472 [7:16:40<06:35,  3.72it/s]
                                                         
{'loss': 2.5292, 'learning_rate': 1.0356827339524663e-07, 'epoch': 15.94}
+
100%|█████████▉| 370000/371472 [7:16:40<06:35,  3.72it/s]
100%|█████████▉| 370001/371472 [7:16:40<06:47,  3.61it/s]
100%|█████████▉| 370002/371472 [7:16:41<07:36,  3.22it/s]
100%|█████████▉| 370003/371472 [7:16:41<07:14,  3.38it/s]
100%|█████████▉| 370004/371472 [7:16:41<06:54,  3.54it/s]
100%|█████████▉| 370005/371472 [7:16:42<06:37,  3.69it/s]
100%|█████████▉| 370006/371472 [7:16:42<06:36,  3.69it/s]
100%|█████████▉| 370007/371472 [7:16:42<06:32,  3.74it/s]
100%|█████████▉| 370008/371472 [7:16:42<06:36,  3.70it/s]
100%|█████████▉| 370009/371472 [7:16:43<06:23,  3.82it/s]
100%|█████████▉| 370010/371472 [7:16:43<06:20,  3.84it/s]
100%|█████████▉| 370011/371472 [7:16:43<06:42,  3.63it/s]
100%|█████████▉| 370012/371472 [7:16:43<06:39,  3.66it/s]
100%|█████████▉| 370013/371472 [7:16:44<06:37,  3.67it/s]
100%|█████████▉| 370014/371472 [7:16:44<06:30,  3.73it/s]
100%|█████████▉| 370015/371472 [7:16:44<06:30,  3.73it/s]
100%|█████████▉| 370016/371472 [7:16:44<06:28,  3.74it/s]
100%|█████████▉| 370017/371472 [7:16:45<06:30,  3.73it/s]
100%|█████████▉| 370018/371472 [7:16:45<07:00,  3.46it/s]
100%|█████████▉| 370019/371472 [7:16:45<07:11,  3.37it/s]
100%|█████████▉| 370020/371472 [7:16:46<07:06,  3.40it/s]
                                                         
{'loss': 2.7151, 'learning_rate': 1.0351979141976769e-07, 'epoch': 15.94}
+
100%|█████████▉| 370020/371472 [7:16:46<07:06,  3.40it/s]
100%|█████████▉| 370021/371472 [7:16:46<07:17,  3.32it/s]
100%|█████████▉| 370022/371472 [7:16:46<07:08,  3.38it/s]
100%|█████████▉| 370023/371472 [7:16:47<07:11,  3.36it/s]
100%|█████████▉| 370024/371472 [7:16:47<06:47,  3.55it/s]
100%|█████████▉| 370025/371472 [7:16:47<06:37,  3.64it/s]
100%|█████████▉| 370026/371472 [7:16:47<07:08,  3.37it/s]
100%|█████████▉| 370027/371472 [7:16:48<06:40,  3.60it/s]
100%|█████████▉| 370028/371472 [7:16:48<06:23,  3.77it/s]
100%|█████████▉| 370029/371472 [7:16:48<06:23,  3.77it/s]
100%|█████████▉| 370030/371472 [7:16:48<06:25,  3.74it/s]
100%|█████████▉| 370031/371472 [7:16:49<06:22,  3.76it/s]
100%|█████████▉| 370032/371472 [7:16:49<06:45,  3.55it/s]
100%|█████████▉| 370033/371472 [7:16:49<06:57,  3.45it/s]
100%|█████████▉| 370034/371472 [7:16:50<07:06,  3.37it/s]
100%|█████████▉| 370035/371472 [7:16:50<06:44,  3.55it/s]
100%|█████████▉| 370036/371472 [7:16:50<07:03,  3.39it/s]
100%|█████████▉| 370037/371472 [7:16:51<06:53,  3.47it/s]
100%|█████████▉| 370038/371472 [7:16:51<06:44,  3.54it/s]
100%|█████████▉| 370039/371472 [7:16:51<06:33,  3.64it/s]
100%|█████████▉| 370040/371472 [7:16:51<06:32,  3.65it/s]
                                                         
{'loss': 2.5154, 'learning_rate': 1.0347130944428883e-07, 'epoch': 15.94}
+
100%|█████████▉| 370040/371472 [7:16:51<06:32,  3.65it/s]
100%|█████████▉| 370041/371472 [7:16:52<06:35,  3.62it/s]
100%|█████████▉| 370042/371472 [7:16:52<06:41,  3.56it/s]
100%|█████████▉| 370043/371472 [7:16:52<08:31,  2.79it/s]
100%|█████████▉| 370044/371472 [7:16:53<07:46,  3.06it/s]
100%|█████████▉| 370045/371472 [7:16:53<07:39,  3.10it/s]
100%|█████████▉| 370046/371472 [7:16:53<07:09,  3.32it/s]
100%|█████████▉| 370047/371472 [7:16:53<06:53,  3.45it/s]
100%|█████████▉| 370048/371472 [7:16:54<06:37,  3.58it/s]
100%|█████████▉| 370049/371472 [7:16:54<06:51,  3.45it/s]
100%|█████████▉| 370050/371472 [7:16:54<07:03,  3.36it/s]
100%|█████████▉| 370051/371472 [7:16:55<06:49,  3.47it/s]
100%|█████████▉| 370052/371472 [7:16:55<07:20,  3.22it/s]
100%|█████████▉| 370053/371472 [7:16:55<07:00,  3.38it/s]
100%|█████████▉| 370054/371472 [7:16:56<07:06,  3.32it/s]
100%|█████████▉| 370055/371472 [7:16:56<06:51,  3.44it/s]
100%|█████████▉| 370056/371472 [7:16:56<06:41,  3.52it/s]
100%|█████████▉| 370057/371472 [7:16:56<07:00,  3.37it/s]
100%|█████████▉| 370058/371472 [7:16:57<06:40,  3.53it/s]
100%|█████████▉| 370059/371472 [7:16:57<06:30,  3.62it/s]
100%|█████████▉| 370060/371472 [7:16:57<06:24,  3.67it/s]
                                                         
{'loss': 2.5904, 'learning_rate': 1.034228274688099e-07, 'epoch': 15.94}
+
100%|█████████▉| 370060/371472 [7:16:57<06:24,  3.67it/s]
100%|█████████▉| 370061/371472 [7:16:58<06:29,  3.63it/s]
100%|█████████▉| 370062/371472 [7:16:58<06:16,  3.75it/s]
100%|█████████▉| 370063/371472 [7:16:58<06:19,  3.71it/s]
100%|█████████▉| 370064/371472 [7:16:58<06:16,  3.74it/s]
100%|█████████▉| 370065/371472 [7:16:59<06:17,  3.73it/s]
100%|█████████▉| 370066/371472 [7:16:59<06:14,  3.75it/s]
100%|█████████▉| 370067/371472 [7:16:59<06:10,  3.79it/s]
100%|█████████▉| 370068/371472 [7:16:59<06:20,  3.69it/s]
100%|█████████▉| 370069/371472 [7:17:00<06:29,  3.60it/s]
100%|█████████▉| 370070/371472 [7:17:00<06:17,  3.71it/s]
100%|█████████▉| 370071/371472 [7:17:00<06:30,  3.59it/s]
100%|█████████▉| 370072/371472 [7:17:00<06:28,  3.60it/s]
100%|█████████▉| 370073/371472 [7:17:01<06:22,  3.66it/s]
100%|█████████▉| 370074/371472 [7:17:01<06:25,  3.62it/s]
100%|█████████▉| 370075/371472 [7:17:01<06:26,  3.62it/s]
100%|█████████▉| 370076/371472 [7:17:02<06:53,  3.37it/s]
100%|█��███████▉| 370077/371472 [7:17:02<06:45,  3.44it/s]
100%|█████████▉| 370078/371472 [7:17:02<06:34,  3.54it/s]
100%|█████████▉| 370079/371472 [7:17:02<06:32,  3.55it/s]
100%|█████████▉| 370080/371472 [7:17:03<06:51,  3.38it/s]
                                                         
{'loss': 2.6093, 'learning_rate': 1.0337434549333106e-07, 'epoch': 15.94}
+
100%|█████████▉| 370080/371472 [7:17:03<06:51,  3.38it/s]
100%|█████████▉| 370081/371472 [7:17:03<06:25,  3.61it/s]
100%|█████████▉| 370082/371472 [7:17:03<06:30,  3.56it/s]
100%|█████████▉| 370083/371472 [7:17:04<06:39,  3.48it/s]
100%|█████████▉| 370084/371472 [7:17:04<06:24,  3.61it/s]
100%|█████████▉| 370085/371472 [7:17:04<07:04,  3.26it/s]
100%|█████████▉| 370086/371472 [7:17:05<07:21,  3.14it/s]
100%|█████████▉| 370087/371472 [7:17:05<06:56,  3.32it/s]
100%|█████████▉| 370088/371472 [7:17:05<07:24,  3.12it/s]
100%|█████████▉| 370089/371472 [7:17:05<07:02,  3.27it/s]
100%|█████████▉| 370090/371472 [7:17:06<06:48,  3.38it/s]
100%|█████████▉| 370091/371472 [7:17:06<06:40,  3.45it/s]
100%|█████████▉| 370092/371472 [7:17:06<06:52,  3.35it/s]
100%|█████████▉| 370093/371472 [7:17:07<06:46,  3.39it/s]
100%|█████████▉| 370094/371472 [7:17:07<06:43,  3.42it/s]
100%|█████████▉| 370095/371472 [7:17:07<06:30,  3.53it/s]
100%|█████████▉| 370096/371472 [7:17:07<06:27,  3.55it/s]
100%|█████████▉| 370097/371472 [7:17:08<06:38,  3.45it/s]
100%|█████████▉| 370098/371472 [7:17:08<06:26,  3.55it/s]
100%|█████████▉| 370099/371472 [7:17:08<06:38,  3.45it/s]
100%|█████████▉| 370100/371472 [7:17:09<06:24,  3.57it/s]
                                                         
{'loss': 2.531, 'learning_rate': 1.0332586351785212e-07, 'epoch': 15.94}
+
100%|█████████▉| 370100/371472 [7:17:09<06:24,  3.57it/s]
100%|█████████▉| 370101/371472 [7:17:09<06:15,  3.65it/s]
100%|█████████▉| 370102/371472 [7:17:09<06:05,  3.75it/s]
100%|█████████▉| 370103/371472 [7:17:09<06:04,  3.75it/s]
100%|█████████▉| 370104/371472 [7:17:10<06:04,  3.75it/s]
100%|█████████▉| 370105/371472 [7:17:10<06:11,  3.68it/s]
100%|█████████▉| 370106/371472 [7:17:10<06:27,  3.53it/s]
100%|█████████▉| 370107/371472 [7:17:11<06:26,  3.53it/s]
100%|█████████▉| 370108/371472 [7:17:11<06:18,  3.60it/s]
100%|█████████▉| 370109/371472 [7:17:11<06:22,  3.56it/s]
100%|█████████▉| 370110/371472 [7:17:11<06:23,  3.55it/s]
100%|█████████▉| 370111/371472 [7:17:12<06:31,  3.48it/s]
100%|█████████▉| 370112/371472 [7:17:12<06:35,  3.44it/s]
100%|█████████▉| 370113/371472 [7:17:12<06:13,  3.63it/s]
100%|█████████▉| 370114/371472 [7:17:12<06:02,  3.74it/s]
100%|█████████▉| 370115/371472 [7:17:13<06:03,  3.73it/s]
100%|█████████▉| 370116/371472 [7:17:13<05:52,  3.85it/s]
100%|█████████▉| 370117/371472 [7:17:13<06:01,  3.74it/s]
100%|█████████▉| 370118/371472 [7:17:14<06:13,  3.63it/s]
100%|█████████▉| 370119/371472 [7:17:14<06:03,  3.72it/s]
100%|█████████▉| 370120/371472 [7:17:14<06:36,  3.41it/s]
                                                         
{'loss': 2.5033, 'learning_rate': 1.0327738154237327e-07, 'epoch': 15.94}
+
100%|█████████▉| 370120/371472 [7:17:14<06:36,  3.41it/s]
100%|█████████▉| 370121/371472 [7:17:14<06:24,  3.51it/s]
100%|█████████▉| 370122/371472 [7:17:15<08:05,  2.78it/s]
100%|█████████▉| 370123/371472 [7:17:15<07:12,  3.12it/s]
100%|█████████▉| 370124/371472 [7:17:15<06:39,  3.37it/s]
100%|█████████▉| 370125/371472 [7:17:16<06:27,  3.48it/s]
100%|█████████▉| 370126/371472 [7:17:16<06:29,  3.45it/s]
100%|█████████▉| 370127/371472 [7:17:16<06:21,  3.53it/s]
100%|█████████▉| 370128/371472 [7:17:17<06:24,  3.49it/s]
100%|█████████▉| 370129/371472 [7:17:17<06:29,  3.45it/s]
100%|█████████▉| 370130/371472 [7:17:17<06:50,  3.27it/s]
100%|█████████▉| 370131/371472 [7:17:17<06:39,  3.35it/s]
100%|█████████▉| 370132/371472 [7:17:18<06:51,  3.26it/s]
100%|█████████▉| 370133/371472 [7:17:18<06:37,  3.37it/s]
100%|█████████▉| 370134/371472 [7:17:18<06:35,  3.38it/s]
100%|█████████▉| 370135/371472 [7:17:19<06:19,  3.52it/s]
100%|█████████▉| 370136/371472 [7:17:19<06:21,  3.50it/s]
100%|█████████▉| 370137/371472 [7:17:19<06:16,  3.55it/s]
100%|█████████▉| 370138/371472 [7:17:19<06:04,  3.66it/s]
100%|█████████▉| 370139/371472 [7:17:20<06:08,  3.61it/s]
100%|█████████▉| 370140/371472 [7:17:20<06:03,  3.66it/s]
                                                         
{'loss': 2.5202, 'learning_rate': 1.0322889956689433e-07, 'epoch': 15.94}
+
100%|█████████▉| 370140/371472 [7:17:20<06:03,  3.66it/s]
100%|█████████▉| 370141/371472 [7:17:20<06:06,  3.63it/s]
100%|█████████▉| 370142/371472 [7:17:21<06:06,  3.63it/s]
100%|█████████▉| 370143/371472 [7:17:21<06:04,  3.64it/s]
100%|█████████▉| 370144/371472 [7:17:21<05:52,  3.77it/s]
100%|█████████▉| 370145/371472 [7:17:21<06:02,  3.66it/s]
100%|█████████▉| 370146/371472 [7:17:22<06:33,  3.37it/s]
100%|█████████▉| 370147/371472 [7:17:22<07:05,  3.11it/s]
100%|█████████▉| 370148/371472 [7:17:22<06:48,  3.24it/s]
100%|█████████▉| 370149/371472 [7:17:23<06:33,  3.36it/s]
100%|█████████▉| 370150/371472 [7:17:23<06:49,  3.23it/s]
100%|█████████▉| 370151/371472 [7:17:23<06:33,  3.35it/s]
100%|█████████▉| 370152/371472 [7:17:24<06:26,  3.41it/s]
100%|█████████▉| 370153/371472 [7:17:24<06:38,  3.31it/s]
100%|█████████▉| 370154/371472 [7:17:24<07:09,  3.07it/s]
100%|█████████▉| 370155/371472 [7:17:24<06:49,  3.22it/s]
100%|█████████▉| 370156/371472 [7:17:25<06:33,  3.35it/s]
100%|█████████▉| 370157/371472 [7:17:25<06:29,  3.38it/s]
100%|█████████▉| 370158/371472 [7:17:25<06:55,  3.16it/s]
100%|█████████▉| 370159/371472 [7:17:26<07:07,  3.07it/s]
100%|█████████▉| 370160/371472 [7:17:26<07:14,  3.02it/s]
                                                         
{'loss': 2.6734, 'learning_rate': 1.031804175914155e-07, 'epoch': 15.94}
+
100%|█████████▉| 370160/371472 [7:17:26<07:14,  3.02it/s]
100%|█████████▉| 370161/371472 [7:17:26<07:18,  2.99it/s]
100%|█████████▉| 370162/371472 [7:17:27<07:05,  3.08it/s]
100%|█████████▉| 370163/371472 [7:17:27<06:37,  3.30it/s]
100%|█████████▉| 370164/371472 [7:17:27<06:18,  3.45it/s]
100%|█████████▉| 370165/371472 [7:17:28<06:01,  3.61it/s]
100%|█████████▉| 370166/371472 [7:17:28<05:52,  3.70it/s]
100%|█████████▉| 370167/371472 [7:17:28<05:38,  3.85it/s]
100%|█████████▉| 370168/371472 [7:17:28<05:57,  3.65it/s]
100%|█████████▉| 370169/371472 [7:17:29<05:51,  3.71it/s]
100%|█████████▉| 370170/371472 [7:17:29<05:57,  3.64it/s]
100%|█████████▉| 370171/371472 [7:17:29<06:16,  3.45it/s]
100%|█████████▉| 370172/371472 [7:17:29<06:22,  3.40it/s]
100%|█████████▉| 370173/371472 [7:17:30<06:31,  3.32it/s]
100%|█████████▉| 370174/371472 [7:17:30<06:52,  3.14it/s]
100%|█████████▉| 370175/371472 [7:17:30<06:42,  3.22it/s]
100%|█████████▉| 370176/371472 [7:17:31<06:31,  3.31it/s]
100%|█████████▉| 370177/371472 [7:17:31<06:12,  3.48it/s]
100%|█████████▉| 370178/371472 [7:17:31<06:15,  3.45it/s]
100%|█████████▉| 370179/371472 [7:17:32<06:29,  3.32it/s]
100%|█████████▉| 370180/371472 [7:17:32<06:15,  3.44it/s]
                                                         
{'loss': 2.4675, 'learning_rate': 1.0313193561593655e-07, 'epoch': 15.94}
+
100%|█████████▉| 370180/371472 [7:17:32<06:15,  3.44it/s]
100%|█████████▉| 370181/371472 [7:17:32<06:03,  3.55it/s]
100%|█████████▉| 370182/371472 [7:17:32<05:52,  3.66it/s]
100%|█████████▉| 370183/371472 [7:17:33<05:53,  3.64it/s]
100%|█████████▉| 370184/371472 [7:17:33<05:44,  3.74it/s]
100%|█████████▉| 370185/371472 [7:17:33<05:36,  3.83it/s]
100%|█████████▉| 370186/371472 [7:17:33<05:53,  3.64it/s]
100%|█████████▉| 370187/371472 [7:17:34<05:55,  3.61it/s]
100%|█████████▉| 370188/371472 [7:17:34<05:46,  3.70it/s]
100%|█████████▉| 370189/371472 [7:17:34<05:37,  3.80it/s]
100%|█████████▉| 370190/371472 [7:17:35<05:38,  3.79it/s]
100%|█████████▉| 370191/371472 [7:17:35<06:11,  3.45it/s]
100%|█��███████▉| 370192/371472 [7:17:35<06:15,  3.41it/s]
100%|█████████▉| 370193/371472 [7:17:35<06:04,  3.50it/s]
100%|█████████▉| 370194/371472 [7:17:36<07:06,  3.00it/s]
100%|█████████▉| 370195/371472 [7:17:36<06:57,  3.06it/s]
100%|█████████▉| 370196/371472 [7:17:36<06:24,  3.32it/s]
100%|█████████▉| 370197/371472 [7:17:37<06:02,  3.52it/s]
100%|█████████▉| 370198/371472 [7:17:37<05:55,  3.58it/s]
100%|█████████▉| 370199/371472 [7:17:37<05:53,  3.60it/s]
100%|█████████▉| 370200/371472 [7:17:38<05:52,  3.61it/s]
                                                         
{'loss': 2.4743, 'learning_rate': 1.030834536404577e-07, 'epoch': 15.95}
+
100%|█████████▉| 370200/371472 [7:17:38<05:52,  3.61it/s]
100%|█████████▉| 370201/371472 [7:17:38<06:04,  3.49it/s]
100%|█████████▉| 370202/371472 [7:17:38<06:10,  3.43it/s]
100%|█████████▉| 370203/371472 [7:17:38<06:15,  3.38it/s]
100%|█████████▉| 370204/371472 [7:17:39<06:14,  3.38it/s]
100%|█████████▉| 370205/371472 [7:17:39<06:13,  3.39it/s]
100%|█████████▉| 370206/371472 [7:17:39<06:03,  3.49it/s]
100%|█████████▉| 370207/371472 [7:17:40<05:57,  3.54it/s]
100%|█████████▉| 370208/371472 [7:17:40<05:44,  3.67it/s]
100%|█████████▉| 370209/371472 [7:17:40<05:44,  3.66it/s]
100%|█████████▉| 370210/371472 [7:17:40<06:01,  3.49it/s]
100%|█████████▉| 370211/371472 [7:17:41<05:53,  3.57it/s]
100%|█████████▉| 370212/371472 [7:17:41<05:43,  3.67it/s]
100%|█████████▉| 370213/371472 [7:17:41<06:04,  3.46it/s]
100%|█████████▉| 370214/371472 [7:17:42<05:56,  3.53it/s]
100%|█████████▉| 370215/371472 [7:17:42<05:47,  3.61it/s]
100%|█████████▉| 370216/371472 [7:17:42<05:52,  3.56it/s]
100%|█████████▉| 370217/371472 [7:17:42<06:28,  3.23it/s]
100%|█████████▉| 370218/371472 [7:17:43<06:07,  3.41it/s]
100%|█████████▉| 370219/371472 [7:17:43<05:47,  3.60it/s]
100%|█████████▉| 370220/371472 [7:17:43<05:39,  3.69it/s]
                                                         
{'loss': 2.5568, 'learning_rate': 1.0303497166497876e-07, 'epoch': 15.95}
+
100%|█████████▉| 370220/371472 [7:17:43<05:39,  3.69it/s]
100%|█████████▉| 370221/371472 [7:17:44<06:04,  3.43it/s]
100%|█████████▉| 370222/371472 [7:17:44<06:07,  3.40it/s]
100%|█████████▉| 370223/371472 [7:17:44<05:52,  3.54it/s]
100%|█████████▉| 370224/371472 [7:17:44<05:58,  3.48it/s]
100%|█████████▉| 370225/371472 [7:17:45<05:58,  3.48it/s]
100%|█████████▉| 370226/371472 [7:17:45<05:59,  3.47it/s]
100%|█████████▉| 370227/371472 [7:17:45<06:04,  3.42it/s]
100%|█████████▉| 370228/371472 [7:17:46<05:59,  3.46it/s]
100%|█████████▉| 370229/371472 [7:17:46<05:46,  3.59it/s]
100%|█████████▉| 370230/371472 [7:17:46<06:20,  3.26it/s]
100%|█████████▉| 370231/371472 [7:17:46<06:24,  3.22it/s]
100%|█████████▉| 370232/371472 [7:17:47<06:19,  3.27it/s]
100%|█████████▉| 370233/371472 [7:17:47<06:19,  3.27it/s]
100%|█████████▉| 370234/371472 [7:17:47<06:08,  3.36it/s]
100%|█████████▉| 370235/371472 [7:17:48<06:00,  3.43it/s]
100%|█████████▉| 370236/371472 [7:17:48<06:01,  3.42it/s]
100%|█████████▉| 370237/371472 [7:17:48<05:54,  3.48it/s]
100%|█████████▉| 370238/371472 [7:17:49<06:19,  3.25it/s]
100%|█████████▉| 370239/371472 [7:17:49<06:10,  3.33it/s]
100%|█████████▉| 370240/371472 [7:17:49<05:59,  3.42it/s]
                                                         
{'loss': 2.5494, 'learning_rate': 1.0298648968949993e-07, 'epoch': 15.95}
+
100%|█████████▉| 370240/371472 [7:17:49<05:59,  3.42it/s]
100%|█████████▉| 370241/371472 [7:17:49<05:54,  3.47it/s]
100%|█████████▉| 370242/371472 [7:17:50<05:47,  3.54it/s]
100%|█████████▉| 370243/371472 [7:17:50<05:42,  3.59it/s]
100%|█████████▉| 370244/371472 [7:17:50<05:30,  3.71it/s]
100%|█████████▉| 370245/371472 [7:17:50<05:19,  3.84it/s]
100%|█████████▉| 370246/371472 [7:17:51<05:25,  3.76it/s]
100%|█████████▉| 370247/371472 [7:17:51<05:32,  3.69it/s]
100%|█████████▉| 370248/371472 [7:17:51<05:39,  3.61it/s]
100%|█████████▉| 370249/371472 [7:17:52<05:32,  3.68it/s]
100%|█████████▉| 370250/371472 [7:17:52<05:35,  3.64it/s]
100%|█████████▉| 370251/371472 [7:17:52<05:43,  3.56it/s]
100%|█████████▉| 370252/371472 [7:17:52<05:38,  3.61it/s]
100%|█████████▉| 370253/371472 [7:17:53<05:31,  3.67it/s]
100%|█████████▉| 370254/371472 [7:17:53<05:25,  3.74it/s]
100%|█████████▉| 370255/371472 [7:17:53<05:15,  3.85it/s]
100%|█████████▉| 370256/371472 [7:17:53<05:26,  3.72it/s]
100%|█████████▉| 370257/371472 [7:17:54<05:33,  3.64it/s]
100%|█████████▉| 370258/371472 [7:17:54<05:23,  3.75it/s]
100%|█████████▉| 370259/371472 [7:17:54<05:39,  3.58it/s]
100%|█████████▉| 370260/371472 [7:17:55<05:55,  3.41it/s]
                                                         
{'loss': 2.5455, 'learning_rate': 1.0293800771402099e-07, 'epoch': 15.95}
+
100%|█████████▉| 370260/371472 [7:17:55<05:55,  3.41it/s]
100%|█████████▉| 370261/371472 [7:17:55<05:59,  3.37it/s]
100%|█████████▉| 370262/371472 [7:17:55<05:51,  3.45it/s]
100%|█████████▉| 370263/371472 [7:17:55<05:45,  3.50it/s]
100%|█████████▉| 370264/371472 [7:17:56<05:33,  3.62it/s]
100%|█████████▉| 370265/371472 [7:17:56<06:02,  3.33it/s]
100%|█████████▉| 370266/371472 [7:17:56<05:51,  3.43it/s]
100%|█████████▉| 370267/371472 [7:17:57<05:54,  3.40it/s]
100%|█████████▉| 370268/371472 [7:17:57<05:43,  3.51it/s]
100%|█████████▉| 370269/371472 [7:17:57<05:38,  3.55it/s]
100%|█████████▉| 370270/371472 [7:17:57<05:44,  3.49it/s]
100%|█████████▉| 370271/371472 [7:17:58<05:49,  3.43it/s]
100%|█████████▉| 370272/371472 [7:17:58<05:39,  3.54it/s]
100%|█████████▉| 370273/371472 [7:17:58<05:28,  3.65it/s]
100%|█████████▉| 370274/371472 [7:17:59<05:37,  3.55it/s]
100%|█████████▉| 370275/371472 [7:17:59<05:24,  3.69it/s]
100%|█████████▉| 370276/371472 [7:17:59<05:20,  3.73it/s]
100%|█████████▉| 370277/371472 [7:17:59<05:13,  3.81it/s]
100%|█████████▉| 370278/371472 [7:18:00<05:13,  3.81it/s]
100%|█████████▉| 370279/371472 [7:18:00<05:12,  3.82it/s]
100%|█████████▉| 370280/371472 [7:18:00<05:22,  3.69it/s]
                                                         
{'loss': 2.5731, 'learning_rate': 1.0288952573854214e-07, 'epoch': 15.95}
+
100%|█████████▉| 370280/371472 [7:18:00<05:22,  3.69it/s]
100%|█████████▉| 370281/371472 [7:18:01<05:53,  3.37it/s]
100%|█████████▉| 370282/371472 [7:18:01<05:57,  3.33it/s]
100%|█████████▉| 370283/371472 [7:18:01<06:11,  3.20it/s]
100%|█████████▉| 370284/371472 [7:18:01<05:53,  3.36it/s]
100%|█████████▉| 370285/371472 [7:18:02<05:46,  3.42it/s]
100%|█████████▉| 370286/371472 [7:18:02<05:36,  3.53it/s]
100%|█████████▉| 370287/371472 [7:18:02<05:32,  3.56it/s]
100%|█████████▉| 370288/371472 [7:18:03<05:25,  3.63it/s]
100%|█████████▉| 370289/371472 [7:18:03<05:17,  3.72it/s]
100%|█████████▉| 370290/371472 [7:18:03<05:34,  3.53it/s]
100%|█████████▉| 370291/371472 [7:18:03<05:27,  3.61it/s]
100%|█████████▉| 370292/371472 [7:18:04<05:30,  3.57it/s]
100%|█████████▉| 370293/371472 [7:18:04<05:48,  3.39it/s]
100%|█████████▉| 370294/371472 [7:18:04<05:37,  3.49it/s]
100%|█████████▉| 370295/371472 [7:18:05<05:30,  3.56it/s]
100%|█████████▉| 370296/371472 [7:18:05<05:25,  3.61it/s]
100%|█████████▉| 370297/371472 [7:18:05<05:24,  3.63it/s]
100%|█████████▉| 370298/371472 [7:18:05<05:30,  3.55it/s]
100%|█████████▉| 370299/371472 [7:18:06<05:27,  3.58it/s]
100%|█████████▉| 370300/371472 [7:18:06<05:40,  3.45it/s]
                                                         
{'loss': 2.4017, 'learning_rate': 1.028410437630632e-07, 'epoch': 15.95}
+
100%|█████████▉| 370300/371472 [7:18:06<05:40,  3.45it/s]
100%|█████████▉| 370301/371472 [7:18:06<05:39,  3.45it/s]
100%|█████████▉| 370302/371472 [7:18:06<05:24,  3.60it/s]
100%|█████████▉| 370303/371472 [7:18:07<05:15,  3.71it/s]
100%|█████████▉| 370304/371472 [7:18:07<05:15,  3.71it/s]
100%|█████████▉| 370305/371472 [7:18:07<05:04,  3.83it/s]
100%|█████████▉| 370306/371472 [7:18:07<05:05,  3.82it/s]
100%|█��███████▉| 370307/371472 [7:18:08<05:06,  3.80it/s]
100%|█████████▉| 370308/371472 [7:18:08<05:02,  3.85it/s]
100%|█████████▉| 370309/371472 [7:18:08<05:13,  3.71it/s]
100%|█████████▉| 370310/371472 [7:18:09<05:14,  3.69it/s]
100%|█████████▉| 370311/371472 [7:18:09<05:16,  3.66it/s]
100%|█████████▉| 370312/371472 [7:18:09<05:13,  3.70it/s]
100%|█████████▉| 370313/371472 [7:18:09<05:11,  3.72it/s]
100%|█████████▉| 370314/371472 [7:18:10<05:08,  3.75it/s]
100%|█████████▉| 370315/371472 [7:18:10<05:00,  3.84it/s]
100%|█████████▉| 370316/371472 [7:18:10<05:05,  3.78it/s]
100%|█████████▉| 370317/371472 [7:18:10<05:11,  3.71it/s]
100%|█████████▉| 370318/371472 [7:18:11<05:11,  3.71it/s]
100%|█████████▉| 370319/371472 [7:18:11<05:16,  3.64it/s]
100%|█████████▉| 370320/371472 [7:18:11<05:28,  3.51it/s]
                                                         
{'loss': 2.6731, 'learning_rate': 1.0279256178758425e-07, 'epoch': 15.95}
+
100%|█████████▉| 370320/371472 [7:18:11<05:28,  3.51it/s]
100%|█████████▉| 370321/371472 [7:18:12<05:30,  3.49it/s]
100%|█████████▉| 370322/371472 [7:18:12<05:25,  3.53it/s]
100%|█████████▉| 370323/371472 [7:18:12<05:09,  3.71it/s]
100%|█████████▉| 370324/371472 [7:18:12<05:28,  3.50it/s]
100%|█████████▉| 370325/371472 [7:18:13<05:30,  3.47it/s]
100%|█████████▉| 370326/371472 [7:18:13<05:42,  3.35it/s]
100%|█████████▉| 370327/371472 [7:18:13<05:28,  3.49it/s]
100%|█████████▉| 370328/371472 [7:18:14<05:25,  3.52it/s]
100%|█████████▉| 370329/371472 [7:18:14<05:18,  3.59it/s]
100%|█████████▉| 370330/371472 [7:18:14<05:11,  3.67it/s]
100%|█████████▉| 370331/371472 [7:18:15<06:00,  3.17it/s]
100%|█████████▉| 370332/371472 [7:18:15<05:55,  3.20it/s]
100%|█████████▉| 370333/371472 [7:18:15<05:38,  3.37it/s]
100%|█████████▉| 370334/371472 [7:18:15<05:26,  3.48it/s]
100%|█████████▉| 370335/371472 [7:18:16<05:34,  3.40it/s]
100%|█████████▉| 370336/371472 [7:18:16<05:44,  3.30it/s]
100%|█████████▉| 370337/371472 [7:18:16<05:50,  3.23it/s]
100%|█████████▉| 370338/371472 [7:18:17<05:30,  3.43it/s]
100%|█████████▉| 370339/371472 [7:18:17<05:23,  3.50it/s]
100%|█████████▉| 370340/371472 [7:18:17<05:14,  3.60it/s]
                                                         
{'loss': 2.6703, 'learning_rate': 1.0274407981210542e-07, 'epoch': 15.95}
+
100%|█████████▉| 370340/371472 [7:18:17<05:14,  3.60it/s]
100%|█████████▉| 370341/371472 [7:18:17<05:27,  3.45it/s]
100%|█████████▉| 370342/371472 [7:18:18<05:18,  3.55it/s]
100%|█████████▉| 370343/371472 [7:18:18<05:10,  3.63it/s]
100%|█████████▉| 370344/371472 [7:18:18<05:25,  3.47it/s]
100%|█████████▉| 370345/371472 [7:18:19<05:14,  3.59it/s]
100%|█████████▉| 370346/371472 [7:18:19<05:11,  3.61it/s]
100%|█████████▉| 370347/371472 [7:18:19<05:04,  3.70it/s]
100%|█████████▉| 370348/371472 [7:18:19<05:13,  3.58it/s]
100%|█████████▉| 370349/371472 [7:18:20<05:12,  3.60it/s]
100%|█████████▉| 370350/371472 [7:18:20<05:01,  3.72it/s]
100%|█████████▉| 370351/371472 [7:18:20<04:59,  3.74it/s]
100%|█████████▉| 370352/371472 [7:18:21<05:46,  3.24it/s]
100%|█████████▉| 370353/371472 [7:18:21<05:33,  3.35it/s]
100%|█████████▉| 370354/371472 [7:18:21<05:30,  3.38it/s]
100%|█████████▉| 370355/371472 [7:18:21<05:25,  3.44it/s]
100%|█████████▉| 370356/371472 [7:18:22<05:22,  3.46it/s]
100%|█████████▉| 370357/371472 [7:18:22<05:13,  3.56it/s]
100%|█████████▉| 370358/371472 [7:18:22<05:37,  3.30it/s]
100%|█████████▉| 370359/371472 [7:18:23<05:37,  3.30it/s]
100%|█████████▉| 370360/371472 [7:18:23<05:43,  3.23it/s]
                                                         
{'loss': 2.5806, 'learning_rate': 1.0269559783662647e-07, 'epoch': 15.95}
+
100%|█████████▉| 370360/371472 [7:18:23<05:43,  3.23it/s]
100%|█████████▉| 370361/371472 [7:18:23<05:28,  3.38it/s]
100%|█████████▉| 370362/371472 [7:18:23<05:28,  3.38it/s]
100%|█████████▉| 370363/371472 [7:18:24<05:20,  3.47it/s]
100%|█████████▉| 370364/371472 [7:18:24<05:11,  3.55it/s]
100%|█████████▉| 370365/371472 [7:18:24<05:01,  3.68it/s]
100%|█████████▉| 370366/371472 [7:18:25<04:58,  3.70it/s]
100%|█████████▉| 370367/371472 [7:18:25<05:11,  3.55it/s]
100%|█████████▉| 370368/371472 [7:18:25<05:03,  3.63it/s]
100%|█████████▉| 370369/371472 [7:18:25<05:10,  3.55it/s]
100%|█████████▉| 370370/371472 [7:18:26<05:01,  3.66it/s]
100%|█████████▉| 370371/371472 [7:18:26<04:51,  3.78it/s]
100%|█████████▉| 370372/371472 [7:18:26<05:20,  3.44it/s]
100%|█████████▉| 370373/371472 [7:18:27<05:57,  3.08it/s]
100%|█████████▉| 370374/371472 [7:18:27<05:40,  3.22it/s]
100%|█████████▉| 370375/371472 [7:18:27<05:23,  3.39it/s]
100%|█████████▉| 370376/371472 [7:18:27<05:12,  3.51it/s]
100%|█████████▉| 370377/371472 [7:18:28<05:03,  3.61it/s]
100%|█████████▉| 370378/371472 [7:18:28<04:58,  3.67it/s]
100%|█████████▉| 370379/371472 [7:18:28<05:10,  3.52it/s]
100%|█████████▉| 370380/371472 [7:18:29<05:12,  3.49it/s]
                                                         
{'loss': 2.567, 'learning_rate': 1.0264711586114763e-07, 'epoch': 15.95}
+
100%|█████████▉| 370380/371472 [7:18:29<05:12,  3.49it/s]
100%|█████████▉| 370381/371472 [7:18:29<05:05,  3.57it/s]
100%|█████████▉| 370382/371472 [7:18:29<05:08,  3.53it/s]
100%|█████████▉| 370383/371472 [7:18:29<05:00,  3.63it/s]
100%|█████████▉| 370384/371472 [7:18:30<05:03,  3.59it/s]
100%|█████████▉| 370385/371472 [7:18:30<05:03,  3.58it/s]
100%|█████████▉| 370386/371472 [7:18:30<05:01,  3.60it/s]
100%|█████████▉| 370387/371472 [7:18:30<04:54,  3.68it/s]
100%|█████████▉| 370388/371472 [7:18:31<05:00,  3.61it/s]
100%|█████████▉| 370389/371472 [7:18:31<05:08,  3.51it/s]
100%|█████████▉| 370390/371472 [7:18:31<05:24,  3.33it/s]
100%|█████████▉| 370391/371472 [7:18:32<05:23,  3.35it/s]
100%|█████████▉| 370392/371472 [7:18:32<05:35,  3.22it/s]
100%|█████████▉| 370393/371472 [7:18:32<05:22,  3.35it/s]
100%|█████████▉| 370394/371472 [7:18:33<05:11,  3.46it/s]
100%|█████████▉| 370395/371472 [7:18:33<05:25,  3.31it/s]
100%|█████████▉| 370396/371472 [7:18:33<05:21,  3.35it/s]
100%|█████████▉| 370397/371472 [7:18:34<05:21,  3.34it/s]
100%|█████████▉| 370398/371472 [7:18:34<05:15,  3.40it/s]
100%|█████████▉| 370399/371472 [7:18:34<05:20,  3.34it/s]
100%|█████████▉| 370400/371472 [7:18:34<05:14,  3.41it/s]
                                                         
{'loss': 2.6239, 'learning_rate': 1.0259863388566868e-07, 'epoch': 15.95}
+
100%|█████████▉| 370400/371472 [7:18:34<05:14,  3.41it/s]
100%|█████████▉| 370401/371472 [7:18:35<05:05,  3.50it/s]
100%|█████████▉| 370402/371472 [7:18:35<04:58,  3.59it/s]
100%|█████████▉| 370403/371472 [7:18:35<05:02,  3.54it/s]
100%|█████████▉| 370404/371472 [7:18:35<05:08,  3.46it/s]
100%|█████████▉| 370405/371472 [7:18:36<05:01,  3.54it/s]
100%|█████████▉| 370406/371472 [7:18:36<05:11,  3.43it/s]
100%|█████████▉| 370407/371472 [7:18:36<05:08,  3.45it/s]
100%|█████████▉| 370408/371472 [7:18:37<05:08,  3.45it/s]
100%|█████████▉| 370409/371472 [7:18:37<05:04,  3.49it/s]
100%|█████████▉| 370410/371472 [7:18:37<05:01,  3.52it/s]
100%|█████████▉| 370411/371472 [7:18:38<05:16,  3.36it/s]
100%|█████████▉| 370412/371472 [7:18:38<05:09,  3.43it/s]
100%|█████████▉| 370413/371472 [7:18:38<05:55,  2.98it/s]
100%|█████████▉| 370414/371472 [7:18:39<05:44,  3.07it/s]
100%|█████████▉| 370415/371472 [7:18:39<05:24,  3.25it/s]
100%|█████████▉| 370416/371472 [7:18:39<05:24,  3.25it/s]
100%|█████████▉| 370417/371472 [7:18:39<05:26,  3.23it/s]
100%|█████████▉| 370418/371472 [7:18:40<05:12,  3.37it/s]
100%|█████████▉| 370419/371472 [7:18:40<04:57,  3.54it/s]
100%|█████████▉| 370420/371472 [7:18:40<04:51,  3.61it/s]
                                                         
{'loss': 2.5695, 'learning_rate': 1.0255015191018984e-07, 'epoch': 15.95}
+
100%|█████████▉| 370420/371472 [7:18:40<04:51,  3.61it/s]
100%|█████████▉| 370421/371472 [7:18:41<04:57,  3.53it/s]
100%|█���███████▉| 370422/371472 [7:18:41<04:53,  3.58it/s]
100%|█████████▉| 370423/371472 [7:18:41<05:04,  3.45it/s]
100%|█████████▉| 370424/371472 [7:18:41<05:32,  3.16it/s]
100%|█████████▉| 370425/371472 [7:18:42<05:22,  3.25it/s]
100%|█████████▉| 370426/371472 [7:18:42<04:57,  3.52it/s]
100%|█████████▉| 370427/371472 [7:18:42<05:22,  3.25it/s]
100%|█████████▉| 370428/371472 [7:18:43<05:21,  3.25it/s]
100%|█████████▉| 370429/371472 [7:18:43<05:09,  3.37it/s]
100%|█████████▉| 370430/371472 [7:18:43<05:03,  3.44it/s]
100%|█████████▉| 370431/371472 [7:18:43<04:52,  3.56it/s]
100%|█████████▉| 370432/371472 [7:18:44<04:43,  3.67it/s]
100%|█████████▉| 370433/371472 [7:18:44<04:47,  3.62it/s]
100%|█████████▉| 370434/371472 [7:18:44<04:48,  3.59it/s]
100%|█████████▉| 370435/371472 [7:18:45<04:40,  3.69it/s]
100%|█████████▉| 370436/371472 [7:18:45<04:40,  3.70it/s]
100%|█████████▉| 370437/371472 [7:18:45<04:37,  3.73it/s]
100%|█████████▉| 370438/371472 [7:18:45<04:35,  3.75it/s]
100%|█████████▉| 370439/371472 [7:18:46<04:43,  3.65it/s]
100%|█████████▉| 370440/371472 [7:18:46<04:50,  3.55it/s]
                                                         
{'loss': 2.5061, 'learning_rate': 1.0250166993471089e-07, 'epoch': 15.96}
+
100%|█████████▉| 370440/371472 [7:18:46<04:50,  3.55it/s]
100%|█████████▉| 370441/371472 [7:18:46<04:49,  3.56it/s]
100%|█████████▉| 370442/371472 [7:18:46<04:42,  3.65it/s]
100%|█████████▉| 370443/371472 [7:18:47<04:42,  3.65it/s]
100%|█████████▉| 370444/371472 [7:18:47<05:01,  3.41it/s]
100%|█████████▉| 370445/371472 [7:18:47<04:49,  3.55it/s]
100%|█████████▉| 370446/371472 [7:18:48<04:41,  3.64it/s]
100%|█████████▉| 370447/371472 [7:18:48<05:00,  3.41it/s]
100%|█████████▉| 370448/371472 [7:18:48<04:55,  3.47it/s]
100%|█████████▉| 370449/371472 [7:18:48<04:46,  3.57it/s]
100%|█████████▉| 370450/371472 [7:18:49<05:40,  3.00it/s]
100%|█████████▉| 370451/371472 [7:18:49<05:15,  3.23it/s]
100%|█████████▉| 370452/371472 [7:18:49<05:11,  3.28it/s]
100%|█████████▉| 370453/371472 [7:18:50<04:52,  3.49it/s]
100%|█████████▉| 370454/371472 [7:18:50<05:10,  3.28it/s]
100%|█████████▉| 370455/371472 [7:18:50<05:14,  3.24it/s]
100%|█████████▉| 370456/371472 [7:18:51<05:01,  3.37it/s]
100%|█████████▉| 370457/371472 [7:18:51<04:47,  3.53it/s]
100%|█████████▉| 370458/371472 [7:18:51<04:51,  3.48it/s]
100%|█████████▉| 370459/371472 [7:18:52<04:54,  3.44it/s]
100%|█████████▉| 370460/371472 [7:18:52<04:42,  3.59it/s]
                                                         
{'loss': 2.6981, 'learning_rate': 1.0245318795923206e-07, 'epoch': 15.96}
+
100%|█████████▉| 370460/371472 [7:18:52<04:42,  3.59it/s]
100%|█████████▉| 370461/371472 [7:18:52<04:56,  3.41it/s]
100%|█████████▉| 370462/371472 [7:18:52<04:54,  3.43it/s]
100%|█████████▉| 370463/371472 [7:18:53<04:49,  3.48it/s]
100%|█████████▉| 370464/371472 [7:18:53<04:49,  3.48it/s]
100%|█████████▉| 370465/371472 [7:18:53<04:41,  3.57it/s]
100%|█████████▉| 370466/371472 [7:18:53<04:36,  3.64it/s]
100%|█████████▉| 370467/371472 [7:18:54<04:54,  3.41it/s]
100%|█████████▉| 370468/371472 [7:18:54<05:02,  3.32it/s]
100%|█████████▉| 370469/371472 [7:18:54<05:12,  3.21it/s]
100%|█████████▉| 370470/371472 [7:18:55<05:00,  3.34it/s]
100%|█████████▉| 370471/371472 [7:18:55<04:57,  3.36it/s]
100%|█████████▉| 370472/371472 [7:18:55<04:53,  3.41it/s]
100%|█████████▉| 370473/371472 [7:18:56<04:50,  3.44it/s]
100%|█████████▉| 370474/371472 [7:18:56<04:37,  3.60it/s]
100%|█████████▉| 370475/371472 [7:18:56<04:40,  3.56it/s]
100%|█████████▉| 370476/371472 [7:18:56<04:34,  3.63it/s]
100%|█████████▉| 370477/371472 [7:18:57<04:30,  3.68it/s]
100%|█████████▉| 370478/371472 [7:18:57<04:23,  3.78it/s]
100%|█████████▉| 370479/371472 [7:18:57<04:44,  3.49it/s]
100%|█████████▉| 370480/371472 [7:18:58<04:56,  3.34it/s]
                                                         
{'loss': 2.5289, 'learning_rate': 1.0240470598375312e-07, 'epoch': 15.96}
+
100%|█████████▉| 370480/371472 [7:18:58<04:56,  3.34it/s]
100%|█████████▉| 370481/371472 [7:18:58<04:42,  3.51it/s]
100%|█████████▉| 370482/371472 [7:18:58<04:32,  3.63it/s]
100%|█████████▉| 370483/371472 [7:18:58<04:38,  3.55it/s]
100%|█████████▉| 370484/371472 [7:18:59<05:06,  3.22it/s]
100%|█████████▉| 370485/371472 [7:18:59<04:59,  3.30it/s]
100%|█████████▉| 370486/371472 [7:18:59<04:48,  3.42it/s]
100%|█████████▉| 370487/371472 [7:19:00<04:54,  3.34it/s]
100%|█████████▉| 370488/371472 [7:19:00<04:51,  3.38it/s]
100%|█████████▉| 370489/371472 [7:19:00<04:42,  3.48it/s]
100%|█████████▉| 370490/371472 [7:19:00<04:39,  3.52it/s]
100%|█████████▉| 370491/371472 [7:19:01<04:41,  3.49it/s]
100%|█████████▉| 370492/371472 [7:19:01<04:32,  3.60it/s]
100%|█████████▉| 370493/371472 [7:19:01<04:33,  3.58it/s]
100%|█████████▉| 370494/371472 [7:19:02<04:58,  3.27it/s]
100%|█████████▉| 370495/371472 [7:19:02<05:13,  3.12it/s]
100%|█████████▉| 370496/371472 [7:19:02<04:57,  3.28it/s]
100%|█████████▉| 370497/371472 [7:19:03<04:45,  3.41it/s]
100%|█████████▉| 370498/371472 [7:19:03<04:39,  3.49it/s]
100%|█████████▉| 370499/371472 [7:19:03<04:45,  3.41it/s]
100%|█████████▉| 370500/371472 [7:19:04<06:26,  2.52it/s]
                                                         
{'loss': 2.4728, 'learning_rate': 1.0235622400827427e-07, 'epoch': 15.96}
+
100%|█████████▉| 370500/371472 [7:19:04<06:26,  2.52it/s]
100%|█████████▉| 370501/371472 [7:19:04<06:05,  2.66it/s]
100%|█████████▉| 370502/371472 [7:19:04<05:33,  2.91it/s]
100%|█████████▉| 370503/371472 [7:19:05<05:04,  3.18it/s]
100%|█████████▉| 370504/371472 [7:19:05<05:01,  3.21it/s]
100%|█████████▉| 370505/371472 [7:19:05<05:10,  3.12it/s]
100%|█████████▉| 370506/371472 [7:19:06<04:54,  3.28it/s]
100%|█████████▉| 370507/371472 [7:19:06<04:58,  3.23it/s]
100%|█████████▉| 370508/371472 [7:19:06<05:09,  3.12it/s]
100%|█████████▉| 370509/371472 [7:19:06<04:57,  3.24it/s]
100%|█████████▉| 370510/371472 [7:19:07<05:04,  3.16it/s]
100%|█████████▉| 370511/371472 [7:19:07<05:11,  3.08it/s]
100%|█████████▉| 370512/371472 [7:19:07<04:59,  3.21it/s]
100%|█████████▉| 370513/371472 [7:19:08<04:48,  3.33it/s]
100%|█████████▉| 370514/371472 [7:19:08<04:35,  3.48it/s]
100%|█████████▉| 370515/371472 [7:19:08<04:26,  3.59it/s]
100%|█████████▉| 370516/371472 [7:19:08<04:22,  3.65it/s]
100%|█████████▉| 370517/371472 [7:19:09<04:23,  3.63it/s]
100%|█████████▉| 370518/371472 [7:19:09<04:15,  3.74it/s]
100%|█████████▉| 370519/371472 [7:19:09<04:20,  3.66it/s]
100%|█████████▉| 370520/371472 [7:19:10<04:16,  3.71it/s]
                                                         
{'loss': 2.468, 'learning_rate': 1.0230774203279532e-07, 'epoch': 15.96}
+
100%|█████████▉| 370520/371472 [7:19:10<04:16,  3.71it/s]
100%|█████████▉| 370521/371472 [7:19:10<04:16,  3.70it/s]
100%|█████████▉| 370522/371472 [7:19:10<04:05,  3.88it/s]
100%|█████████▉| 370523/371472 [7:19:10<04:02,  3.92it/s]
100%|█████████▉| 370524/371472 [7:19:11<04:11,  3.77it/s]
100%|█████████▉| 370525/371472 [7:19:11<04:13,  3.73it/s]
100%|█████████▉| 370526/371472 [7:19:11<04:10,  3.78it/s]
100%|█████████▉| 370527/371472 [7:19:11<04:14,  3.71it/s]
100%|█████████▉| 370528/371472 [7:19:12<04:19,  3.64it/s]
100%|█████████▉| 370529/371472 [7:19:12<04:11,  3.74it/s]
100%|█████████▉| 370530/371472 [7:19:12<04:18,  3.64it/s]
100%|█████████▉| 370531/371472 [7:19:13<04:25,  3.55it/s]
100%|█████████▉| 370532/371472 [7:19:13<04:30,  3.48it/s]
100%|█████████▉| 370533/371472 [7:19:13<04:18,  3.63it/s]
100%|█████████▉| 370534/371472 [7:19:13<04:15,  3.67it/s]
100%|█████████▉| 370535/371472 [7:19:14<04:25,  3.53it/s]
100%|█████████▉| 370536/371472 [7:19:14<04:36,  3.38it/s]
100%|█████████▉| 370537/371472 [7:19:14<04:22,  3.57it/s]
100%|█████████▉| 370538/371472 [7:19:15<04:25,  3.52it/s]
100%|█████████▉| 370539/371472 [7:19:15<04:18,  3.60it/s]
100%|█████████▉| 370540/371472 [7:19:15<04:16,  3.63it/s]
                                                         
{'loss': 2.5743, 'learning_rate': 1.0225926005731649e-07, 'epoch': 15.96}
+
100%|█████████▉| 370540/371472 [7:19:15<04:16,  3.63it/s]
100%|█████████▉| 370541/371472 [7:19:15<04:13,  3.68it/s]
100%|█████████▉| 370542/371472 [7:19:16<04:30,  3.43it/s]
100%|█████████▉| 370543/371472 [7:19:16<04:28,  3.46it/s]
100%|█████████▉| 370544/371472 [7:19:16<04:32,  3.40it/s]
100%|█████████▉| 370545/371472 [7:19:16<04:22,  3.54it/s]
100%|█████████▉| 370546/371472 [7:19:17<04:22,  3.53it/s]
100%|█████████▉| 370547/371472 [7:19:17<04:13,  3.65it/s]
100%|█████████▉| 370548/371472 [7:19:17<04:07,  3.73it/s]
100%|█████████▉| 370549/371472 [7:19:18<04:09,  3.70it/s]
100%|█████████▉| 370550/371472 [7:19:18<04:02,  3.80it/s]
100%|█████████▉| 370551/371472 [7:19:18<04:15,  3.61it/s]
100%|█████████▉| 370552/371472 [7:19:18<04:19,  3.54it/s]
100%|█████████▉| 370553/371472 [7:19:19<04:06,  3.73it/s]
100%|█████████▉| 370554/371472 [7:19:19<04:06,  3.72it/s]
100%|█████████▉| 370555/371472 [7:19:19<04:19,  3.53it/s]
100%|█████████▉| 370556/371472 [7:19:20<04:17,  3.56it/s]
100%|█████████▉| 370557/371472 [7:19:20<04:12,  3.62it/s]
100%|█████████▉| 370558/371472 [7:19:20<04:19,  3.52it/s]
100%|█████████▉| 370559/371472 [7:19:20<04:17,  3.54it/s]
100%|█████████▉| 370560/371472 [7:19:21<04:13,  3.60it/s]
                                                         
{'loss': 2.6394, 'learning_rate': 1.0221077808183755e-07, 'epoch': 15.96}
+
100%|█████████▉| 370560/371472 [7:19:21<04:13,  3.60it/s]
100%|█████████▉| 370561/371472 [7:19:21<04:09,  3.65it/s]
100%|█████████▉| 370562/371472 [7:19:21<04:26,  3.41it/s]
100%|█████████▉| 370563/371472 [7:19:22<04:31,  3.35it/s]
100%|█████████▉| 370564/371472 [7:19:22<04:22,  3.46it/s]
100%|█████████▉| 370565/371472 [7:19:22<04:21,  3.46it/s]
100%|█████████▉| 370566/371472 [7:19:22<04:13,  3.58it/s]
100%|█████████▉| 370567/371472 [7:19:23<04:07,  3.65it/s]
100%|█████████▉| 370568/371472 [7:19:23<04:08,  3.63it/s]
100%|█████████▉| 370569/371472 [7:19:23<04:06,  3.66it/s]
100%|█████████▉| 370570/371472 [7:19:23<04:00,  3.75it/s]
100%|█████████▉| 370571/371472 [7:19:24<04:22,  3.43it/s]
100%|█████████▉| 370572/371472 [7:19:24<04:17,  3.49it/s]
100%|█████████▉| 370573/371472 [7:19:24<04:06,  3.65it/s]
100%|█████████▉| 370574/371472 [7:19:25<04:12,  3.56it/s]
100%|█████████▉| 370575/371472 [7:19:25<04:09,  3.59it/s]
100%|█████████▉| 370576/371472 [7:19:25<04:10,  3.58it/s]
100%|█████████▉| 370577/371472 [7:19:25<04:29,  3.33it/s]
100%|█████████▉| 370578/371472 [7:19:26<04:44,  3.15it/s]
100%|█████████▉| 370579/371472 [7:19:26<04:33,  3.27it/s]
100%|█████████▉| 370580/371472 [7:19:26<04:40,  3.19it/s]
                                                         
{'loss': 2.6088, 'learning_rate': 1.021622961063587e-07, 'epoch': 15.96}
+
100%|█████████▉| 370580/371472 [7:19:26<04:40,  3.19it/s]
100%|█████████▉| 370581/371472 [7:19:27<04:26,  3.34it/s]
100%|█████████▉| 370582/371472 [7:19:27<04:18,  3.45it/s]
100%|█████████▉| 370583/371472 [7:19:27<04:11,  3.54it/s]
100%|█████████▉| 370584/371472 [7:19:28<04:06,  3.60it/s]
100%|█████████▉| 370585/371472 [7:19:28<04:00,  3.69it/s]
100%|█████████▉| 370586/371472 [7:19:28<03:57,  3.73it/s]
100%|█████████▉| 370587/371472 [7:19:28<04:02,  3.66it/s]
100%|█████████▉| 370588/371472 [7:19:29<03:59,  3.70it/s]
100%|█████████▉| 370589/371472 [7:19:29<04:07,  3.57it/s]
100%|█████████▉| 370590/371472 [7:19:29<04:01,  3.65it/s]
100%|█████████▉| 370591/371472 [7:19:29<04:08,  3.54it/s]
100%|█████████▉| 370592/371472 [7:19:30<04:08,  3.55it/s]
100%|█████████▉| 370593/371472 [7:19:30<04:03,  3.61it/s]
100%|█████████▉| 370594/371472 [7:19:30<04:01,  3.63it/s]
100%|█████████▉| 370595/371472 [7:19:31<03:58,  3.68it/s]
100%|█████████▉| 370596/371472 [7:19:31<04:04,  3.58it/s]
100%|██████��██▉| 370597/371472 [7:19:31<04:17,  3.39it/s]
100%|█████████▉| 370598/371472 [7:19:31<04:10,  3.49it/s]
100%|█████████▉| 370599/371472 [7:19:32<04:11,  3.47it/s]
100%|█████████▉| 370600/371472 [7:19:32<04:16,  3.40it/s]
                                                         
{'loss': 2.6182, 'learning_rate': 1.0211381413087976e-07, 'epoch': 15.96}
+
100%|█████████▉| 370600/371472 [7:19:32<04:16,  3.40it/s]
100%|█████████▉| 370601/371472 [7:19:32<04:21,  3.33it/s]
100%|█████████▉| 370602/371472 [7:19:33<04:16,  3.39it/s]
100%|█████████▉| 370603/371472 [7:19:33<04:09,  3.49it/s]
100%|█████████▉| 370604/371472 [7:19:33<04:17,  3.37it/s]
100%|█████████▉| 370605/371472 [7:19:33<04:09,  3.47it/s]
100%|█████████▉| 370606/371472 [7:19:34<04:18,  3.35it/s]
100%|█████████▉| 370607/371472 [7:19:34<04:10,  3.46it/s]
100%|█████████▉| 370608/371472 [7:19:34<04:07,  3.49it/s]
100%|█████████▉| 370609/371472 [7:19:35<04:08,  3.47it/s]
100%|█████████▉| 370610/371472 [7:19:35<04:01,  3.57it/s]
100%|█████████▉| 370611/371472 [7:19:35<04:25,  3.24it/s]
100%|█████████▉| 370612/371472 [7:19:36<04:19,  3.32it/s]
100%|█████████▉| 370613/371472 [7:19:36<04:05,  3.49it/s]
100%|█████████▉| 370614/371472 [7:19:36<04:06,  3.48it/s]
100%|█████████▉| 370615/371472 [7:19:36<03:58,  3.59it/s]
100%|█████████▉| 370616/371472 [7:19:37<03:59,  3.58it/s]
100%|█████████▉| 370617/371472 [7:19:37<03:54,  3.64it/s]
100%|█████████▉| 370618/371472 [7:19:37<04:05,  3.47it/s]
100%|█████████▉| 370619/371472 [7:19:37<03:54,  3.63it/s]
100%|█████████▉| 370620/371472 [7:19:38<03:45,  3.78it/s]
                                                         
{'loss': 2.6647, 'learning_rate': 1.0206533215540092e-07, 'epoch': 15.96}
+
100%|█████████▉| 370620/371472 [7:19:38<03:45,  3.78it/s]
100%|█████████▉| 370621/371472 [7:19:38<03:55,  3.61it/s]
100%|█████████▉| 370622/371472 [7:19:38<03:51,  3.68it/s]
100%|█████████▉| 370623/371472 [7:19:39<03:49,  3.70it/s]
100%|█████████▉| 370624/371472 [7:19:39<04:02,  3.50it/s]
100%|█████████▉| 370625/371472 [7:19:39<04:00,  3.52it/s]
100%|█████████▉| 370626/371472 [7:19:39<04:08,  3.40it/s]
100%|█████████▉| 370627/371472 [7:19:40<04:10,  3.38it/s]
100%|█████████▉| 370628/371472 [7:19:40<04:19,  3.25it/s]
100%|█████████▉| 370629/371472 [7:19:40<04:01,  3.49it/s]
100%|█████████▉| 370630/371472 [7:19:41<03:57,  3.55it/s]
100%|█████████▉| 370631/371472 [7:19:41<03:49,  3.66it/s]
100%|█████████▉| 370632/371472 [7:19:41<03:46,  3.70it/s]
100%|█████████▉| 370633/371472 [7:19:41<03:46,  3.70it/s]
100%|█████████▉| 370634/371472 [7:19:42<03:41,  3.78it/s]
100%|█████████▉| 370635/371472 [7:19:42<03:37,  3.86it/s]
100%|█████████▉| 370636/371472 [7:19:42<03:52,  3.59it/s]
100%|█████████▉| 370637/371472 [7:19:43<03:58,  3.50it/s]
100%|█████████▉| 370638/371472 [7:19:43<03:56,  3.52it/s]
100%|█████████▉| 370639/371472 [7:19:43<03:57,  3.51it/s]
100%|█████████▉| 370640/371472 [7:19:43<03:53,  3.57it/s]
                                                         
{'loss': 2.633, 'learning_rate': 1.0201685017992198e-07, 'epoch': 15.96}
+
100%|█████████▉| 370640/371472 [7:19:43<03:53,  3.57it/s]
100%|█████████▉| 370641/371472 [7:19:44<03:44,  3.70it/s]
100%|█████████▉| 370642/371472 [7:19:44<03:44,  3.70it/s]
100%|█████████▉| 370643/371472 [7:19:44<03:41,  3.75it/s]
100%|█████████▉| 370644/371472 [7:19:44<03:39,  3.78it/s]
100%|█████████▉| 370645/371472 [7:19:45<03:38,  3.79it/s]
100%|█████████▉| 370646/371472 [7:19:45<03:56,  3.50it/s]
100%|█████████▉| 370647/371472 [7:19:45<04:00,  3.43it/s]
100%|█████████▉| 370648/371472 [7:19:46<04:01,  3.41it/s]
100%|█████████▉| 370649/371472 [7:19:46<03:51,  3.55it/s]
100%|█████████▉| 370650/371472 [7:19:46<03:43,  3.68it/s]
100%|█████████▉| 370651/371472 [7:19:46<03:43,  3.68it/s]
100%|█████████▉| 370652/371472 [7:19:47<04:01,  3.39it/s]
100%|█████████▉| 370653/371472 [7:19:47<04:02,  3.37it/s]
100%|█████████▉| 370654/371472 [7:19:47<03:58,  3.43it/s]
100%|█████████▉| 370655/371472 [7:19:48<03:56,  3.45it/s]
100%|█████████▉| 370656/371472 [7:19:48<03:45,  3.62it/s]
100%|█████████▉| 370657/371472 [7:19:48<03:39,  3.72it/s]
100%|█████████▉| 370658/371472 [7:19:48<03:42,  3.65it/s]
100%|█████████▉| 370659/371472 [7:19:49<03:40,  3.69it/s]
100%|█████████▉| 370660/371472 [7:19:49<03:58,  3.40it/s]
                                                         
{'loss': 2.5665, 'learning_rate': 1.0196836820444313e-07, 'epoch': 15.97}
+
100%|█████████▉| 370660/371472 [7:19:49<03:58,  3.40it/s]
100%|█████████▉| 370661/371472 [7:19:49<03:45,  3.60it/s]
100%|█████████▉| 370662/371472 [7:19:49<03:42,  3.65it/s]
100%|█████████▉| 370663/371472 [7:19:50<03:36,  3.74it/s]
100%|█████████▉| 370664/371472 [7:19:50<03:53,  3.46it/s]
100%|█████████▉| 370665/371472 [7:19:50<04:05,  3.29it/s]
100%|█████████▉| 370666/371472 [7:19:51<03:56,  3.41it/s]
100%|█████████▉| 370667/371472 [7:19:51<03:47,  3.53it/s]
100%|█████████▉| 370668/371472 [7:19:51<03:51,  3.47it/s]
100%|█████████▉| 370669/371472 [7:19:51<03:40,  3.64it/s]
100%|█████████▉| 370670/371472 [7:19:52<03:56,  3.39it/s]
100%|█████████▉| 370671/371472 [7:19:52<03:48,  3.51it/s]
100%|█████████▉| 370672/371472 [7:19:52<03:46,  3.53it/s]
100%|█████████▉| 370673/371472 [7:19:53<03:42,  3.60it/s]
100%|█████████▉| 370674/371472 [7:19:53<03:35,  3.70it/s]
100%|█████████▉| 370675/371472 [7:19:53<03:37,  3.66it/s]
100%|█████████▉| 370676/371472 [7:19:53<03:49,  3.46it/s]
100%|█████████▉| 370677/371472 [7:19:54<03:41,  3.60it/s]
100%|█████████▉| 370678/371472 [7:19:54<03:41,  3.59it/s]
100%|█████████▉| 370679/371472 [7:19:54<03:34,  3.70it/s]
100%|█████████▉| 370680/371472 [7:19:55<03:44,  3.53it/s]
                                                         
{'loss': 2.7128, 'learning_rate': 1.0191988622896419e-07, 'epoch': 15.97}
+
100%|█████████▉| 370680/371472 [7:19:55<03:44,  3.53it/s]
100%|█████████▉| 370681/371472 [7:19:55<03:36,  3.66it/s]
100%|█████████▉| 370682/371472 [7:19:55<03:29,  3.77it/s]
100%|█████████▉| 370683/371472 [7:19:55<03:43,  3.53it/s]
100%|█████████▉| 370684/371472 [7:19:56<03:48,  3.45it/s]
100%|█████████▉| 370685/371472 [7:19:56<03:44,  3.51it/s]
100%|█████████▉| 370686/371472 [7:19:56<03:50,  3.40it/s]
100%|█████████▉| 370687/371472 [7:19:57<03:49,  3.41it/s]
100%|█████████▉| 370688/371472 [7:19:57<03:49,  3.42it/s]
100%|█████████▉| 370689/371472 [7:19:57<03:48,  3.43it/s]
100%|█████████▉| 370690/371472 [7:19:57<03:51,  3.38it/s]
100%|█████████▉| 370691/371472 [7:19:58<03:45,  3.46it/s]
100%|█████████▉| 370692/371472 [7:19:58<04:02,  3.22it/s]
100%|█████████▉| 370693/371472 [7:19:58<04:02,  3.21it/s]
100%|█████████▉| 370694/371472 [7:19:59<04:06,  3.15it/s]
100%|█████████▉| 370695/371472 [7:19:59<03:49,  3.38it/s]
100%|█████████▉| 370696/371472 [7:19:59<03:38,  3.55it/s]
100%|█████████▉| 370697/371472 [7:20:00<03:40,  3.51it/s]
100%|█████████▉| 370698/371472 [7:20:00<03:43,  3.46it/s]
100%|█████████▉| 370699/371472 [7:20:00<03:44,  3.44it/s]
100%|█████████▉| 370700/371472 [7:20:00<03:42,  3.48it/s]
                                                         
{'loss': 2.3996, 'learning_rate': 1.0187140425348535e-07, 'epoch': 15.97}
+
100%|█████████▉| 370700/371472 [7:20:00<03:42,  3.48it/s]
100%|█████████▉| 370701/371472 [7:20:01<03:41,  3.48it/s]
100%|█████████▉| 370702/371472 [7:20:01<03:51,  3.33it/s]
100%|█████████▉| 370703/371472 [7:20:01<04:04,  3.15it/s]
100%|█████████▉| 370704/371472 [7:20:02<03:49,  3.34it/s]
100%|█████████▉| 370705/371472 [7:20:02<03:42,  3.45it/s]
100%|█████████▉| 370706/371472 [7:20:02<03:39,  3.49it/s]
100%|█████████▉| 370707/371472 [7:20:02<03:33,  3.59it/s]
100%|█████████▉| 370708/371472 [7:20:03<03:32,  3.60it/s]
100%|█████████▉| 370709/371472 [7:20:03<03:30,  3.63it/s]
100%|█████████▉| 370710/371472 [7:20:03<03:22,  3.77it/s]
100%|█████████▉| 370711/371472 [7:20:03<03:21,  3.79it/s]
100%|██████���██▉| 370712/371472 [7:20:04<03:21,  3.77it/s]
100%|█████████▉| 370713/371472 [7:20:04<03:19,  3.80it/s]
100%|█████████▉| 370714/371472 [7:20:04<03:16,  3.85it/s]
100%|█████████▉| 370715/371472 [7:20:05<03:26,  3.67it/s]
100%|█████████▉| 370716/371472 [7:20:05<03:44,  3.37it/s]
100%|█████████▉| 370717/371472 [7:20:05<03:50,  3.27it/s]
100%|█████████▉| 370718/371472 [7:20:06<03:43,  3.38it/s]
100%|█████████▉| 370719/371472 [7:20:06<03:53,  3.23it/s]
100%|█████████▉| 370720/371472 [7:20:06<03:55,  3.20it/s]
                                                         
{'loss': 2.4959, 'learning_rate': 1.0182292227800641e-07, 'epoch': 15.97}
+
100%|█████████▉| 370720/371472 [7:20:06<03:55,  3.20it/s]
100%|█████████▉| 370721/371472 [7:20:06<03:41,  3.38it/s]
100%|█████████▉| 370722/371472 [7:20:07<03:35,  3.48it/s]
100%|█████████▉| 370723/371472 [7:20:07<03:45,  3.31it/s]
100%|█████████▉| 370724/371472 [7:20:07<03:41,  3.38it/s]
100%|█████████▉| 370725/371472 [7:20:08<03:36,  3.44it/s]
100%|█████████▉| 370726/371472 [7:20:08<03:34,  3.48it/s]
100%|█████████▉| 370727/371472 [7:20:08<03:27,  3.60it/s]
100%|█████████▉| 370728/371472 [7:20:08<03:30,  3.53it/s]
100%|█████████▉| 370729/371472 [7:20:09<03:32,  3.50it/s]
100%|█████████▉| 370730/371472 [7:20:09<03:29,  3.54it/s]
100%|█████████▉| 370731/371472 [7:20:09<03:22,  3.66it/s]
100%|█████████▉| 370732/371472 [7:20:10<03:22,  3.66it/s]
100%|█████████▉| 370733/371472 [7:20:10<03:27,  3.57it/s]
100%|█████████▉| 370734/371472 [7:20:10<03:20,  3.68it/s]
100%|█████████▉| 370735/371472 [7:20:10<03:19,  3.70it/s]
100%|█████████▉| 370736/371472 [7:20:11<03:11,  3.84it/s]
100%|█████████▉| 370737/371472 [7:20:11<03:15,  3.76it/s]
100%|█████████▉| 370738/371472 [7:20:11<03:12,  3.81it/s]
100%|█████████▉| 370739/371472 [7:20:11<03:30,  3.48it/s]
100%|█████████▉| 370740/371472 [7:20:12<03:24,  3.59it/s]
                                                         
{'loss': 2.5477, 'learning_rate': 1.0177444030252756e-07, 'epoch': 15.97}
+
100%|█████████▉| 370740/371472 [7:20:12<03:24,  3.59it/s]
100%|█████████▉| 370741/371472 [7:20:12<03:38,  3.34it/s]
100%|█████████▉| 370742/371472 [7:20:12<03:35,  3.39it/s]
100%|█████████▉| 370743/371472 [7:20:13<03:28,  3.50it/s]
100%|█████████▉| 370744/371472 [7:20:13<03:28,  3.49it/s]
100%|█████████▉| 370745/371472 [7:20:13<03:28,  3.48it/s]
100%|█████████▉| 370746/371472 [7:20:13<03:24,  3.56it/s]
100%|█████████▉| 370747/371472 [7:20:14<03:19,  3.63it/s]
100%|█████████▉| 370748/371472 [7:20:14<03:16,  3.69it/s]
100%|█████████▉| 370749/371472 [7:20:14<03:17,  3.67it/s]
100%|█████████▉| 370750/371472 [7:20:15<03:22,  3.57it/s]
100%|█████████▉| 370751/371472 [7:20:15<03:21,  3.58it/s]
100%|█████████▉| 370752/371472 [7:20:15<03:17,  3.64it/s]
100%|█████████▉| 370753/371472 [7:20:15<03:17,  3.64it/s]
100%|█████████▉| 370754/371472 [7:20:16<03:21,  3.56it/s]
100%|█████████▉| 370755/371472 [7:20:16<03:14,  3.69it/s]
100%|█████████▉| 370756/371472 [7:20:16<03:23,  3.51it/s]
100%|█████████▉| 370757/371472 [7:20:16<03:18,  3.61it/s]
100%|█████████▉| 370758/371472 [7:20:17<03:15,  3.65it/s]
100%|█████████▉| 370759/371472 [7:20:17<03:12,  3.70it/s]
100%|█████████▉| 370760/371472 [7:20:17<03:33,  3.33it/s]
                                                         
{'loss': 2.6031, 'learning_rate': 1.0172595832704862e-07, 'epoch': 15.97}
+
100%|█████████▉| 370760/371472 [7:20:17<03:33,  3.33it/s]
100%|█████████▉| 370761/371472 [7:20:18<03:20,  3.55it/s]
100%|█████████▉| 370762/371472 [7:20:18<03:18,  3.57it/s]
100%|█████████▉| 370763/371472 [7:20:18<03:11,  3.70it/s]
100%|█████████▉| 370764/371472 [7:20:18<03:09,  3.73it/s]
100%|█████████▉| 370765/371472 [7:20:19<03:14,  3.64it/s]
100%|█████████▉| 370766/371472 [7:20:19<03:10,  3.71it/s]
100%|█████████▉| 370767/371472 [7:20:19<03:18,  3.56it/s]
100%|█████████▉| 370768/371472 [7:20:20<03:10,  3.70it/s]
100%|█████████▉| 370769/371472 [7:20:20<03:05,  3.79it/s]
100%|█████████▉| 370770/371472 [7:20:20<03:07,  3.75it/s]
100%|█████████▉| 370771/371472 [7:20:20<03:20,  3.50it/s]
100%|█████████▉| 370772/371472 [7:20:21<03:17,  3.55it/s]
100%|█████████▉| 370773/371472 [7:20:21<03:13,  3.61it/s]
100%|█████████▉| 370774/371472 [7:20:21<03:20,  3.49it/s]
100%|█████████▉| 370775/371472 [7:20:21<03:16,  3.55it/s]
100%|█████████▉| 370776/371472 [7:20:22<03:16,  3.54it/s]
100%|█████████▉| 370777/371472 [7:20:22<03:18,  3.51it/s]
100%|█████████▉| 370778/371472 [7:20:22<03:14,  3.56it/s]
100%|█████████▉| 370779/371472 [7:20:23<03:10,  3.64it/s]
100%|█████████▉| 370780/371472 [7:20:23<03:12,  3.60it/s]
                                                         
{'loss': 2.7387, 'learning_rate': 1.0167747635156978e-07, 'epoch': 15.97}
+
100%|█████████▉| 370780/371472 [7:20:23<03:12,  3.60it/s]
100%|█████████▉| 370781/371472 [7:20:23<03:10,  3.63it/s]
100%|█████████▉| 370782/371472 [7:20:23<03:17,  3.50it/s]
100%|█████████▉| 370783/371472 [7:20:24<03:12,  3.58it/s]
100%|█████████▉| 370784/371472 [7:20:24<03:12,  3.58it/s]
100%|█████████▉| 370785/371472 [7:20:24<03:06,  3.68it/s]
100%|█████████▉| 370786/371472 [7:20:25<03:12,  3.57it/s]
100%|█████████▉| 370787/371472 [7:20:25<03:11,  3.58it/s]
100%|█████████▉| 370788/371472 [7:20:25<03:06,  3.67it/s]
100%|█████████▉| 370789/371472 [7:20:25<03:21,  3.39it/s]
100%|█████████▉| 370790/371472 [7:20:26<03:14,  3.50it/s]
100%|█████████▉| 370791/371472 [7:20:26<03:08,  3.60it/s]
100%|█████████▉| 370792/371472 [7:20:26<03:12,  3.53it/s]
100%|█████████▉| 370793/371472 [7:20:27<03:19,  3.40it/s]
100%|█████████▉| 370794/371472 [7:20:27<03:17,  3.44it/s]
100%|█████████▉| 370795/371472 [7:20:27<03:26,  3.28it/s]
100%|█████████▉| 370796/371472 [7:20:28<03:31,  3.20it/s]
100%|█████████▉| 370797/371472 [7:20:28<03:39,  3.07it/s]
100%|█████████▉| 370798/371472 [7:20:28<03:28,  3.24it/s]
100%|█████████▉| 370799/371472 [7:20:28<03:33,  3.15it/s]
100%|█████████▉| 370800/371472 [7:20:29<03:37,  3.10it/s]
                                                         
{'loss': 2.502, 'learning_rate': 1.0162899437609086e-07, 'epoch': 15.97}
+
100%|█████████▉| 370800/371472 [7:20:29<03:37,  3.10it/s]
100%|█████████▉| 370801/371472 [7:20:29<03:47,  2.95it/s]
100%|█████████▉| 370802/371472 [7:20:29<03:34,  3.12it/s]
100%|█████████▉| 370803/371472 [7:20:30<03:21,  3.32it/s]
100%|█████████▉| 370804/371472 [7:20:30<03:14,  3.43it/s]
100%|█████████▉| 370805/371472 [7:20:30<03:20,  3.32it/s]
100%|█████████▉| 370806/371472 [7:20:31<03:11,  3.48it/s]
100%|█████████▉| 370807/371472 [7:20:31<03:15,  3.41it/s]
100%|█████████▉| 370808/371472 [7:20:31<03:19,  3.33it/s]
100%|█████████▉| 370809/371472 [7:20:32<03:22,  3.27it/s]
100%|█████████▉| 370810/371472 [7:20:32<03:15,  3.38it/s]
100%|█████████▉| 370811/371472 [7:20:32<03:18,  3.33it/s]
100%|█████████▉| 370812/371472 [7:20:32<03:06,  3.53it/s]
100%|█████████▉| 370813/371472 [7:20:33<03:09,  3.48it/s]
100%|█████████▉| 370814/371472 [7:20:33<03:17,  3.33it/s]
100%|█████████▉| 370815/371472 [7:20:33<03:08,  3.49it/s]
100%|█████████▉| 370816/371472 [7:20:33<03:02,  3.59it/s]
100%|█████████▉| 370817/371472 [7:20:34<03:17,  3.32it/s]
100%|█████████▉| 370818/371472 [7:20:34<03:11,  3.41it/s]
100%|█████████▉| 370819/371472 [7:20:34<03:14,  3.36it/s]
100%|█████████▉| 370820/371472 [7:20:35<03:07,  3.48it/s]
                                                         
{'loss': 2.5374, 'learning_rate': 1.0158051240061199e-07, 'epoch': 15.97}
+
100%|█████████▉| 370820/371472 [7:20:35<03:07,  3.48it/s]
100%|█████████▉| 370821/371472 [7:20:35<03:00,  3.60it/s]
100%|█████████▉| 370822/371472 [7:20:35<02:57,  3.66it/s]
100%|█████████▉| 370823/371472 [7:20:35<02:59,  3.62it/s]
100%|█████████▉| 370824/371472 [7:20:36<02:58,  3.63it/s]
100%|█████████▉| 370825/371472 [7:20:36<02:56,  3.67it/s]
100%|█████████▉| 370826/371472 [7:20:36<02:52,  3.74it/s]
100%|█████████▉| 370827/371472 [7:20:37<02:47,  3.84it/s]
100%|█████████▉| 370828/371472 [7:20:37<02:46,  3.87it/s]
100%|█████████▉| 370829/371472 [7:20:37<02:48,  3.83it/s]
100%|█████████▉| 370830/371472 [7:20:37<02:44,  3.91it/s]
100%|█████████▉| 370831/371472 [7:20:38<02:50,  3.77it/s]
100%|█████████▉| 370832/371472 [7:20:38<02:49,  3.78it/s]
100%|█████████▉| 370833/371472 [7:20:38<02:45,  3.87it/s]
100%|█████████▉| 370834/371472 [7:20:38<03:00,  3.53it/s]
100%|█████████▉| 370835/371472 [7:20:39<02:55,  3.64it/s]
100%|█████████▉| 370836/371472 [7:20:39<03:03,  3.47it/s]
100%|█████████▉| 370837/371472 [7:20:39<03:16,  3.23it/s]
100%|█████████▉| 370838/371472 [7:20:40<03:14,  3.27it/s]
100%|█████████▉| 370839/371472 [7:20:40<03:44,  2.82it/s]
100%|█████████▉| 370840/371472 [7:20:40<03:42,  2.84it/s]
                                                         
{'loss': 2.6133, 'learning_rate': 1.0153203042513305e-07, 'epoch': 15.97}
+
100%|█████████▉| 370840/371472 [7:20:40<03:42,  2.84it/s]
100%|█████████▉| 370841/371472 [7:20:41<03:24,  3.08it/s]
100%|█████████▉| 370842/371472 [7:20:41<03:15,  3.22it/s]
100%|█████████▉| 370843/371472 [7:20:41<03:21,  3.12it/s]
100%|█████████▉| 370844/371472 [7:20:42<03:14,  3.22it/s]
100%|█████████▉| 370845/371472 [7:20:42<03:09,  3.31it/s]
100%|█████████▉| 370846/371472 [7:20:42<03:15,  3.19it/s]
100%|█████████▉| 370847/371472 [7:20:43<03:27,  3.01it/s]
100%|█████████▉| 370848/371472 [7:20:43<03:25,  3.04it/s]
100%|█████████▉| 370849/371472 [7:20:43<03:18,  3.14it/s]
100%|█████████▉| 370850/371472 [7:20:44<03:07,  3.31it/s]
100%|█████████▉| 370851/371472 [7:20:44<03:14,  3.20it/s]
100%|█████████▉| 370852/371472 [7:20:44<03:01,  3.42it/s]
100%|█████████▉| 370853/371472 [7:20:44<03:00,  3.43it/s]
100%|█████████▉| 370854/371472 [7:20:45<03:06,  3.31it/s]
100%|█████████▉| 370855/371472 [7:20:45<02:58,  3.45it/s]
100%|█████████▉| 370856/371472 [7:20:45<02:54,  3.53it/s]
100%|█████████▉| 370857/371472 [7:20:45<02:45,  3.73it/s]
100%|█████████▉| 370858/371472 [7:20:46<02:52,  3.57it/s]
100%|█████████▉| 370859/371472 [7:20:46<03:06,  3.28it/s]
100%|█████████▉| 370860/371472 [7:20:46<03:01,  3.37it/s]
                                                         
{'loss': 2.4041, 'learning_rate': 1.0148354844965411e-07, 'epoch': 15.97}
+
100%|█████████▉| 370860/371472 [7:20:46<03:01,  3.37it/s]
100%|█████████▉| 370861/371472 [7:20:47<02:55,  3.48it/s]
100%|█████████▉| 370862/371472 [7:20:47<03:07,  3.26it/s]
100%|█████████▉| 370863/371472 [7:20:47<02:56,  3.46it/s]
100%|█████████▉| 370864/371472 [7:20:48<02:46,  3.66it/s]
100%|█████████▉| 370865/371472 [7:20:48<02:45,  3.68it/s]
100%|█████████▉| 370866/371472 [7:20:48<02:44,  3.69it/s]
100%|█████████▉| 370867/371472 [7:20:48<02:50,  3.55it/s]
100%|█████████▉| 370868/371472 [7:20:49<02:46,  3.62it/s]
100%|█████████▉| 370869/371472 [7:20:49<02:49,  3.56it/s]
100%|█████████▉| 370870/371472 [7:20:49<02:43,  3.69it/s]
100%|█████████▉| 370871/371472 [7:20:49<02:43,  3.67it/s]
100%|█████████▉| 370872/371472 [7:20:50<02:57,  3.39it/s]
100%|█████████▉| 370873/371472 [7:20:50<02:52,  3.47it/s]
100%|█████████▉| 370874/371472 [7:20:50<02:57,  3.36it/s]
100%|█████████▉| 370875/371472 [7:20:51<02:51,  3.48it/s]
100%|█████████▉| 370876/371472 [7:20:51<02:57,  3.35it/s]
100%|█████████▉| 370877/371472 [7:20:51<02:54,  3.40it/s]
100%|█████████▉| 370878/371472 [7:20:52<02:59,  3.31it/s]
100%|█████████▉| 370879/371472 [7:20:52<02:59,  3.31it/s]
100%|█████████▉| 370880/371472 [7:20:52<02:57,  3.34it/s]
                                                         
{'loss': 2.5459, 'learning_rate': 1.0143506647417527e-07, 'epoch': 15.97}
+
100%|█████████▉| 370880/371472 [7:20:52<02:57,  3.34it/s]
100%|█████████▉| 370881/371472 [7:20:52<02:48,  3.51it/s]
100%|█████████▉| 370882/371472 [7:20:53<02:43,  3.62it/s]
100%|█████████▉| 370883/371472 [7:20:53<02:39,  3.69it/s]
100%|█████████▉| 370884/371472 [7:20:53<02:57,  3.30it/s]
100%|█████████▉| 370885/371472 [7:20:54<02:55,  3.35it/s]
100%|█████████▉| 370886/371472 [7:20:54<03:08,  3.10it/s]
100%|█████████▉| 370887/371472 [7:20:54<03:15,  3.00it/s]
100%|█████████▉| 370888/371472 [7:20:55<03:07,  3.11it/s]
100%|█████████▉| 370889/371472 [7:20:55<03:00,  3.23it/s]
100%|█████████▉| 370890/371472 [7:20:55<02:56,  3.30it/s]
100%|█████████▉| 370891/371472 [7:20:55<02:45,  3.50it/s]
100%|█████████▉| 370892/371472 [7:20:56<02:41,  3.59it/s]
100%|█████████▉| 370893/371472 [7:20:56<02:51,  3.38it/s]
100%|█████████▉| 370894/371472 [7:20:56<02:47,  3.45it/s]
100%|█████████▉| 370895/371472 [7:20:57<02:44,  3.51it/s]
100%|█████████▉| 370896/371472 [7:20:57<02:43,  3.52it/s]
100%|█████████▉| 370897/371472 [7:20:57<02:41,  3.55it/s]
100%|█████████▉| 370898/371472 [7:20:57<02:40,  3.58it/s]
100%|█████████▉| 370899/371472 [7:20:58<02:41,  3.54it/s]
100%|█████████▉| 370900/371472 [7:20:58<02:40,  3.57it/s]
                                                         
{'loss': 2.5954, 'learning_rate': 1.0138658449869633e-07, 'epoch': 15.98}
+
100%|█████████▉| 370900/371472 [7:20:58<02:40,  3.57it/s]
100%|█████████▉| 370901/371472 [7:20:58<02:40,  3.57it/s]
100%|█████████▉| 370902/371472 [7:20:59<02:35,  3.66it/s]
100%|█████████▉| 370903/371472 [7:20:59<02:47,  3.40it/s]
100%|█████████▉| 370904/371472 [7:20:59<02:40,  3.53it/s]
100%|█████████▉| 370905/371472 [7:20:59<02:34,  3.67it/s]
100%|█████████▉| 370906/371472 [7:21:00<02:34,  3.66it/s]
100%|█████████▉| 370907/371472 [7:21:00<02:34,  3.67it/s]
100%|█████████▉| 370908/371472 [7:21:00<02:28,  3.80it/s]
100%|█████████▉| 370909/371472 [7:21:01<02:38,  3.54it/s]
100%|█████████▉| 370910/371472 [7:21:01<02:37,  3.58it/s]
100%|█████████▉| 370911/371472 [7:21:01<02:30,  3.73it/s]
100%|█████████▉| 370912/371472 [7:21:01<02:37,  3.55it/s]
100%|█████████▉| 370913/371472 [7:21:02<02:35,  3.59it/s]
100%|█████████▉| 370914/371472 [7:21:02<02:29,  3.74it/s]
100%|█████████▉| 370915/371472 [7:21:02<02:33,  3.62it/s]
100%|█████████▉| 370916/371472 [7:21:02<02:34,  3.61it/s]
100%|█████████▉| 370917/371472 [7:21:03<02:30,  3.69it/s]
100%|█████████▉| 370918/371472 [7:21:03<02:36,  3.53it/s]
100%|█████████▉| 370919/371472 [7:21:03<02:30,  3.68it/s]
100%|█████████▉| 370920/371472 [7:21:03<02:26,  3.77it/s]
                                                         
{'loss': 2.5845, 'learning_rate': 1.0133810252321748e-07, 'epoch': 15.98}
+
100%|█████████▉| 370920/371472 [7:21:03<02:26,  3.77it/s]
100%|█████████▉| 370921/371472 [7:21:04<02:26,  3.77it/s]
100%|█████████▉| 370922/371472 [7:21:04<02:24,  3.81it/s]
100%|█████████▉| 370923/371472 [7:21:04<02:20,  3.91it/s]
100%|█████████▉| 370924/371472 [7:21:05<02:33,  3.57it/s]
100%|█████████▉| 370925/371472 [7:21:05<02:31,  3.61it/s]
100%|█████████▉| 370926/371472 [7:21:05<02:36,  3.49it/s]
100%|█████████▉| 370927/371472 [7:21:06<02:44,  3.31it/s]
100%|█████████▉| 370928/371472 [7:21:06<02:46,  3.26it/s]
100%|█████████▉| 370929/371472 [7:21:06<02:47,  3.24it/s]
100%|█████████▉| 370930/371472 [7:21:06<02:42,  3.33it/s]
100%|█████████▉| 370931/371472 [7:21:07<02:48,  3.22it/s]
100%|█████████▉| 370932/371472 [7:21:07<02:40,  3.37it/s]
100%|█████████▉| 370933/371472 [7:21:07<02:55,  3.06it/s]
100%|█████████▉| 370934/371472 [7:21:08<02:44,  3.28it/s]
100%|█████████▉| 370935/371472 [7:21:08<02:36,  3.42it/s]
100%|█████████▉| 370936/371472 [7:21:08<02:31,  3.53it/s]
100%|█████████▉| 370937/371472 [7:21:08<02:29,  3.58it/s]
100%|█████████▉| 370938/371472 [7:21:09<02:34,  3.46it/s]
100%|█████████▉| 370939/371472 [7:21:09<02:25,  3.66it/s]
100%|█████████▉| 370940/371472 [7:21:09<02:25,  3.65it/s]
                                                         
{'loss': 2.4867, 'learning_rate': 1.0128962054773854e-07, 'epoch': 15.98}
+
100%|█████████▉| 370940/371472 [7:21:09<02:25,  3.65it/s]
100%|█████████▉| 370941/371472 [7:21:10<02:23,  3.70it/s]
100%|█████���███▉| 370942/371472 [7:21:10<02:22,  3.73it/s]
100%|█████████▉| 370943/371472 [7:21:10<02:21,  3.75it/s]
100%|█████████▉| 370944/371472 [7:21:10<02:21,  3.73it/s]
100%|█████████▉| 370945/371472 [7:21:11<02:25,  3.62it/s]
100%|█████████▉| 370946/371472 [7:21:11<02:25,  3.63it/s]
100%|█████████▉| 370947/371472 [7:21:11<02:20,  3.73it/s]
100%|█████████▉| 370948/371472 [7:21:11<02:20,  3.73it/s]
100%|█████████▉| 370949/371472 [7:21:12<02:22,  3.66it/s]
100%|█████████▉| 370950/371472 [7:21:12<02:43,  3.20it/s]
100%|█████████▉| 370951/371472 [7:21:12<02:42,  3.22it/s]
100%|█████████▉| 370952/371472 [7:21:13<02:37,  3.30it/s]
100%|█████████▉| 370953/371472 [7:21:13<02:31,  3.43it/s]
100%|█████████▉| 370954/371472 [7:21:13<02:29,  3.47it/s]
100%|█████████▉| 370955/371472 [7:21:14<02:35,  3.33it/s]
100%|█████████▉| 370956/371472 [7:21:14<02:31,  3.40it/s]
100%|█████████▉| 370957/371472 [7:21:14<02:39,  3.23it/s]
100%|█████████▉| 370958/371472 [7:21:14<02:31,  3.39it/s]
100%|█████████▉| 370959/371472 [7:21:15<02:29,  3.43it/s]
100%|█████████▉| 370960/371472 [7:21:15<02:24,  3.56it/s]
                                                         
{'loss': 2.4833, 'learning_rate': 1.012411385722597e-07, 'epoch': 15.98}
+
100%|█████████▉| 370960/371472 [7:21:15<02:24,  3.56it/s]
100%|█████████▉| 370961/371472 [7:21:15<02:25,  3.50it/s]
100%|█████████▉| 370962/371472 [7:21:16<02:21,  3.60it/s]
100%|█████████▉| 370963/371472 [7:21:16<02:15,  3.74it/s]
100%|█████████▉| 370964/371472 [7:21:16<02:18,  3.66it/s]
100%|█████████▉| 370965/371472 [7:21:16<02:26,  3.46it/s]
100%|█████████▉| 370966/371472 [7:21:17<02:47,  3.01it/s]
100%|█████████▉| 370967/371472 [7:21:17<02:43,  3.08it/s]
100%|█████████▉| 370968/371472 [7:21:17<02:32,  3.30it/s]
100%|█████████▉| 370969/371472 [7:21:18<02:29,  3.37it/s]
100%|█████████▉| 370970/371472 [7:21:18<02:29,  3.36it/s]
100%|█████████▉| 370971/371472 [7:21:18<02:24,  3.46it/s]
100%|█████████▉| 370972/371472 [7:21:19<02:21,  3.52it/s]
100%|█████████▉| 370973/371472 [7:21:19<02:17,  3.62it/s]
100%|█████████▉| 370974/371472 [7:21:19<02:13,  3.73it/s]
100%|█████████▉| 370975/371472 [7:21:19<02:19,  3.56it/s]
100%|█████████▉| 370976/371472 [7:21:20<02:17,  3.61it/s]
100%|█████████▉| 370977/371472 [7:21:20<02:21,  3.49it/s]
100%|█████████▉| 370978/371472 [7:21:20<02:24,  3.41it/s]
100%|█████████▉| 370979/371472 [7:21:21<02:19,  3.54it/s]
100%|█████████▉| 370980/371472 [7:21:21<02:14,  3.66it/s]
                                                         
{'loss': 2.4574, 'learning_rate': 1.0119265659678076e-07, 'epoch': 15.98}
+
100%|█████████▉| 370980/371472 [7:21:21<02:14,  3.66it/s]
100%|█████████▉| 370981/371472 [7:21:21<02:35,  3.16it/s]
100%|█████████▉| 370982/371472 [7:21:21<02:33,  3.19it/s]
100%|█████████▉| 370983/371472 [7:21:22<02:23,  3.40it/s]
100%|█████████▉| 370984/371472 [7:21:22<02:16,  3.57it/s]
100%|█████████▉| 370985/371472 [7:21:22<02:17,  3.53it/s]
100%|█████████▉| 370986/371472 [7:21:23<02:14,  3.62it/s]
100%|█████████▉| 370987/371472 [7:21:23<02:15,  3.57it/s]
100%|█████████▉| 370988/371472 [7:21:23<02:12,  3.66it/s]
100%|█████████▉| 370989/371472 [7:21:23<02:24,  3.34it/s]
100%|█████████▉| 370990/371472 [7:21:24<02:17,  3.51it/s]
100%|█████████▉| 370991/371472 [7:21:24<02:26,  3.27it/s]
100%|█████████▉| 370992/371472 [7:21:24<02:22,  3.36it/s]
100%|█████████▉| 370993/371472 [7:21:25<02:21,  3.38it/s]
100%|█████████▉| 370994/371472 [7:21:25<02:20,  3.41it/s]
100%|█████████▉| 370995/371472 [7:21:25<02:14,  3.54it/s]
100%|█████████▉| 370996/371472 [7:21:25<02:12,  3.60it/s]
100%|█████████▉| 370997/371472 [7:21:26<02:13,  3.56it/s]
100%|█████████▉| 370998/371472 [7:21:26<02:11,  3.61it/s]
100%|█████████▉| 370999/371472 [7:21:26<02:12,  3.58it/s]
100%|█████████▉| 371000/371472 [7:21:27<02:10,  3.62it/s]
                                                         
{'loss': 2.5831, 'learning_rate': 1.0114417462130191e-07, 'epoch': 15.98}
+
100%|█████████▉| 371000/371472 [7:21:27<02:10,  3.62it/s]
100%|█████████▉| 371001/371472 [7:21:27<02:10,  3.60it/s]
100%|█████████▉| 371002/371472 [7:21:27<02:12,  3.54it/s]
100%|█████████▉| 371003/371472 [7:21:27<02:15,  3.45it/s]
100%|█████████▉| 371004/371472 [7:21:28<02:20,  3.34it/s]
100%|█████████▉| 371005/371472 [7:21:28<02:15,  3.44it/s]
100%|█████████▉| 371006/371472 [7:21:28<02:12,  3.51it/s]
100%|█████████▉| 371007/371472 [7:21:29<02:07,  3.65it/s]
100%|█████████▉| 371008/371472 [7:21:29<02:13,  3.48it/s]
100%|█████████▉| 371009/371472 [7:21:29<02:10,  3.54it/s]
100%|█████████▉| 371010/371472 [7:21:29<02:12,  3.49it/s]
100%|█████████▉| 371011/371472 [7:21:30<02:14,  3.44it/s]
100%|█████████▉| 371012/371472 [7:21:30<02:09,  3.56it/s]
100%|█████████▉| 371013/371472 [7:21:30<02:07,  3.61it/s]
100%|█████████▉| 371014/371472 [7:21:31<02:05,  3.64it/s]
100%|█████████▉| 371015/371472 [7:21:31<02:08,  3.55it/s]
100%|█████████▉| 371016/371472 [7:21:31<02:05,  3.64it/s]
100%|█████████▉| 371017/371472 [7:21:31<02:11,  3.45it/s]
100%|█████████▉| 371018/371472 [7:21:32<02:21,  3.20it/s]
100%|█████████▉| 371019/371472 [7:21:32<02:13,  3.39it/s]
100%|█████████▉| 371020/371472 [7:21:32<02:14,  3.37it/s]
                                                         
{'loss': 2.6038, 'learning_rate': 1.0109569264582297e-07, 'epoch': 15.98}
+
100%|█████████▉| 371020/371472 [7:21:32<02:14,  3.37it/s]
100%|█████████▉| 371021/371472 [7:21:33<02:05,  3.58it/s]
100%|█████████▉| 371022/371472 [7:21:33<02:13,  3.38it/s]
100%|█████████▉| 371023/371472 [7:21:33<02:10,  3.43it/s]
100%|█████████▉| 371024/371472 [7:21:33<02:03,  3.63it/s]
100%|█████████▉| 371025/371472 [7:21:34<02:11,  3.39it/s]
100%|█████████▉| 371026/371472 [7:21:34<02:10,  3.42it/s]
100%|█████████▉| 371027/371472 [7:21:34<02:08,  3.45it/s]
100%|█████████▉| 371028/371472 [7:21:35<02:05,  3.55it/s]
100%|█████████▉| 371029/371472 [7:21:35<02:14,  3.28it/s]
100%|█████████▉| 371030/371472 [7:21:35<02:13,  3.32it/s]
100%|█████████▉| 371031/371472 [7:21:35<02:05,  3.52it/s]
100%|█████████▉| 371032/371472 [7:21:36<02:00,  3.66it/s]
100%|█████████▉| 371033/371472 [7:21:36<01:56,  3.77it/s]
100%|█████████▉| 371034/371472 [7:21:36<02:07,  3.42it/s]
100%|█████████▉| 371035/371472 [7:21:37<02:03,  3.54it/s]
100%|█████████▉| 371036/371472 [7:21:37<02:00,  3.62it/s]
100%|█████████▉| 371037/371472 [7:21:37<01:56,  3.74it/s]
100%|█████████▉| 371038/371472 [7:21:37<01:58,  3.68it/s]
100%|█████████▉| 371039/371472 [7:21:38<01:54,  3.77it/s]
100%|█████████▉| 371040/371472 [7:21:38<01:54,  3.77it/s]
                                                         
{'loss': 2.6344, 'learning_rate': 1.0104721067034412e-07, 'epoch': 15.98}
+
100%|█████████▉| 371040/371472 [7:21:38<01:54,  3.77it/s]
100%|█████████▉| 371041/371472 [7:21:38<01:58,  3.64it/s]
100%|█████████▉| 371042/371472 [7:21:38<01:55,  3.71it/s]
100%|█████████▉| 371043/371472 [7:21:39<01:55,  3.73it/s]
100%|█████████▉| 371044/371472 [7:21:39<01:54,  3.73it/s]
100%|█████████▉| 371045/371472 [7:21:39<01:57,  3.64it/s]
100%|█████████▉| 371046/371472 [7:21:40<01:58,  3.59it/s]
100%|█████████▉| 371047/371472 [7:21:40<02:02,  3.47it/s]
100%|█████████▉| 371048/371472 [7:21:40<02:08,  3.30it/s]
100%|█████████▉| 371049/371472 [7:21:40<02:04,  3.40it/s]
100%|█████████▉| 371050/371472 [7:21:41<02:02,  3.44it/s]
100%|█████████▉| 371051/371472 [7:21:41<02:07,  3.30it/s]
100%|█████████▉| 371052/371472 [7:21:41<02:02,  3.43it/s]
100%|█████████▉| 371053/371472 [7:21:42<01:57,  3.58it/s]
100%|█████████▉| 371054/371472 [7:21:42<01:55,  3.63it/s]
100%|█████████▉| 371055/371472 [7:21:42<01:54,  3.65it/s]
100%|█████████▉| 371056/371472 [7:21:42<01:51,  3.73it/s]
100%|█████████▉| 371057/371472 [7:21:43<01:59,  3.46it/s]
100%|█████████▉| 371058/371472 [7:21:43<02:00,  3.43it/s]
100%|█████████▉| 371059/371472 [7:21:43<01:59,  3.46it/s]
100%|█████████▉| 371060/371472 [7:21:44<01:56,  3.54it/s]
                                                         
{'loss': 2.6479, 'learning_rate': 1.0099872869486518e-07, 'epoch': 15.98}
+
100%|█████████▉| 371060/371472 [7:21:44<01:56,  3.54it/s]
100%|█████████▉| 371061/371472 [7:21:44<01:54,  3.60it/s]
100%|█████████▉| 371062/371472 [7:21:44<01:49,  3.73it/s]
100%|█████████▉| 371063/371472 [7:21:44<01:49,  3.74it/s]
100%|█████████▉| 371064/371472 [7:21:45<01:50,  3.69it/s]
100%|█████████▉| 371065/371472 [7:21:45<01:48,  3.75it/s]
100%|█████████▉| 371066/371472 [7:21:45<01:49,  3.72it/s]
100%|█████████▉| 371067/371472 [7:21:45<01:47,  3.77it/s]
100%|█████████▉| 371068/371472 [7:21:46<01:47,  3.76it/s]
100%|█████████▉| 371069/371472 [7:21:46<01:48,  3.71it/s]
100%|█████████▉| 371070/371472 [7:21:46<01:53,  3.54it/s]
100%|█████████▉| 371071/371472 [7:21:47<02:02,  3.28it/s]
100%|█████████▉| 371072/371472 [7:21:47<02:01,  3.29it/s]
100%|█████████▉| 371073/371472 [7:21:47<02:04,  3.21it/s]
100%|█████████▉| 371074/371472 [7:21:48<01:56,  3.42it/s]
100%|█████████▉| 371075/371472 [7:21:48<01:51,  3.57it/s]
100%|█████████▉| 371076/371472 [7:21:48<01:51,  3.55it/s]
100%|█████████▉| 371077/371472 [7:21:48<01:51,  3.55it/s]
100%|█████████▉| 371078/371472 [7:21:49<01:48,  3.63it/s]
100%|█████████▉| 371079/371472 [7:21:49<01:46,  3.70it/s]
100%|█████████▉| 371080/371472 [7:21:49<01:43,  3.80it/s]
                                                         
{'loss': 2.6846, 'learning_rate': 1.0095024671938635e-07, 'epoch': 15.98}
+
100%|█████████▉| 371080/371472 [7:21:49<01:43,  3.80it/s]
100%|█████████▉| 371081/371472 [7:21:49<01:43,  3.77it/s]
100%|█████████▉| 371082/371472 [7:21:50<01:56,  3.33it/s]
100%|█████████▉| 371083/371472 [7:21:50<02:05,  3.10it/s]
100%|█████████▉| 371084/371472 [7:21:50<02:03,  3.14it/s]
100%|█████████▉| 371085/371472 [7:21:51<02:01,  3.18it/s]
100%|█████████▉| 371086/371472 [7:21:51<01:59,  3.24it/s]
100%|█████████▉| 371087/371472 [7:21:51<01:54,  3.35it/s]
100%|█████████▉| 371088/371472 [7:21:52<01:52,  3.41it/s]
100%|█████████▉| 371089/371472 [7:21:52<01:55,  3.32it/s]
100%|█████████▉| 371090/371472 [7:21:52<01:53,  3.36it/s]
100%|█████████▉| 371091/371472 [7:21:52<01:49,  3.49it/s]
100%|█████████▉| 371092/371472 [7:21:53<01:48,  3.51it/s]
100%|█████████▉| 371093/371472 [7:21:53<01:45,  3.59it/s]
100%|█████████▉| 371094/371472 [7:21:53<01:49,  3.45it/s]
100%|█████████▉| 371095/371472 [7:21:54<01:50,  3.42it/s]
100%|█████████▉| 371096/371472 [7:21:54<01:52,  3.35it/s]
100%|█████████▉| 371097/371472 [7:21:54<01:52,  3.34it/s]
100%|█████████▉| 371098/371472 [7:21:55<01:48,  3.43it/s]
100%|█████████▉| 371099/371472 [7:21:55<01:46,  3.51it/s]
100%|█████████▉| 371100/371472 [7:21:55<01:46,  3.49it/s]
                                                         
{'loss': 2.5628, 'learning_rate': 1.009017647439074e-07, 'epoch': 15.98}
+
100%|█████████▉| 371100/371472 [7:21:55<01:46,  3.49it/s]
100%|█████████▉| 371101/371472 [7:21:55<01:51,  3.34it/s]
100%|█████████▉| 371102/371472 [7:21:56<01:45,  3.50it/s]
100%|█████████▉| 371103/371472 [7:21:56<01:40,  3.67it/s]
100%|█████████▉| 371104/371472 [7:21:56<01:47,  3.44it/s]
100%|█████████▉| 371105/371472 [7:21:57<01:46,  3.44it/s]
100%|█████████▉| 371106/371472 [7:21:57<01:50,  3.31it/s]
100%|█████████▉| 371107/371472 [7:21:57<01:49,  3.33it/s]
100%|█████████▉| 371108/371472 [7:21:57<01:46,  3.43it/s]
100%|█████████▉| 371109/371472 [7:21:58<01:42,  3.54it/s]
100%|█████████▉| 371110/371472 [7:21:58<01:42,  3.53it/s]
100%|█████████▉| 371111/371472 [7:21:58<01:41,  3.54it/s]
100%|█████████▉| 371112/371472 [7:21:59<01:42,  3.53it/s]
100%|█████████▉| 371113/371472 [7:21:59<01:39,  3.61it/s]
100%|█████████▉| 371114/371472 [7:21:59<01:39,  3.61it/s]
100%|█████████▉| 371115/371472 [7:21:59<01:39,  3.58it/s]
100%|█████████▉| 371116/371472 [7:22:00<01:47,  3.31it/s]
100%|█████████▉| 371117/371472 [7:22:00<01:43,  3.43it/s]
100%|█████████▉| 371118/371472 [7:22:00<01:44,  3.40it/s]
100%|█████████▉| 371119/371472 [7:22:01<01:39,  3.54it/s]
100%|█████████▉| 371120/371472 [7:22:01<01:38,  3.57it/s]
                                                         
{'loss': 2.5493, 'learning_rate': 1.0085328276842855e-07, 'epoch': 15.98}
+
100%|█████████▉| 371120/371472 [7:22:01<01:38,  3.57it/s]
100%|█████████▉| 371121/371472 [7:22:01<01:41,  3.47it/s]
100%|█████████▉| 371122/371472 [7:22:01<01:38,  3.54it/s]
100%|█████████▉| 371123/371472 [7:22:02<01:36,  3.61it/s]
100%|█████████▉| 371124/371472 [7:22:02<01:34,  3.67it/s]
100%|█████████▉| 371125/371472 [7:22:02<01:32,  3.77it/s]
100%|█████████▉| 371126/371472 [7:22:02<01:31,  3.76it/s]
100%|█████████▉| 371127/371472 [7:22:03<01:31,  3.75it/s]
100%|█████████▉| 371128/371472 [7:22:03<01:36,  3.55it/s]
100%|█████████▉| 371129/371472 [7:22:03<01:36,  3.55it/s]
100%|█████████▉| 371130/371472 [7:22:04<01:38,  3.47it/s]
100%|█████████▉| 371131/371472 [7:22:04<01:37,  3.50it/s]
100%|█████████▉| 371132/371472 [7:22:04<01:34,  3.60it/s]
100%|█████████▉| 371133/371472 [7:22:04<01:41,  3.35it/s]
100%|█████████▉| 371134/371472 [7:22:05<01:45,  3.20it/s]
100%|█████████▉| 371135/371472 [7:22:05<01:41,  3.33it/s]
100%|█████████▉| 371136/371472 [7:22:05<01:46,  3.17it/s]
100%|█████████▉| 371137/371472 [7:22:06<01:39,  3.36it/s]
100%|█████████▉| 371138/371472 [7:22:06<01:38,  3.39it/s]
100%|█████████▉| 371139/371472 [7:22:06<01:38,  3.39it/s]
100%|█████████▉| 371140/371472 [7:22:07<01:42,  3.23it/s]
                                                         
{'loss': 2.4884, 'learning_rate': 1.0080480079294961e-07, 'epoch': 15.99}
+
100%|█████████▉| 371140/371472 [7:22:07<01:42,  3.23it/s]
100%|█████████▉| 371141/371472 [7:22:07<01:40,  3.28it/s]
100%|█████████▉| 371142/371472 [7:22:07<01:36,  3.43it/s]
100%|█████████▉| 371143/371472 [7:22:07<01:34,  3.49it/s]
100%|█████████▉| 371144/371472 [7:22:08<01:31,  3.58it/s]
100%|█████████▉| 371145/371472 [7:22:08<01:29,  3.67it/s]
100%|█████████▉| 371146/371472 [7:22:08<01:28,  3.69it/s]
100%|█████████▉| 371147/371472 [7:22:09<01:30,  3.59it/s]
100%|█████████▉| 371148/371472 [7:22:09<01:29,  3.60it/s]
100%|█████████▉| 371149/371472 [7:22:09<01:29,  3.62it/s]
100%|█████████▉| 371150/371472 [7:22:09<01:28,  3.63it/s]
100%|█████████▉| 371151/371472 [7:22:10<01:27,  3.67it/s]
100%|█████████▉| 371152/371472 [7:22:10<01:30,  3.52it/s]
100%|█████████▉| 371153/371472 [7:22:10<01:37,  3.29it/s]
100%|█████████▉| 371154/371472 [7:22:11<01:33,  3.42it/s]
100%|█████████▉| 371155/371472 [7:22:11<01:53,  2.80it/s]
100%|█████████▉| 371156/371472 [7:22:11<01:46,  2.97it/s]
100%|█████████▉| 371157/371472 [7:22:12<01:40,  3.12it/s]
100%|█████████▉| 371158/371472 [7:22:12<01:34,  3.31it/s]
100%|█████████▉| 371159/371472 [7:22:12<01:36,  3.23it/s]
100%|█████████▉| 371160/371472 [7:22:12<01:30,  3.44it/s]
                                                         
{'loss': 2.6734, 'learning_rate': 1.0075631881747078e-07, 'epoch': 15.99}
+
100%|█████████▉| 371160/371472 [7:22:12<01:30,  3.44it/s]
100%|█████████▉| 371161/371472 [7:22:13<01:28,  3.53it/s]
100%|█████████▉| 371162/371472 [7:22:13<01:23,  3.69it/s]
100%|█████████▉| 371163/371472 [7:22:13<01:23,  3.72it/s]
100%|█████████▉| 371164/371472 [7:22:14<01:23,  3.70it/s]
100%|█████████▉| 371165/371472 [7:22:14<01:22,  3.70it/s]
100%|█████████▉| 371166/371472 [7:22:14<01:25,  3.58it/s]
100%|█████████▉| 371167/371472 [7:22:14<01:24,  3.60it/s]
100%|█████████▉| 371168/371472 [7:22:15<01:23,  3.64it/s]
100%|█████████▉| 371169/371472 [7:22:15<01:23,  3.61it/s]
100%|█████████▉| 371170/371472 [7:22:15<01:22,  3.68it/s]
100%|█████████▉| 371171/371472 [7:22:15<01:22,  3.66it/s]
100%|█████████▉| 371172/371472 [7:22:16<01:22,  3.62it/s]
100%|█████████▉| 371173/371472 [7:22:16<01:20,  3.69it/s]
100%|█████████▉| 371174/371472 [7:22:16<01:25,  3.49it/s]
100%|█████████▉| 371175/371472 [7:22:17<01:23,  3.55it/s]
100%|█████████▉| 371176/371472 [7:22:17<01:19,  3.73it/s]
100%|█████████▉| 371177/371472 [7:22:17<01:27,  3.39it/s]
100%|█████████▉| 371178/371472 [7:22:17<01:22,  3.58it/s]
100%|█████████▉| 371179/371472 [7:22:18<01:19,  3.69it/s]
100%|█████████▉| 371180/371472 [7:22:18<01:17,  3.76it/s]
                                                         
{'loss': 2.6717, 'learning_rate': 1.0070783684199184e-07, 'epoch': 15.99}
+
100%|█████████▉| 371180/371472 [7:22:18<01:17,  3.76it/s]
100%|█████████▉| 371181/371472 [7:22:18<01:16,  3.80it/s]
100%|█████████▉| 371182/371472 [7:22:18<01:17,  3.76it/s]
100%|█████████▉| 371183/371472 [7:22:19<01:17,  3.74it/s]
100%|█████████▉| 371184/371472 [7:22:19<01:17,  3.70it/s]
100%|█████████▉| 371185/371472 [7:22:19<01:14,  3.84it/s]
100%|█████████▉| 371186/371472 [7:22:19<01:15,  3.80it/s]
100%|█████████▉| 371187/371472 [7:22:20<01:17,  3.66it/s]
100%|█████████▉| 371188/371472 [7:22:20<01:19,  3.59it/s]
100%|█████████▉| 371189/371472 [7:22:20<01:23,  3.38it/s]
100%|█████████▉| 371190/371472 [7:22:21<01:31,  3.08it/s]
100%|█████████▉| 371191/371472 [7:22:21<01:31,  3.07it/s]
100%|█████████▉| 371192/371472 [7:22:21<01:26,  3.25it/s]
100%|█████████▉| 371193/371472 [7:22:22<01:21,  3.44it/s]
100%|█████████▉| 371194/371472 [7:22:22<01:20,  3.46it/s]
100%|█████████▉| 371195/371472 [7:22:22<01:18,  3.54it/s]
100%|█████████▉| 371196/371472 [7:22:22<01:15,  3.64it/s]
100%|█████████▉| 371197/371472 [7:22:23<01:13,  3.75it/s]
100%|█████████▉| 371198/371472 [7:22:23<01:12,  3.80it/s]
100%|█████████▉| 371199/371472 [7:22:23<01:10,  3.90it/s]
100%|█████████▉| 371200/371472 [7:22:23<01:11,  3.78it/s]
                                                         
{'loss': 2.4994, 'learning_rate': 1.0065935486651299e-07, 'epoch': 15.99}
+
100%|█████████▉| 371200/371472 [7:22:23<01:11,  3.78it/s]
100%|█████████▉| 371201/371472 [7:22:24<01:11,  3.77it/s]
100%|█████████▉| 371202/371472 [7:22:24<01:11,  3.79it/s]
100%|█████████▉| 371203/371472 [7:22:24<01:14,  3.63it/s]
100%|█████████▉| 371204/371472 [7:22:25<01:13,  3.65it/s]
100%|█████████▉| 371205/371472 [7:22:25<01:13,  3.63it/s]
100%|█████████▉| 371206/371472 [7:22:25<01:14,  3.55it/s]
100%|█████████▉| 371207/371472 [7:22:25<01:17,  3.43it/s]
100%|█████████▉| 371208/371472 [7:22:26<01:15,  3.50it/s]
100%|█████████▉| 371209/371472 [7:22:26<01:15,  3.49it/s]
100%|█████████▉| 371210/371472 [7:22:26<01:14,  3.53it/s]
100%|█████████▉| 371211/371472 [7:22:27<01:13,  3.54it/s]
100%|█████████▉| 371212/371472 [7:22:27<01:11,  3.64it/s]
100%|█████████▉| 371213/371472 [7:22:27<01:08,  3.75it/s]
100%|█████████▉| 371214/371472 [7:22:27<01:07,  3.84it/s]
100%|█████████▉| 371215/371472 [7:22:28<01:07,  3.78it/s]
100%|█████████▉| 371216/371472 [7:22:28<01:11,  3.57it/s]
100%|█████████▉| 371217/371472 [7:22:28<01:15,  3.40it/s]
100%|█████████▉| 371218/371472 [7:22:29<01:20,  3.16it/s]
100%|█████████▉| 371219/371472 [7:22:29<01:16,  3.31it/s]
100%|█████████▉| 371220/371472 [7:22:29<01:11,  3.51it/s]
                                                         
{'loss': 2.572, 'learning_rate': 1.0061087289103406e-07, 'epoch': 15.99}
+
100%|█████████▉| 371220/371472 [7:22:29<01:11,  3.51it/s]
100%|█████████▉| 371221/371472 [7:22:29<01:13,  3.41it/s]
100%|█████████▉| 371222/371472 [7:22:30<01:11,  3.51it/s]
100%|█████████▉| 371223/371472 [7:22:30<01:12,  3.44it/s]
100%|█████████▉| 371224/371472 [7:22:30<01:10,  3.52it/s]
100%|█████████▉| 371225/371472 [7:22:31<01:08,  3.62it/s]
100%|█████████▉| 371226/371472 [7:22:31<01:09,  3.54it/s]
100%|█████████▉| 371227/371472 [7:22:31<01:12,  3.36it/s]
100%|█████████▉| 371228/371472 [7:22:31<01:09,  3.52it/s]
100%|█████████▉| 371229/371472 [7:22:32<01:11,  3.39it/s]
100%|█████████▉| 371230/371472 [7:22:32<01:09,  3.49it/s]
100%|█████████▉| 371231/371472 [7:22:32<01:13,  3.26it/s]
100%|█████████▉| 371232/371472 [7:22:33<01:10,  3.40it/s]
100%|█████████▉| 371233/371472 [7:22:33<01:10,  3.40it/s]
100%|█████████▉| 371234/371472 [7:22:33<01:13,  3.24it/s]
100%|█████████▉| 371235/371472 [7:22:34<01:10,  3.38it/s]
100%|█████████▉| 371236/371472 [7:22:34<01:07,  3.47it/s]
100%|█████████▉| 371237/371472 [7:22:34<01:10,  3.35it/s]
100%|█████████▉| 371238/371472 [7:22:34<01:09,  3.35it/s]
100%|█████████▉| 371239/371472 [7:22:35<01:05,  3.55it/s]
100%|█████████▉| 371240/371472 [7:22:35<01:02,  3.69it/s]
                                                         
{'loss': 2.6734, 'learning_rate': 1.0056239091555521e-07, 'epoch': 15.99}
+
100%|█████████▉| 371240/371472 [7:22:35<01:02,  3.69it/s]
100%|█████████▉| 371241/371472 [7:22:35<01:06,  3.49it/s]
100%|█████████▉| 371242/371472 [7:22:36<01:07,  3.39it/s]
100%|█████████▉| 371243/371472 [7:22:36<01:04,  3.56it/s]
100%|█████████▉| 371244/371472 [7:22:36<01:02,  3.63it/s]
100%|█████████▉| 371245/371472 [7:22:36<01:03,  3.60it/s]
100%|█████████▉| 371246/371472 [7:22:37<01:03,  3.58it/s]
100%|█████████▉| 371247/371472 [7:22:37<01:00,  3.72it/s]
100%|█████████▉| 371248/371472 [7:22:37<01:00,  3.68it/s]
100%|█████████▉| 371249/371472 [7:22:37<01:00,  3.66it/s]
100%|█████████▉| 371250/371472 [7:22:38<01:05,  3.37it/s]
100%|█████████▉| 371251/371472 [7:22:38<01:01,  3.59it/s]
100%|█████████▉| 371252/371472 [7:22:38<00:59,  3.69it/s]
100%|█████████▉| 371253/371472 [7:22:39<00:59,  3.68it/s]
100%|█████████▉| 371254/371472 [7:22:39<01:01,  3.57it/s]
100%|█████████▉| 371255/371472 [7:22:39<01:01,  3.54it/s]
100%|█████████▉| 371256/371472 [7:22:39<01:01,  3.52it/s]
100%|█████████▉| 371257/371472 [7:22:40<00:58,  3.66it/s]
100%|█████████▉| 371258/371472 [7:22:40<00:57,  3.75it/s]
100%|█████████▉| 371259/371472 [7:22:40<00:56,  3.74it/s]
100%|█████████▉| 371260/371472 [7:22:40<00:58,  3.64it/s]
                                                         
{'loss': 2.573, 'learning_rate': 1.0051390894007627e-07, 'epoch': 15.99}
+
100%|█████████▉| 371260/371472 [7:22:41<00:58,  3.64it/s]
100%|█████████▉| 371261/371472 [7:22:41<00:57,  3.65it/s]
100%|█████████▉| 371262/371472 [7:22:41<01:00,  3.44it/s]
100%|█████████▉| 371263/371472 [7:22:41<01:00,  3.46it/s]
100%|█████████▉| 371264/371472 [7:22:42<01:05,  3.18it/s]
100%|█████████▉| 371265/371472 [7:22:42<01:01,  3.36it/s]
100%|█████████▉| 371266/371472 [7:22:42<00:59,  3.48it/s]
100%|█████████▉| 371267/371472 [7:22:43<00:58,  3.51it/s]
100%|█████████▉| 371268/371472 [7:22:43<00:56,  3.62it/s]
100%|█████████▉| 371269/371472 [7:22:43<00:56,  3.58it/s]
100%|█████████▉| 371270/371472 [7:22:43<00:59,  3.41it/s]
100%|█████████▉| 371271/371472 [7:22:44<00:56,  3.55it/s]
100%|█████████▉| 371272/371472 [7:22:44<00:54,  3.69it/s]
100%|█████████▉| 371273/371472 [7:22:44<00:57,  3.47it/s]
100%|█████████▉| 371274/371472 [7:22:44<00:54,  3.66it/s]
100%|█████████▉| 371275/371472 [7:22:45<00:55,  3.54it/s]
100%|█████████▉| 371276/371472 [7:22:45<00:54,  3.62it/s]
100%|█████████▉| 371277/371472 [7:22:45<00:53,  3.66it/s]
100%|█████████▉| 371278/371472 [7:22:46<01:02,  3.10it/s]
100%|█████████▉| 371279/371472 [7:22:46<00:58,  3.31it/s]
100%|█████████▉| 371280/371472 [7:22:46<00:57,  3.34it/s]
                                                         
{'loss': 2.5661, 'learning_rate': 1.0046542696459742e-07, 'epoch': 15.99}
+
100%|█████████▉| 371280/371472 [7:22:46<00:57,  3.34it/s]
100%|█████████▉| 371281/371472 [7:22:47<00:55,  3.44it/s]
100%|█████████▉| 371282/371472 [7:22:47<00:52,  3.61it/s]
100%|█████████▉| 371283/371472 [7:22:47<00:53,  3.51it/s]
100%|█████████▉| 371284/371472 [7:22:47<00:52,  3.57it/s]
100%|█████████▉| 371285/371472 [7:22:48<00:49,  3.76it/s]
100%|█████████▉| 371286/371472 [7:22:48<00:51,  3.62it/s]
100%|█████████▉| 371287/371472 [7:22:48<00:50,  3.66it/s]
100%|█████████▉| 371288/371472 [7:22:49<00:53,  3.47it/s]
100%|█████████▉| 371289/371472 [7:22:49<00:50,  3.62it/s]
100%|█████████▉| 371290/371472 [7:22:49<00:56,  3.22it/s]
100%|█████████▉| 371291/371472 [7:22:49<00:54,  3.33it/s]
100%|█████████▉| 371292/371472 [7:22:50<00:51,  3.52it/s]
100%|█████████▉| 371293/371472 [7:22:50<00:50,  3.52it/s]
100%|█████████▉| 371294/371472 [7:22:50<00:49,  3.61it/s]
100%|█████████▉| 371295/371472 [7:22:51<00:50,  3.48it/s]
100%|█████████▉| 371296/371472 [7:22:51<00:52,  3.34it/s]
100%|█████████▉| 371297/371472 [7:22:51<00:52,  3.34it/s]
100%|█████████▉| 371298/371472 [7:22:51<00:52,  3.34it/s]
100%|█████████▉| 371299/371472 [7:22:52<00:50,  3.45it/s]
100%|█████████▉| 371300/371472 [7:22:52<00:48,  3.53it/s]
                                                         
{'loss': 2.5232, 'learning_rate': 1.0041694498911848e-07, 'epoch': 15.99}
+
100%|█████████▉| 371300/371472 [7:22:52<00:48,  3.53it/s]
100%|█████████▉| 371301/371472 [7:22:52<00:49,  3.42it/s]
100%|█████████▉| 371302/371472 [7:22:53<00:51,  3.29it/s]
100%|█████████▉| 371303/371472 [7:22:53<00:50,  3.36it/s]
100%|█████████▉| 371304/371472 [7:22:53<00:49,  3.42it/s]
100%|█████████▉| 371305/371472 [7:22:53<00:47,  3.52it/s]
100%|█████████▉| 371306/371472 [7:22:54<00:47,  3.53it/s]
100%|█████████▉| 371307/371472 [7:22:54<00:46,  3.52it/s]
100%|█████████▉| 371308/371472 [7:22:54<00:47,  3.48it/s]
100%|█████████▉| 371309/371472 [7:22:55<00:45,  3.61it/s]
100%|█████████▉| 371310/371472 [7:22:55<00:45,  3.57it/s]
100%|█████████▉| 371311/371472 [7:22:55<00:43,  3.69it/s]
100%|█████████▉| 371312/371472 [7:22:55<00:43,  3.69it/s]
100%|█████████▉| 371313/371472 [7:22:56<00:42,  3.74it/s]
100%|█████████▉| 371314/371472 [7:22:56<00:45,  3.49it/s]
100%|█████████▉| 371315/371472 [7:22:56<00:47,  3.33it/s]
100%|█████████▉| 371316/371472 [7:22:57<00:44,  3.50it/s]
100%|█████████▉| 371317/371472 [7:22:57<00:43,  3.57it/s]
100%|█████████▉| 371318/371472 [7:22:57<00:44,  3.50it/s]
100%|█████████▉| 371319/371472 [7:22:57<00:47,  3.25it/s]
100%|█████████▉| 371320/371472 [7:22:58<00:45,  3.34it/s]
                                                         
{'loss': 2.5395, 'learning_rate': 1.0036846301363964e-07, 'epoch': 15.99}
+
100%|█████████▉| 371320/371472 [7:22:58<00:45,  3.34it/s]
100%|█████████▉| 371321/371472 [7:22:58<00:44,  3.39it/s]
100%|█████████▉| 371322/371472 [7:22:58<00:43,  3.45it/s]
100%|█████████▉| 371323/371472 [7:22:59<00:42,  3.51it/s]
100%|█████████▉| 371324/371472 [7:22:59<00:43,  3.40it/s]
100%|█████████▉| 371325/371472 [7:22:59<00:40,  3.60it/s]
100%|█████████▉| 371326/371472 [7:22:59<00:40,  3.59it/s]
100%|█████████▉| 371327/371472 [7:23:00<00:39,  3.71it/s]
100%|█████████▉| 371328/371472 [7:23:00<00:39,  3.61it/s]
100%|█████████▉| 371329/371472 [7:23:00<00:38,  3.72it/s]
100%|█████████▉| 371330/371472 [7:23:01<00:38,  3.71it/s]
100%|█████████▉| 371331/371472 [7:23:01<00:38,  3.62it/s]
100%|█████████▉| 371332/371472 [7:23:01<00:39,  3.55it/s]
100%|█████████▉| 371333/371472 [7:23:01<00:39,  3.55it/s]
100%|█████████▉| 371334/371472 [7:23:02<00:37,  3.70it/s]
100%|█████████▉| 371335/371472 [7:23:02<00:35,  3.83it/s]
100%|█████████▉| 371336/371472 [7:23:02<00:36,  3.78it/s]
100%|█████████▉| 371337/371472 [7:23:02<00:36,  3.67it/s]
100%|█████████▉| 371338/371472 [7:23:03<00:36,  3.68it/s]
100%|█████████▉| 371339/371472 [7:23:03<00:36,  3.67it/s]
100%|█████████▉| 371340/371472 [7:23:03<00:34,  3.80it/s]
                                                         
{'loss': 2.5615, 'learning_rate': 1.003199810381607e-07, 'epoch': 15.99}
+
100%|█████████▉| 371340/371472 [7:23:03<00:34,  3.80it/s]
100%|█████████▉| 371341/371472 [7:23:03<00:34,  3.80it/s]
100%|█████████▉| 371342/371472 [7:23:04<00:34,  3.72it/s]
100%|█████████▉| 371343/371472 [7:23:04<00:34,  3.77it/s]
100%|█████████▉| 371344/371472 [7:23:04<00:34,  3.75it/s]
100%|█████████▉| 371345/371472 [7:23:05<00:35,  3.61it/s]
100%|█████████▉| 371346/371472 [7:23:05<00:34,  3.69it/s]
100%|█████████▉| 371347/371472 [7:23:05<00:33,  3.75it/s]
100%|█████████▉| 371348/371472 [7:23:05<00:34,  3.59it/s]
100%|█████████▉| 371349/371472 [7:23:06<00:35,  3.44it/s]
100%|█████████▉| 371350/371472 [7:23:06<00:36,  3.39it/s]
100%|█████████▉| 371351/371472 [7:23:06<00:35,  3.41it/s]
100%|█████████▉| 371352/371472 [7:23:07<00:33,  3.57it/s]
100%|█████████▉| 371353/371472 [7:23:07<00:33,  3.56it/s]
100%|█████████▉| 371354/371472 [7:23:07<00:32,  3.64it/s]
100%|█████████▉| 371355/371472 [7:23:07<00:31,  3.71it/s]
100%|█████████▉| 371356/371472 [7:23:08<00:31,  3.69it/s]
100%|█████████▉| 371357/371472 [7:23:08<00:33,  3.40it/s]
100%|█████████▉| 371358/371472 [7:23:08<00:33,  3.42it/s]
100%|█████████▉| 371359/371472 [7:23:09<00:32,  3.42it/s]
100%|█████████▉| 371360/371472 [7:23:09<00:35,  3.17it/s]
                                                         
{'loss': 2.5744, 'learning_rate': 1.0027149906268185e-07, 'epoch': 16.0}
+
100%|█████████▉| 371360/371472 [7:23:09<00:35,  3.17it/s]
100%|█████████▉| 371361/371472 [7:23:09<00:35,  3.08it/s]
100%|█████████▉| 371362/371472 [7:23:10<00:34,  3.23it/s]
100%|█████████▉| 371363/371472 [7:23:10<00:32,  3.37it/s]
100%|█████████▉| 371364/371472 [7:23:10<00:33,  3.25it/s]
100%|█████████▉| 371365/371472 [7:23:11<00:36,  2.95it/s]
100%|█████████▉| 371366/371472 [7:23:11<00:35,  3.01it/s]
100%|█████████▉| 371367/371472 [7:23:11<00:34,  3.04it/s]
100%|█████████▉| 371368/371472 [7:23:12<00:33,  3.12it/s]
100%|█████████▉| 371369/371472 [7:23:12<00:32,  3.20it/s]
100%|█████████▉| 371370/371472 [7:23:12<00:30,  3.36it/s]
100%|█████████▉| 371371/371472 [7:23:12<00:30,  3.26it/s]
100%|█████████▉| 371372/371472 [7:23:13<00:30,  3.26it/s]
100%|█████████▉| 371373/371472 [7:23:13<00:29,  3.34it/s]
100%|█████████▉| 371374/371472 [7:23:13<00:29,  3.37it/s]
100%|█████████▉| 371375/371472 [7:23:14<00:31,  3.13it/s]
100%|█████████▉| 371376/371472 [7:23:14<00:29,  3.28it/s]
100%|█████████▉| 371377/371472 [7:23:14<00:26,  3.52it/s]
100%|█████████▉| 371378/371472 [7:23:14<00:27,  3.39it/s]
100%|█████████▉| 371379/371472 [7:23:15<00:26,  3.45it/s]
100%|█████████▉| 371380/371472 [7:23:15<00:26,  3.51it/s]
                                                         
{'loss': 2.602, 'learning_rate': 1.0022301708720291e-07, 'epoch': 16.0}
+
100%|█████████▉| 371380/371472 [7:23:15<00:26,  3.51it/s]
100%|█████████▉| 371381/371472 [7:23:15<00:27,  3.33it/s]
100%|█████████▉| 371382/371472 [7:23:16<00:27,  3.27it/s]
100%|█████████▉| 371383/371472 [7:23:16<00:26,  3.34it/s]
100%|█████████▉| 371384/371472 [7:23:16<00:25,  3.39it/s]
100%|█████████▉| 371385/371472 [7:23:17<00:25,  3.46it/s]
100%|█████████▉| 371386/371472 [7:23:17<00:24,  3.46it/s]
100%|█████████▉| 371387/371472 [7:23:17<00:24,  3.50it/s]
100%|█████████▉| 371388/371472 [7:23:17<00:24,  3.42it/s]
100%|█████████▉| 371389/371472 [7:23:18<00:24,  3.43it/s]
100%|█████████▉| 371390/371472 [7:23:18<00:24,  3.40it/s]
100%|█████████▉| 371391/371472 [7:23:18<00:23,  3.49it/s]
100%|█████████▉| 371392/371472 [7:23:19<00:24,  3.32it/s]
100%|█████████▉| 371393/371472 [7:23:19<00:23,  3.41it/s]
100%|█████████▉| 371394/371472 [7:23:19<00:22,  3.47it/s]
100%|█████████▉| 371395/371472 [7:23:19<00:22,  3.47it/s]
100%|█████████▉| 371396/371472 [7:23:20<00:21,  3.53it/s]
100%|█████████▉| 371397/371472 [7:23:20<00:20,  3.70it/s]
100%|█████████▉| 371398/371472 [7:23:20<00:19,  3.73it/s]
100%|█████████▉| 371399/371472 [7:23:20<00:19,  3.70it/s]
100%|█████████▉| 371400/371472 [7:23:21<00:21,  3.40it/s]
                                                         
{'loss': 2.7026, 'learning_rate': 1.0017453511172397e-07, 'epoch': 16.0}
+
100%|█████████▉| 371400/371472 [7:23:21<00:21,  3.40it/s]
100%|█████████▉| 371401/371472 [7:23:21<00:21,  3.26it/s]
100%|█████████▉| 371402/371472 [7:23:21<00:21,  3.30it/s]
100%|█████████▉| 371403/371472 [7:23:22<00:20,  3.34it/s]
100%|█████████▉| 371404/371472 [7:23:22<00:21,  3.13it/s]
100%|█████████▉| 371405/371472 [7:23:22<00:20,  3.33it/s]
100%|█████████▉| 371406/371472 [7:23:23<00:18,  3.51it/s]
100%|█████████▉| 371407/371472 [7:23:23<00:17,  3.64it/s]
100%|█████████▉| 371408/371472 [7:23:23<00:17,  3.71it/s]
100%|█████████▉| 371409/371472 [7:23:23<00:17,  3.55it/s]
100%|█████████▉| 371410/371472 [7:23:24<00:16,  3.68it/s]
100%|█████████▉| 371411/371472 [7:23:24<00:18,  3.25it/s]
100%|█████████▉| 371412/371472 [7:23:24<00:18,  3.16it/s]
100%|█████████▉| 371413/371472 [7:23:25<00:18,  3.14it/s]
100%|█████████▉| 371414/371472 [7:23:25<00:19,  2.96it/s]
100%|█████████▉| 371415/371472 [7:23:25<00:18,  3.13it/s]
100%|█████████▉| 371416/371472 [7:23:26<00:16,  3.31it/s]
100%|█████████▉| 371417/371472 [7:23:26<00:16,  3.39it/s]
100%|█████████▉| 371418/371472 [7:23:26<00:15,  3.40it/s]
100%|█████████▉| 371419/371472 [7:23:27<00:15,  3.37it/s]
100%|█████████▉| 371420/371472 [7:23:27<00:14,  3.51it/s]
                                                         
{'loss': 2.564, 'learning_rate': 1.0012605313624513e-07, 'epoch': 16.0}
+
100%|█████████▉| 371420/371472 [7:23:27<00:14,  3.51it/s]
100%|█████████▉| 371421/371472 [7:23:27<00:14,  3.61it/s]
100%|█████████▉| 371422/371472 [7:23:27<00:13,  3.75it/s]
100%|█████████▉| 371423/371472 [7:23:28<00:13,  3.70it/s]
100%|█████████▉| 371424/371472 [7:23:28<00:13,  3.47it/s]
100%|█████████▉| 371425/371472 [7:23:28<00:13,  3.38it/s]
100%|█████████▉| 371426/371472 [7:23:29<00:13,  3.29it/s]
100%|█████████▉| 371427/371472 [7:23:29<00:12,  3.48it/s]
100%|█████████▉| 371428/371472 [7:23:29<00:12,  3.51it/s]
100%|█████████▉| 371429/371472 [7:23:29<00:11,  3.70it/s]
100%|█████████▉| 371430/371472 [7:23:30<00:10,  3.86it/s]
100%|█████████▉| 371431/371472 [7:23:30<00:10,  3.86it/s]
100%|█████████▉| 371432/371472 [7:23:30<00:10,  3.82it/s]
100%|█████████▉| 371433/371472 [7:23:30<00:11,  3.50it/s]
100%|█████████▉| 371434/371472 [7:23:31<00:10,  3.50it/s]
100%|█████████▉| 371435/371472 [7:23:31<00:10,  3.49it/s]
100%|█████████▉| 371436/371472 [7:23:31<00:10,  3.50it/s]
100%|█████████▉| 371437/371472 [7:23:32<00:09,  3.65it/s]
100%|█████████▉| 371438/371472 [7:23:32<00:10,  3.23it/s]
100%|█████████▉| 371439/371472 [7:23:32<00:09,  3.36it/s]
100%|█████████▉| 371440/371472 [7:23:33<00:09,  3.24it/s]
                                                         
{'loss': 2.6131, 'learning_rate': 1.0007757116076619e-07, 'epoch': 16.0}
+
100%|█████████▉| 371440/371472 [7:23:33<00:09,  3.24it/s]
100%|█████████▉| 371441/371472 [7:23:33<00:09,  3.38it/s]
100%|█████████▉| 371442/371472 [7:23:33<00:08,  3.60it/s]
100%|█████████▉| 371443/371472 [7:23:33<00:08,  3.62it/s]
100%|█████████▉| 371444/371472 [7:23:34<00:07,  3.73it/s]
100%|█████████▉| 371445/371472 [7:23:34<00:07,  3.54it/s]
100%|█████████▉| 371446/371472 [7:23:34<00:07,  3.36it/s]
100%|█████████▉| 371447/371472 [7:23:34<00:06,  3.58it/s]
100%|█████████▉| 371448/371472 [7:23:35<00:06,  3.66it/s]
100%|█████████▉| 371449/371472 [7:23:35<00:06,  3.59it/s]
100%|█████████▉| 371450/371472 [7:23:35<00:05,  3.76it/s]
100%|█████████▉| 371451/371472 [7:23:35<00:05,  3.81it/s]
100%|█████████▉| 371452/371472 [7:23:36<00:05,  3.66it/s]
100%|█████████▉| 371453/371472 [7:23:36<00:05,  3.70it/s]
100%|█████████▉| 371454/371472 [7:23:36<00:04,  3.61it/s]
100%|█████████▉| 371455/371472 [7:23:37<00:04,  3.58it/s]
100%|█████████▉| 371456/371472 [7:23:37<00:04,  3.73it/s]
100%|█████████▉| 371457/371472 [7:23:37<00:04,  3.73it/s]
100%|█████████▉| 371458/371472 [7:23:37<00:03,  3.75it/s]
100%|█████████▉| 371459/371472 [7:23:38<00:03,  3.79it/s]
100%|█████████▉| 371460/371472 [7:23:38<00:03,  3.69it/s]
                                                         
{'loss': 2.4045, 'learning_rate': 1.0002908918528734e-07, 'epoch': 16.0}
+
100%|█████████▉| 371460/371472 [7:23:38<00:03,  3.69it/s]
100%|█████████▉| 371461/371472 [7:23:38<00:02,  3.70it/s]
100%|█████████▉| 371462/371472 [7:23:38<00:02,  3.69it/s]
100%|█████████▉| 371463/371472 [7:23:39<00:02,  3.57it/s]
100%|█████████▉| 371464/371472 [7:23:39<00:02,  3.59it/s]
100%|█████████▉| 371465/371472 [7:23:39<00:02,  3.41it/s]
100%|█████████▉| 371466/371472 [7:23:40<00:01,  3.57it/s]
100%|█████████▉| 371467/371472 [7:23:40<00:01,  3.65it/s]
100%|█████████▉| 371468/371472 [7:23:40<00:01,  3.71it/s]
100%|█████████▉| 371469/371472 [7:23:40<00:00,  3.70it/s]
100%|█████████▉| 371470/371472 [7:23:41<00:00,  3.55it/s]
100%|█████████▉| 371471/371472 [7:23:41<00:00,  3.65it/s]
100%|██████████| 371472/371472 [7:23:41<00:00,  3.48it/s]Some non-default generation parameters are set in the model config. These should go into a GenerationConfig file (https://huggingface.co/docs/transformers/generation_strategies#save-a-custom-decoding-strategy-with-your-model) instead. This warning will be raised to an exception in v4.41.
+Non-default generation parameters: {'max_length': 200, 'early_stopping': True, 'num_beams': 5, 'forced_eos_token_id': 2}
+
                                                         
{'train_runtime': 26664.9079, 'train_samples_per_second': 55.724, 'train_steps_per_second': 13.931, 'train_loss': 0.6528488995518832, 'epoch': 16.0}
+
100%|██████████| 371472/371472 [7:24:08<00:00,  3.48it/s]
100%|██████████| 371472/371472 [7:24:08<00:00, 13.94it/s]