Training in progress, epoch 13

logs/events.out.tfevents.1716905625.sphinx2

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:306c64c353025809371aacd11d55db50acb39f6b8a2bc5a8117b5119b36861c2
-size 88574
+oid sha256:31601533b20d2849a14ea579861c813c0a35434560a935bfbecba6703d27a1a8
+size 95469

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0b38fb2fb234db7b1c09baa9b5ae3584f6b4b1ff93eab3dc3cc814b459faca37
+oid sha256:75d2e9312f6acf3730fee259b2478884c3c7191777eec4387c260c51a2516148
 size 281715176

train_job_output.txt

@@ -604,4 +604,36 @@ command outputs:
 
                                                      
 92%|█████████▏| 9850/10682 [1:23:57<06:51,  2.02it/s]
 92%|█████████▏| 9851/10682 [1:23:57<06:51,  2.02it/s]
 92%|█████████▏| 9852/10682 [1:23:58<06:51,  2.02it/s]
 92%|█████████▏| 9853/10682 [1:23:58<06:50,  2.02it/s]
 92%|█████████▏| 9854/10682 [1:23:59<06:49,  2.02it/s]
 92%|█████████▏| 9855/10682 [1:23:59<06:49,  2.02it/s]
 92%|█████████▏| 9856/10682 [1:24:00<06:48,  2.02it/s]
 92%|█████████▏| 9857/10682 [1:24:00<06:48,  2.02it/s]
 92%|█████████▏| 9858/10682 [1:24:01<06:47,  2.02it/s]
 92%|█████████▏| 9859/10682 [1:24:01<06:47,  2.02it/s]
 92%|█████████▏| 9860/10682 [1:24:02<06:46,  2.02it/s]
 92%|█████████▏| 9861/10682 [1:24:02<06:46,  2.02it/s]
 92%|█████████▏| 9862/10682 [1:24:03<06:45,  2.02it/s]
 92%|█████████▏| 9863/10682 [1:24:03<06:45,  2.02it/s]
 92%|█████████▏| 9864/10682 [1:24:04<06:44,  2.02it/s]
 92%|█████████▏| 9865/10682 [1:24:04<06:44,  2.02it/s]
 92%|█████████▏| 9866/10682 [1:24:05<06:43,  2.02it/s]
 92%|█████████▏| 9867/10682 [1:24:05<06:43,  2.02it/s]
 92%|█████████▏| 9868/10682 [1:24:06<06:42,  2.02it/s]
 92%|█████████▏| 9869/10682 [1:24:06<06:42,  2.02it/s]
 92%|█████████▏| 9870/10682 [1:24:07<06:42,  2.02it/s]
 92%|█████████▏| 9871/10682 [1:24:07<06:41,  2.02it/s]
 92%|█████████▏| 9872/10682 [1:24:08<06:41,  2.02it/s]
 92%|█████████▏| 9873/10682 [1:24:08<06:40,  2.02it/s]
 92%|█████████▏| 9874/10682 [1:24:09<06:40,  2.02it/s]
 92%|█████████▏| 9875/10682 [1:24:09<06:39,  2.02it/s]{'loss': 2.435, 'grad_norm': 0.2683054506778717, 'learning_rate': 1.7288216615031272e-05, 'epoch': 12.93}
 
                                                      
 92%|█████████▏| 9875/10682 [1:24:09<06:39,  2.02it/s]
 92%|█████████▏| 9876/10682 [1:24:10<06:40,  2.01it/s]
 92%|█████████▏| 9877/10682 [1:24:10<06:39,  2.02it/s]
 92%|█████████▏| 9878/10682 [1:24:11<06:39,  2.01it/s]
 92%|█████████▏| 9879/10682 [1:24:11<06:38,  2.02it/s]
 92%|█████████▏| 9880/10682 [1:24:12<06:37,  2.02it/s]
 93%|█████████▎| 9881/10682 [1:24:12<06:37,  2.02it/s]
 93%|█████████▎| 9882/10682 [1:24:13<06:36,  2.02it/s]
 93%|█████████▎| 9883/10682 [1:24:13<06:35,  2.02it/s]
 93%|█████████▎| 9884/10682 [1:24:14<06:35,  2.02it/s]
 93%|█████████▎| 9885/10682 [1:24:14<06:34,  2.02it/s]
 93%|█████████▎| 9886/10682 [1:24:15<06:34,  2.02it/s]
 93%|█████████▎| 9887/10682 [1:24:15<06:33,  2.02it/s]
 93%|█████████▎| 9888/10682 [1:24:16<06:32,  2.02it/s]
 93%|█████████▎| 9889/10682 [1:24:16<06:32,  2.02it/s]
 93%|█████████▎| 9890/10682 [1:24:17<06:31,  2.02it/s]
 93%|█████████▎| 9891/10682 [1:24:17<06:30,  2.02it/s]
 93%|█████████▎| 9892/10682 [1:24:18<06:30,  2.02it/s]
 93%|█████████▎| 9893/10682 [1:24:18<06:30,  2.02it/s]
 93%|█████████▎| 9894/10682 [1:24:19<06:29,  2.02it/s]
 93%|█████████▎| 9895/10682 [1:24:19<06:28,  2.02it/s]
 93%|█████████▎| 9896/10682 [1:24:20<06:29,  2.02it/s]
 93%|█████████▎| 9897/10682 [1:24:20<06:28,  2.02it/s]
 93%|█████████▎| 9898/10682 [1:24:21<06:27,  2.02it/s]
 93%|█████████▎| 9899/10682 [1:24:21<06:26,  2.02it/s]
 93%|█████████▎| 9900/10682 [1:24:22<06:26,  2.02it/s]{'loss': 2.4431, 'grad_norm': 0.26754263043403625, 'learning_rate': 1.6239414036870183e-05, 'epoch': 12.97}
 
                                                      
 93%|█████████▎| 9900/10682 [1:24:22<06:26,  2.02it/s]
 93%|█████████▎| 9901/10682 [1:24:22<06:26,  2.02it/s]
 93%|█████████▎| 9902/10682 [1:24:23<06:26,  2.02it/s]
 93%|█████████▎| 9903/10682 [1:24:23<06:25,  2.02it/s]
 93%|█████████▎| 9904/10682 [1:24:24<06:24,  2.02it/s]
 93%|█████████▎| 9905/10682 [1:24:24<06:24,  2.02it/s]
 93%|█████████▎| 9906/10682 [1:24:25<06:23,  2.02it/s]
 93%|█████████▎| 9907/10682 [1:24:25<06:23,  2.02it/s]
 93%|█████████▎| 9908/10682 [1:24:26<06:23,  2.02it/s]
 93%|█████████▎| 9909/10682 [1:24:26<06:22,  2.02it/s]
 93%|█████████▎| 9910/10682 [1:24:27<06:22,  2.02it/s]
 93%|█████████▎| 9911/10682 [1:24:27<06:21,  2.02it/s]
 93%|█████████▎| 9912/10682 [1:24:28<06:21,  2.02it/s]
 93%|█████████▎| 9913/10682 [1:24:28<06:20,  2.02it/s]
 93%|█████████▎| 9914/10682 [1:24:29<06:20,  2.02it/s]
 93%|█████████▎| 9915/10682 [1:24:29<06:19,  2.02it/s]
 93%|█████████▎| 9916/10682 [1:24:30<06:19,  2.02it/s]
 93%|█████████▎| 9917/10682 [1:24:30<06:18,  2.02it/s]
 93%|█████████▎| 9918/10682 [1:24:31<06:18,  2.02it/s]
 93%|█████████▎| 9919/10682 [1:24:31<06:17,  2.02it/s]
 93%|█████████▎| 9920/10682 [1:24:32<06:16,  2.02it/s]
 93%|█████████▎| 9921/10682 [1:24:32<06:16,  2.02it/s]
 93%|█████████▎| 9922/10682 [1:24:33<06:15,  2.02it/s]
 93%|█████████▎| 9923/10682 [1:24:33<06:15,  2.02it/s]
 93%|█████████▎| 9924/10682 [1:24:34<06:14,  2.02it/s]
 93%|█████████▎| 9925/10682 [1:24:34<06:23,  1.97it/s]{'loss': 2.4348, 'grad_norm': 0.2628089487552643, 'learning_rate': 1.5222903086944684e-05, 'epoch': 13.0}
-
                                                      
 93%|█████████▎| 9925/10682 [1:24:34<06:23,  1.97it/s]
 93%|█████████▎| 9926/10682 [1:24:46<49:14,  3.91s/it]
 93%|█████████▎| 9927/10682 [1:24:47<36:18,  2.89s/it]
 93%|█████████▎| 9928/10682 [1:24:47<27:14,  2.17s/it]
 93%|█████████▎| 9929/10682 [1:24:47<20:55,  1.67s/it]
 93%|█████████▎| 9930/10682 [1:24:48<16:30,  1.32s/it]
 93%|█████████▎| 9931/10682 [1:24:48<13:23,  1.07s/it]
 93%|█████████▎| 9932/10682 [1:24:49<11:14,  1.11it/s]
 93%|█████████▎| 9933/10682 [1:24:49<09:43,  1.28it/s]
 93%|█████████▎| 9934/10682 [1:24:50<08:39,  1.44it/s]
 93%|█████████▎| 9935/10682 [1:24:50<07:54,  1.57it/s]
 93%|█████████▎| 9936/10682 [1:24:51<07:23,  1.68it/s]
 93%|█████████▎| 9937/10682 [1:24:51<07:01,  1.77it/s]
 93%|█████████▎| 9938/10682 [1:24:52<06:44,  1.84it/s]
 93%|█████████▎| 9939/10682 [1:24:52<06:33,  1.89it/s]
 93%|█████████▎| 9940/10682 [1:24:53<06:25,  1.92it/s]
 93%|█████████▎| 9941/10682 [1:24:53<06:19,  1.95it/s]
 93%|█████████▎| 9942/10682 [1:24:54<06:15,  1.97it/s]
 93%|█████████▎| 9943/10682 [1:24:54<06:12,  1.99it/s]
 93%|█████████▎| 9944/10682 [1:24:55<06:10,  1.99it/s]
 93%|█████████▎| 9945/10682 [1:24:55<06:08,  2.00it/s]
 93%|█████████▎| 9946/10682 [1:24:56<06:06,  2.01it/s]
 93%|█████████▎| 9947/10682 [1:24:56<06:05,  2.01it/s]
 93%|█████████▎| 9948/10682 [1:24:57<06:05,  2.01it/s]
 93%|█████████▎| 9949/10682 [1:24:57<06:03,  2.02it/s]
+
                                                      
 93%|█████████▎| 9925/10682 [1:24:34<06:23,  1.97it/s]
 93%|█████████▎| 9926/10682 [1:24:46<49:14,  3.91s/it]
 93%|█████████▎| 9927/10682 [1:24:47<36:18,  2.89s/it]
 93%|█████████▎| 9928/10682 [1:24:47<27:14,  2.17s/it]
 93%|█████████▎| 9929/10682 [1:24:47<20:55,  1.67s/it]
 93%|█████████▎| 9930/10682 [1:24:48<16:30,  1.32s/it]
 93%|█████████▎| 9931/10682 [1:24:48<13:23,  1.07s/it]
 93%|█████████▎| 9932/10682 [1:24:49<11:14,  1.11it/s]
 93%|█████████▎| 9933/10682 [1:24:49<09:43,  1.28it/s]
 93%|█████████▎| 9934/10682 [1:24:50<08:39,  1.44it/s]
 93%|█████████▎| 9935/10682 [1:24:50<07:54,  1.57it/s]
 93%|█████████▎| 9936/10682 [1:24:51<07:23,  1.68it/s]
 93%|█████████▎| 9937/10682 [1:24:51<07:01,  1.77it/s]
 93%|█████████▎| 9938/10682 [1:24:52<06:44,  1.84it/s]
 93%|█████████▎| 9939/10682 [1:24:52<06:33,  1.89it/s]
 93%|█████████▎| 9940/10682 [1:24:53<06:25,  1.92it/s]
 93%|█████████▎| 9941/10682 [1:24:53<06:19,  1.95it/s]
 93%|█████████▎| 9942/10682 [1:24:54<06:15,  1.97it/s]
 93%|█████████▎| 9943/10682 [1:24:54<06:12,  1.99it/s]
 93%|█████████▎| 9944/10682 [1:24:55<06:10,  1.99it/s]
 93%|█████████▎| 9945/10682 [1:24:55<06:08,  2.00it/s]
 93%|█████████▎| 9946/10682 [1:24:56<06:06,  2.01it/s]
 93%|█████████▎| 9947/10682 [1:24:56<06:05,  2.01it/s]
 93%|█████████▎| 9948/10682 [1:24:57<06:05,  2.01it/s]
 93%|█████████▎| 9949/10682 [1:24:57<06:03,  2.02it/s]
 93%|█████████▎| 9950/10682 [1:24:58<06:02,  2.02it/s]
                                                      {'loss': 2.4029, 'grad_norm': 0.26635313034057617, 'learning_rate': 1.4238751618640577e-05, 'epoch': 13.03}
+
 93%|█████████▎| 9950/10682 [1:24:58<06:02,  2.02it/s]
 93%|█████████▎| 9951/10682 [1:24:58<06:02,  2.02it/s]
 93%|█████████▎| 9952/10682 [1:24:59<06:01,  2.02it/s]
 93%|█████████▎| 9953/10682 [1:24:59<06:00,  2.02it/s]
 93%|█████████▎| 9954/10682 [1:25:00<06:00,  2.02it/s]
 93%|█████████▎| 9955/10682 [1:25:00<05:59,  2.02it/s]
 93%|█████████▎| 9956/10682 [1:25:01<05:59,  2.02it/s]
 93%|█████████▎| 9957/10682 [1:25:01<05:58,  2.02it/s]
 93%|█████████▎| 9958/10682 [1:25:02<05:58,  2.02it/s]
 93%|█████████▎| 9959/10682 [1:25:02<05:57,  2.02it/s]
 93%|█████████▎| 9960/10682 [1:25:03<05:57,  2.02it/s]
 93%|█████████▎| 9961/10682 [1:25:03<05:56,  2.02it/s]
 93%|█████████▎| 9962/10682 [1:25:04<05:56,  2.02it/s]
 93%|█████████▎| 9963/10682 [1:25:04<05:55,  2.02it/s]
 93%|█████████▎| 9964/10682 [1:25:05<05:55,  2.02it/s]
 93%|█████████▎| 9965/10682 [1:25:05<05:54,  2.02it/s]
 93%|█████████▎| 9966/10682 [1:25:06<05:54,  2.02it/s]
 93%|█████████▎| 9967/10682 [1:25:06<05:54,  2.02it/s]
 93%|█████████▎| 9968/10682 [1:25:07<05:53,  2.02it/s]
 93%|█████████▎| 9969/10682 [1:25:07<05:53,  2.02it/s]
 93%|█████████▎| 9970/10682 [1:25:08<05:52,  2.02it/s]
 93%|█████████▎| 9971/10682 [1:25:08<05:52,  2.02it/s]
 93%|█████████▎| 9972/10682 [1:25:09<05:51,  2.02it/s]
 93%|█████████▎| 9973/10682 [1:25:09<05:51,  2.02it/s]
 93%|█████████▎| 9974/10682 [1:25:10<05:50,  2.02it/s]
 93%|█████████▎| 9975/10682 [1:25:10<05:49,  2.02it/s]{'loss': 2.4068, 'grad_norm': 0.26276275515556335, 'learning_rate': 1.3287025325307511e-05, 'epoch': 13.06}
+
                                                      
 93%|█████████▎| 9975/10682 [1:25:10<05:49,  2.02it/s]
 93%|█████████▎| 9976/10682 [1:25:11<05:49,  2.02it/s]
 93%|█████████▎| 9977/10682 [1:25:11<05:48,  2.02it/s]
 93%|█████████▎| 9978/10682 [1:25:12<05:48,  2.02it/s]
 93%|█████████▎| 9979/10682 [1:25:12<05:48,  2.02it/s]
 93%|█████████▎| 9980/10682 [1:25:13<05:47,  2.02it/s]
 93%|█████████▎| 9981/10682 [1:25:13<05:47,  2.02it/s]
 93%|█████████▎| 9982/10682 [1:25:14<05:46,  2.02it/s]
 93%|█████████▎| 9983/10682 [1:25:14<05:45,  2.02it/s]
 93%|█████████▎| 9984/10682 [1:25:15<05:45,  2.02it/s]
 93%|█████████▎| 9985/10682 [1:25:15<05:44,  2.02it/s]
 93%|█████████▎| 9986/10682 [1:25:16<05:44,  2.02it/s]
 93%|█████████▎| 9987/10682 [1:25:16<05:43,  2.02it/s]
 94%|█████████▎| 9988/10682 [1:25:17<05:43,  2.02it/s]
 94%|█████████▎| 9989/10682 [1:25:17<05:42,  2.02it/s]
 94%|█████████▎| 9990/10682 [1:25:18<05:42,  2.02it/s]
 94%|█████████▎| 9991/10682 [1:25:18<05:41,  2.02it/s]
 94%|█████████▎| 9992/10682 [1:25:19<05:41,  2.02it/s]
 94%|█████████▎| 9993/10682 [1:25:19<05:40,  2.02it/s]
 94%|█████████▎| 9994/10682 [1:25:20<05:40,  2.02it/s]
 94%|█████████▎| 9995/10682 [1:25:20<05:39,  2.02it/s]
 94%|█████████▎| 9996/10682 [1:25:21<05:38,  2.02it/s]
 94%|█████████▎| 9997/10682 [1:25:21<05:38,  2.02it/s]
 94%|█████████▎| 9998/10682 [1:25:22<05:37,  2.03it/s]
 94%|█████████▎| 9999/10682 [1:25:22<05:37,  2.02it/s]
 94%|█████████▎| 10000/10682 [1:25:23<05:37,  2.02it/s]{'loss': 2.3947, 'grad_norm': 0.26599612832069397, 'learning_rate': 1.2367787735873993e-05, 'epoch': 13.1}
+
                                                       
 94%|█████████▎| 10000/10682 [1:25:23<05:37,  2.02it/s]
 94%|█████████▎| 10001/10682 [1:25:23<05:37,  2.02it/s]
 94%|█████████▎| 10002/10682 [1:25:24<05:36,  2.02it/s]
 94%|█████████▎| 10003/10682 [1:25:24<05:36,  2.02it/s]
 94%|█████████▎| 10004/10682 [1:25:25<05:35,  2.02it/s]
 94%|█████████▎| 10005/10682 [1:25:25<05:34,  2.02it/s]
 94%|█████████▎| 10006/10682 [1:25:26<05:34,  2.02it/s]
 94%|█████████▎| 10007/10682 [1:25:26<05:34,  2.02it/s]
 94%|█████████▎| 10008/10682 [1:25:27<05:33,  2.02it/s]
 94%|█████████▎| 10009/10682 [1:25:27<05:33,  2.02it/s]
 94%|█████████▎| 10010/10682 [1:25:28<05:32,  2.02it/s]
 94%|█████████▎| 10011/10682 [1:25:28<05:32,  2.02it/s]
 94%|█████████▎| 10012/10682 [1:25:29<05:31,  2.02it/s]
 94%|█████████▎| 10013/10682 [1:25:29<05:31,  2.02it/s]
 94%|█████████▎| 10014/10682 [1:25:30<05:30,  2.02it/s]
 94%|█████████▍| 10015/10682 [1:25:30<05:30,  2.02it/s]
 94%|█████████▍| 10016/10682 [1:25:31<05:29,  2.02it/s]
 94%|█████████▍| 10017/10682 [1:25:31<05:28,  2.02it/s]
 94%|█████████▍| 10018/10682 [1:25:32<05:28,  2.02it/s]
 94%|█████████▍| 10019/10682 [1:25:32<05:27,  2.02it/s]
 94%|█████████▍| 10020/10682 [1:25:33<05:27,  2.02it/s]
 94%|█████████▍| 10021/10682 [1:25:33<05:26,  2.02it/s]
 94%|█████████▍| 10022/10682 [1:25:34<05:26,  2.02it/s]
 94%|█████████▍| 10023/10682 [1:25:34<05:25,  2.02it/s]
 94%|█████████▍| 10024/10682 [1:25:35<05:25,  2.02it/s]
 94%|█████████▍| 10025/10682 [1:25:35<05:25,  2.02it/s]{'loss': 2.4108, 'grad_norm': 0.2646719515323639, 'learning_rate': 1.1481100210606388e-05, 'epoch': 13.13}
+
                                                       
 94%|█████████▍| 10025/10682 [1:25:35<05:25,  2.02it/s]
 94%|█████████▍| 10026/10682 [1:25:36<05:25,  2.02it/s]
 94%|█████████▍| 10027/10682 [1:25:36<05:24,  2.02it/s]
 94%|█████████▍| 10028/10682 [1:25:37<05:23,  2.02it/s]
 94%|█████████▍| 10029/10682 [1:25:37<05:23,  2.02it/s]
 94%|█████████▍| 10030/10682 [1:25:38<05:23,  2.02it/s]
 94%|█████████▍| 10031/10682 [1:25:38<05:22,  2.02it/s]
 94%|█████████▍| 10032/10682 [1:25:39<05:21,  2.02it/s]
 94%|█████████▍| 10033/10682 [1:25:39<05:21,  2.02it/s]
 94%|█████████▍| 10034/10682 [1:25:39<05:20,  2.02it/s]
 94%|█████████▍| 10035/10682 [1:25:40<05:20,  2.02it/s]
 94%|█████████▍| 10036/10682 [1:25:40<05:19,  2.02it/s]
 94%|█████████▍| 10037/10682 [1:25:41<05:19,  2.02it/s]
 94%|█████████▍| 10038/10682 [1:25:41<05:18,  2.02it/s]
 94%|█████████▍| 10039/10682 [1:25:42<05:18,  2.02it/s]
 94%|█████████▍| 10040/10682 [1:25:42<05:17,  2.02it/s]
 94%|█████████▍| 10041/10682 [1:25:43<05:17,  2.02it/s]
 94%|█████████▍| 10042/10682 [1:25:43<05:16,  2.02it/s]
 94%|█████████▍| 10043/10682 [1:25:44<05:16,  2.02it/s]
 94%|█████████▍| 10044/10682 [1:25:44<05:15,  2.02it/s]
 94%|█████████▍| 10045/10682 [1:25:45<05:15,  2.02it/s]
 94%|█████████▍| 10046/10682 [1:25:45<05:14,  2.02it/s]
 94%|█████████▍| 10047/10682 [1:25:46<05:14,  2.02it/s]
 94%|█████████▍| 10048/10682 [1:25:46<05:14,  2.02it/s]
 94%|█████████▍| 10049/10682 [1:25:47<05:13,  2.02it/s]
 94%|█████████▍| 10050/10682 [1:25:47<05:12,  2.02it/s]
                                                       
+
 94%|█████████▍| 10050/10682 [1:25:47<05:12,  2.02it/s]
 94%|█████████▍| 10051/10682 [1:25:48<05:12,  2.02it/s]
 94%|█████████▍| 10052/10682 [1:25:48<05:11,  2.02it/s]
 94%|█████████▍| 10053/10682 [1:25:49<05:11,  2.02it/s]
 94%|█████████▍| 10054/10682 [1:25:49<05:10,  2.02it/s]
 94%|█████████▍| 10055/10682 [1:25:50<05:10,  2.02it/s]
 94%|█████████▍| 10056/10682 [1:25:50<05:10,  2.02it/s]
 94%|█████████▍| 10057/10682 [1:25:51<05:09,  2.02it/s]
 94%|█████████▍| 10058/10682 [1:25:51<05:09,  2.02it/s]
 94%|█████████▍| 10059/10682 [1:25:52<05:08,  2.02it/s]
 94%|█████████▍| 10060/10682 [1:25:52<05:08,  2.02it/s]
 94%|█████████▍| 10061/10682 [1:25:53<05:07,  2.02it/s]
 94%|█████████▍| 10062/10682 [1:25:53<05:07,  2.02it/s]
 94%|█████████▍| 10063/10682 [1:25:54<05:06,  2.02it/s]
 94%|█████████▍| 10064/10682 [1:25:54<05:06,  2.02it/s]
 94%|█████████▍| 10065/10682 [1:25:55<05:05,  2.02it/s]
 94%|█████████▍| 10066/10682 [1:25:55<05:05,  2.02it/s]
 94%|█████████▍| 10067/10682 [1:25:56<05:04,  2.02it/s]
 94%|█████████▍| 10068/10682 [1:25:56<05:04,  2.02it/s]
 94%|█��███████▍| 10069/10682 [1:25:57<05:03,  2.02it/s]
 94%|█████████▍| 10070/10682 [1:25:57<05:03,  2.02it/s]
 94%|█████████▍| 10071/10682 [1:25:58<05:02,  2.02it/s]
 94%|█████████▍| 10072/10682 [1:25:58<05:02,  2.02it/s]
 94%|█████████▍| 10073/10682 [1:25:59<05:02,  2.01it/s]
 94%|█████████▍| 10074/10682 [1:25:59<05:01,  2.02it/s]
 94%|█████████▍| 10075/10682 [1:26:00<05:01,  2.02it/s]{'loss': 2.4071, 'grad_norm': 0.26652199029922485, 'learning_rate': 9.805609925895964e-06, 'epoch': 13.2}
+
                                                       
 94%|█████████▍| 10075/10682 [1:26:00<05:01,  2.02it/s]
 94%|█████████▍| 10076/10682 [1:26:00<05:00,  2.01it/s]
 94%|█████████▍| 10077/10682 [1:26:01<05:00,  2.01it/s]
 94%|█████████▍| 10078/10682 [1:26:01<04:59,  2.01it/s]
 94%|█████████▍| 10079/10682 [1:26:02<04:59,  2.02it/s]
 94%|█████████▍| 10080/10682 [1:26:02<04:58,  2.02it/s]
 94%|█████████▍| 10081/10682 [1:26:03<04:57,  2.02it/s]
 94%|█████████▍| 10082/10682 [1:26:03<04:56,  2.02it/s]
 94%|█████████▍| 10083/10682 [1:26:04<04:56,  2.02it/s]
 94%|█████████▍| 10084/10682 [1:26:04<04:56,  2.02it/s]
 94%|█████████▍| 10085/10682 [1:26:05<04:55,  2.02it/s]
 94%|█████████▍| 10086/10682 [1:26:05<04:55,  2.02it/s]
 94%|█████████▍| 10087/10682 [1:26:06<04:54,  2.02it/s]
 94%|█████████▍| 10088/10682 [1:26:06<04:54,  2.02it/s]
 94%|█████████▍| 10089/10682 [1:26:07<04:54,  2.02it/s]
 94%|█████████▍| 10090/10682 [1:26:07<04:53,  2.02it/s]
 94%|█████████▍| 10091/10682 [1:26:08<04:52,  2.02it/s]
 94%|█████████▍| 10092/10682 [1:26:08<04:52,  2.02it/s]
 94%|█████████▍| 10093/10682 [1:26:09<04:51,  2.02it/s]
 94%|█████████▍| 10094/10682 [1:26:09<04:51,  2.02it/s]
 95%|█████████▍| 10095/10682 [1:26:10<04:50,  2.02it/s]
 95%|█████████▍| 10096/10682 [1:26:10<04:50,  2.02it/s]
 95%|█████████▍| 10097/10682 [1:26:11<04:49,  2.02it/s]
 95%|█████████▍| 10098/10682 [1:26:11<04:49,  2.02it/s]
 95%|█████████▍| 10099/10682 [1:26:12<04:48,  2.02it/s]
 95%|█████████▍| 10100/10682 [1:26:12<04:48,  2.02it/s]
                                                       
+
 95%|█████████▍| 10100/10682 [1:26:12<04:48,  2.02it/s]
 95%|█████████▍| 10101/10682 [1:26:13<04:48,  2.02it/s]
 95%|█████████▍| 10102/10682 [1:26:13<04:47,  2.02it/s]
 95%|█████████▍| 10103/10682 [1:26:14<04:47,  2.02it/s]
 95%|█████████▍| 10104/10682 [1:26:14<04:46,  2.02it/s]
 95%|█████████▍| 10105/10682 [1:26:15<04:45,  2.02it/s]
 95%|█████████▍| 10106/10682 [1:26:15<04:45,  2.02it/s]
 95%|█████████▍| 10107/10682 [1:26:16<04:44,  2.02it/s]
 95%|█████████▍| 10108/10682 [1:26:16<04:44,  2.02it/s]
 95%|█████████▍| 10109/10682 [1:26:17<04:44,  2.02it/s]
 95%|█████████▍| 10110/10682 [1:26:17<04:43,  2.01it/s]
 95%|█████████▍| 10111/10682 [1:26:18<04:43,  2.02it/s]
 95%|█████████▍| 10112/10682 [1:26:18<04:42,  2.02it/s]
 95%|█████████▍| 10113/10682 [1:26:19<04:41,  2.02it/s]
 95%|█████████▍| 10114/10682 [1:26:19<04:41,  2.02it/s]
 95%|█████████▍| 10115/10682 [1:26:20<04:40,  2.02it/s]
 95%|█████████▍| 10116/10682 [1:26:20<04:40,  2.02it/s]
 95%|█████████▍| 10117/10682 [1:26:21<04:39,  2.02it/s]
 95%|█████████▍| 10118/10682 [1:26:21<04:39,  2.02it/s]
 95%|█████████▍| 10119/10682 [1:26:22<04:38,  2.02it/s]
 95%|█████████▍| 10120/10682 [1:26:22<04:38,  2.02it/s]
 95%|█████████▍| 10121/10682 [1:26:23<04:37,  2.02it/s]
 95%|█████████▍| 10122/10682 [1:26:23<04:37,  2.02it/s]
 95%|█████████▍| 10123/10682 [1:26:24<04:36,  2.02it/s]
 95%|█████████▍| 10124/10682 [1:26:24<04:36,  2.02it/s]
 95%|█████████▍| 10125/10682 [1:26:25<04:36,  2.02it/s]{'loss': 2.4062, 'grad_norm': 0.2640628218650818, 'learning_rate': 8.261001828055447e-06, 'epoch': 13.26}
+
                                                       
 95%|█████████▍| 10125/10682 [1:26:25<04:36,  2.02it/s]
 95%|█████████▍| 10126/10682 [1:26:25<04:35,  2.02it/s]
 95%|█████████▍| 10127/10682 [1:26:26<04:35,  2.02it/s]
 95%|█████████▍| 10128/10682 [1:26:26<04:34,  2.02it/s]
 95%|█████████▍| 10129/10682 [1:26:27<04:34,  2.02it/s]
 95%|█████████▍| 10130/10682 [1:26:27<04:33,  2.02it/s]
 95%|█████████▍| 10131/10682 [1:26:28<04:33,  2.02it/s]
 95%|█████████▍| 10132/10682 [1:26:28<04:32,  2.02it/s]
 95%|█████████▍| 10133/10682 [1:26:29<04:32,  2.02it/s]
 95%|█████████▍| 10134/10682 [1:26:29<04:31,  2.02it/s]
 95%|█████████▍| 10135/10682 [1:26:30<04:31,  2.02it/s]
 95%|█████████▍| 10136/10682 [1:26:30<04:30,  2.01it/s]
 95%|█████████▍| 10137/10682 [1:26:31<04:30,  2.02it/s]
 95%|█████████▍| 10138/10682 [1:26:31<04:29,  2.02it/s]
 95%|█████████▍| 10139/10682 [1:26:32<04:29,  2.02it/s]
 95%|█████████▍| 10140/10682 [1:26:32<04:28,  2.02it/s]
 95%|█████████▍| 10141/10682 [1:26:33<04:28,  2.02it/s]
 95%|█████████▍| 10142/10682 [1:26:33<04:27,  2.02it/s]
 95%|█████████▍| 10143/10682 [1:26:34<04:27,  2.02it/s]
 95%|█████████▍| 10144/10682 [1:26:34<04:26,  2.02it/s]
 95%|█████████▍| 10145/10682 [1:26:34<04:26,  2.01it/s]
 95%|█████████▍| 10146/10682 [1:26:35<04:25,  2.02it/s]
 95%|█████████▍| 10147/10682 [1:26:35<04:25,  2.01it/s]
 95%|█████████▌| 10148/10682 [1:26:36<04:47,  1.85it/s]
 95%|█████████▌| 10149/10682 [1:26:37<04:40,  1.90it/s]
 95%|█████████▌| 10150/10682 [1:26:37<04:34,  1.93it/s]
                                                       {'loss': 2.4023, 'grad_norm': 0.2630499601364136, 'learning_rate': 7.537908845868024e-06, 'epoch': 13.29}
+
 95%|█████████▌| 10150/10682 [1:26:37<04:34,  1.93it/s]
 95%|█████████▌| 10151/10682 [1:26:38<04:31,  1.96it/s]
 95%|█████████▌| 10152/10682 [1:26:38<04:28,  1.97it/s]
 95%|█████████▌| 10153/10682 [1:26:39<04:26,  1.99it/s]
 95%|█████████▌| 10154/10682 [1:26:39<04:24,  1.99it/s]
 95%|█████████▌| 10155/10682 [1:26:40<04:23,  2.00it/s]
 95%|█████████▌| 10156/10682 [1:26:40<04:22,  2.01it/s]
 95%|█████████▌| 10157/10682 [1:26:41<04:21,  2.01it/s]
 95%|█████████▌| 10158/10682 [1:26:41<04:20,  2.01it/s]
 95%|█████████▌| 10159/10682 [1:26:42<04:20,  2.01it/s]
 95%|█████████▌| 10160/10682 [1:26:42<04:19,  2.01it/s]
 95%|█████████▌| 10161/10682 [1:26:43<04:18,  2.01it/s]
 95%|█████████▌| 10162/10682 [1:26:43<04:17,  2.02it/s]
 95%|█████████▌| 10163/10682 [1:26:44<04:17,  2.02it/s]
 95%|█████████▌| 10164/10682 [1:26:44<04:16,  2.02it/s]
 95%|█████████▌| 10165/10682 [1:26:45<04:16,  2.02it/s]
 95%|█████████▌| 10166/10682 [1:26:45<04:15,  2.02it/s]
 95%|█████████▌| 10167/10682 [1:26:46<04:14,  2.02it/s]
 95%|█████████▌| 10168/10682 [1:26:46<04:14,  2.02it/s]
 95%|█████████▌| 10169/10682 [1:26:47<04:13,  2.02it/s]
 95%|█████████▌| 10170/10682 [1:26:47<04:13,  2.02it/s]
 95%|█████████▌| 10171/10682 [1:26:48<04:12,  2.02it/s]
 95%|█████████▌| 10172/10682 [1:26:48<04:12,  2.02it/s]
 95%|█████████▌| 10173/10682 [1:26:49<04:11,  2.02it/s]
 95%|█████████▌| 10174/10682 [1:26:49<04:11,  2.02it/s]
 95%|█████████▌| 10175/10682 [1:26:50<04:10,  2.02it/s]
                                                       {'loss': 2.4103, 'grad_norm': 0.2654673159122467, 'learning_rate': 6.847688328344037e-06, 'epoch': 13.33}
+
 95%|█████████▌| 10175/10682 [1:26:50<04:10,  2.02it/s]
 95%|█████████▌| 10176/10682 [1:26:50<04:10,  2.02it/s]
 95%|█████████▌| 10177/10682 [1:26:50<04:10,  2.02it/s]
 95%|█████████▌| 10178/10682 [1:26:51<04:09,  2.02it/s]
 95%|█████████▌| 10179/10682 [1:26:51<04:09,  2.02it/s]
 95%|█████████▌| 10180/10682 [1:26:52<04:08,  2.02it/s]
 95%|█████████▌| 10181/10682 [1:26:52<04:08,  2.02it/s]
 95%|█████████▌| 10182/10682 [1:26:53<04:07,  2.02it/s]
 95%|█████████▌| 10183/10682 [1:26:53<04:07,  2.02it/s]
 95%|█████████▌| 10184/10682 [1:26:54<04:06,  2.02it/s]
 95%|█████████▌| 10185/10682 [1:26:54<04:06,  2.02it/s]
 95%|█████████▌| 10186/10682 [1:26:55<04:05,  2.02it/s]
 95%|█████████▌| 10187/10682 [1:26:55<04:05,  2.02it/s]
 95%|���████████▌| 10188/10682 [1:26:56<04:05,  2.02it/s]
 95%|█████████▌| 10189/10682 [1:26:56<04:04,  2.02it/s]
 95%|█████████▌| 10190/10682 [1:26:57<04:04,  2.01it/s]
 95%|█████████▌| 10191/10682 [1:26:57<04:03,  2.01it/s]
 95%|█████████▌| 10192/10682 [1:26:58<04:03,  2.02it/s]
 95%|█████████▌| 10193/10682 [1:26:58<04:02,  2.02it/s]
 95%|█████████▌| 10194/10682 [1:26:59<04:02,  2.01it/s]
 95%|█████████▌| 10195/10682 [1:26:59<04:01,  2.01it/s]
 95%|█████████▌| 10196/10682 [1:27:00<04:01,  2.02it/s]
 95%|█████████▌| 10197/10682 [1:27:00<04:00,  2.02it/s]
 95%|█████████▌| 10198/10682 [1:27:01<04:00,  2.01it/s]
 95%|█████████▌| 10199/10682 [1:27:01<03:59,  2.01it/s]
 95%|█████████▌| 10200/10682 [1:27:02<03:59,  2.02it/s]
                                                       
+
 95%|█████████▌| 10200/10682 [1:27:02<03:59,  2.02it/s]
 95%|█████████▌| 10201/10682 [1:27:02<03:58,  2.01it/s]
 96%|█████████▌| 10202/10682 [1:27:03<03:58,  2.02it/s]
 96%|█████████▌| 10203/10682 [1:27:03<03:57,  2.02it/s]
 96%|█████████▌| 10204/10682 [1:27:04<03:56,  2.02it/s]
 96%|█████████▌| 10205/10682 [1:27:04<03:56,  2.02it/s]
 96%|█████████▌| 10206/10682 [1:27:05<03:56,  2.01it/s]
 96%|█████████▌| 10207/10682 [1:27:05<03:55,  2.02it/s]
 96%|█████████▌| 10208/10682 [1:27:06<03:55,  2.02it/s]
 96%|█████████▌| 10209/10682 [1:27:06<03:54,  2.02it/s]
 96%|█████████▌| 10210/10682 [1:27:07<03:54,  2.01it/s]
 96%|█████████▌| 10211/10682 [1:27:07<03:53,  2.02it/s]
 96%|█████████▌| 10212/10682 [1:27:08<03:52,  2.02it/s]
 96%|█████████▌| 10213/10682 [1:27:08<03:52,  2.02it/s]
 96%|█████████▌| 10214/10682 [1:27:09<03:52,  2.02it/s]
 96%|█████████▌| 10215/10682 [1:27:09<03:51,  2.02it/s]
 96%|█████████▌| 10216/10682 [1:27:10<03:50,  2.02it/s]
 96%|█████████▌| 10217/10682 [1:27:10<03:50,  2.02it/s]
 96%|█████████▌| 10218/10682 [1:27:11<03:50,  2.02it/s]
 96%|█████████▌| 10219/10682 [1:27:11<03:50,  2.01it/s]
 96%|█████████▌| 10220/10682 [1:27:12<03:49,  2.01it/s]
 96%|█████████▌| 10221/10682 [1:27:12<03:48,  2.01it/s]
 96%|█████████▌| 10222/10682 [1:27:13<03:48,  2.01it/s]
 96%|█████████▌| 10223/10682 [1:27:13<03:47,  2.02it/s]
 96%|█████████▌| 10224/10682 [1:27:14<03:47,  2.01it/s]
 96%|█████████▌| 10225/10682 [1:27:14<03:46,  2.02it/s]{'loss': 2.4091, 'grad_norm': 0.2631201446056366, 'learning_rate': 5.56604678228706e-06, 'epoch': 13.39}
+
                                                       
 96%|█████████▌| 10225/10682 [1:27:14<03:46,  2.02it/s]
 96%|█████████▌| 10226/10682 [1:27:15<03:46,  2.01it/s]
 96%|█████████▌| 10227/10682 [1:27:15<03:45,  2.01it/s]
 96%|█████████▌| 10228/10682 [1:27:16<03:45,  2.02it/s]
 96%|█████████▌| 10229/10682 [1:27:16<03:44,  2.02it/s]
 96%|█████████▌| 10230/10682 [1:27:17<03:44,  2.02it/s]
 96%|█████████▌| 10231/10682 [1:27:17<03:43,  2.02it/s]
 96%|█████████▌| 10232/10682 [1:27:18<03:42,  2.02it/s]
 96%|█████████▌| 10233/10682 [1:27:18<03:42,  2.02it/s]
 96%|█████████▌| 10234/10682 [1:27:19<03:42,  2.02it/s]
 96%|█████████▌| 10235/10682 [1:27:19<03:41,  2.02it/s]
 96%|█████████▌| 10236/10682 [1:27:20<03:41,  2.02it/s]
 96%|█████████▌| 10237/10682 [1:27:20<03:40,  2.02it/s]
 96%|█████████▌| 10238/10682 [1:27:21<03:40,  2.02it/s]
 96%|█████████▌| 10239/10682 [1:27:21<03:39,  2.02it/s]
 96%|█████████▌| 10240/10682 [1:27:22<03:39,  2.02it/s]
 96%|█████████▌| 10241/10682 [1:27:22<03:38,  2.01it/s]
 96%|█████████▌| 10242/10682 [1:27:23<03:38,  2.02it/s]
 96%|█████████▌| 10243/10682 [1:27:23<03:37,  2.02it/s]
 96%|█████████▌| 10244/10682 [1:27:24<03:37,  2.02it/s]
 96%|█████████▌| 10245/10682 [1:27:24<03:36,  2.02it/s]
 96%|█████████▌| 10246/10682 [1:27:25<03:35,  2.02it/s]
 96%|█████████▌| 10247/10682 [1:27:25<03:35,  2.02it/s]
 96%|█████████▌| 10248/10682 [1:27:26<03:34,  2.02it/s]
 96%|█████████▌| 10249/10682 [1:27:26<03:34,  2.02it/s]
 96%|█████████▌| 10250/10682 [1:27:27<03:34,  2.02it/s]{'loss': 2.408, 'grad_norm': 0.2649598717689514, 'learning_rate': 4.974711304941093e-06, 'epoch': 13.43}
+
                                                       
 96%|█████████▌| 10250/10682 [1:27:27<03:34,  2.02it/s]
 96%|█████████▌| 10251/10682 [1:27:27<03:34,  2.01it/s]
 96%|█████████▌| 10252/10682 [1:27:28<03:33,  2.01it/s]
 96%|█████████▌| 10253/10682 [1:27:28<03:51,  1.86it/s]
 96%|█████████▌| 10254/10682 [1:27:29<03:45,  1.90it/s]
 96%|█████████▌| 10255/10682 [1:27:29<03:41,  1.93it/s]
 96%|█████████▌| 10256/10682 [1:27:30<03:37,  1.96it/s]
 96%|█████████▌| 10257/10682 [1:27:30<03:35,  1.97it/s]
 96%|█████████▌| 10258/10682 [1:27:31<03:33,  1.99it/s]
 96%|█████████▌| 10259/10682 [1:27:31<03:32,  1.99it/s]
 96%|█████████▌| 10260/10682 [1:27:32<03:30,  2.00it/s]
 96%|█████████▌| 10261/10682 [1:27:32<03:30,  2.00it/s]
 96%|█████████▌| 10262/10682 [1:27:33<03:29,  2.01it/s]
 96%|█████████▌| 10263/10682 [1:27:33<03:28,  2.01it/s]
 96%|█████████▌| 10264/10682 [1:27:34<03:27,  2.01it/s]
 96%|█████████▌| 10265/10682 [1:27:34<03:27,  2.01it/s]
 96%|█████████▌| 10266/10682 [1:27:35<03:26,  2.02it/s]
 96%|█████████▌| 10267/10682 [1:27:35<03:25,  2.02it/s]
 96%|█████████▌| 10268/10682 [1:27:36<03:25,  2.02it/s]
 96%|█████████▌| 10269/10682 [1:27:36<03:24,  2.02it/s]
 96%|█████████▌| 10270/10682 [1:27:37<03:24,  2.02it/s]
 96%|█████████▌| 10271/10682 [1:27:37<03:23,  2.02it/s]
 96%|█████████▌| 10272/10682 [1:27:38<03:22,  2.02it/s]
 96%|█████████▌| 10273/10682 [1:27:38<03:22,  2.02it/s]
 96%|█████████▌| 10274/10682 [1:27:39<03:21,  2.02it/s]
 96%|█████████▌| 10275/10682 [1:27:39<03:21,  2.02it/s]{'loss': 2.3975, 'grad_norm': 0.2637172341346741, 'learning_rate': 4.416419388921844e-06, 'epoch': 13.46}
+
                                                       
 96%|█████████▌| 10275/10682 [1:27:39<03:21,  2.02it/s]
 96%|█████████▌| 10276/10682 [1:27:40<03:21,  2.02it/s]
 96%|█████████▌| 10277/10682 [1:27:40<03:20,  2.02it/s]
 96%|█████████▌| 10278/10682 [1:27:41<03:20,  2.02it/s]
 96%|█████████▌| 10279/10682 [1:27:41<03:19,  2.02it/s]
 96%|█████████▌| 10280/10682 [1:27:42<03:19,  2.02it/s]
 96%|█████████▌| 10281/10682 [1:27:42<03:18,  2.02it/s]
 96%|█████████▋| 10282/10682 [1:27:43<03:18,  2.02it/s]
 96%|█████████▋| 10283/10682 [1:27:43<03:17,  2.02it/s]
 96%|█████████▋| 10284/10682 [1:27:44<03:17,  2.02it/s]
 96%|█████████▋| 10285/10682 [1:27:44<03:16,  2.02it/s]
 96%|█████████▋| 10286/10682 [1:27:45<03:16,  2.02it/s]
 96%|█████████▋| 10287/10682 [1:27:45<03:15,  2.02it/s]
 96%|█████████▋| 10288/10682 [1:27:46<03:15,  2.02it/s]
 96%|█████████▋| 10289/10682 [1:27:46<03:14,  2.02it/s]
 96%|█████████▋| 10290/10682 [1:27:47<03:13,  2.02it/s]
 96%|█████████▋| 10291/10682 [1:27:47<03:13,  2.02it/s]
 96%|█████████▋| 10292/10682 [1:27:48<03:12,  2.02it/s]
 96%|█████████▋| 10293/10682 [1:27:48<03:12,  2.02it/s]
 96%|█████████▋| 10294/10682 [1:27:49<03:11,  2.02it/s]
 96%|█████████▋| 10295/10682 [1:27:49<03:11,  2.02it/s]
 96%|█████████▋| 10296/10682 [1:27:50<03:11,  2.02it/s]
 96%|█████████▋| 10297/10682 [1:27:50<03:11,  2.02it/s]
 96%|█████████▋| 10298/10682 [1:27:51<03:10,  2.02it/s]
 96%|█████████▋| 10299/10682 [1:27:51<03:10,  2.02it/s]
 96%|█████████▋| 10300/10682 [1:27:52<03:09,  2.02it/s]{'loss': 2.4022, 'grad_norm': 0.26662492752075195, 'learning_rate': 3.891208300917604e-06, 'epoch': 13.49}
+
                                                       
 96%|█████████▋| 10300/10682 [1:27:52<03:09,  2.02it/s]
 96%|█████████▋| 10301/10682 [1:27:52<03:09,  2.02it/s]
 96%|█████████▋| 10302/10682 [1:27:53<03:08,  2.01it/s]
 96%|█████████▋| 10303/10682 [1:27:53<03:08,  2.02it/s]
 96%|█████████▋| 10304/10682 [1:27:54<03:07,  2.01it/s]
 96%|█████████▋| 10305/10682 [1:27:54<03:06,  2.02it/s]
 96%|█████████▋| 10306/10682 [1:27:55<03:06,  2.02it/s]
 96%|█████████▋| 10307/10682 [1:27:55<03:05,  2.02it/s]
 96%|█████████▋| 10308/10682 [1:27:56<03:04,  2.02it/s]
 97%|█████████▋| 10309/10682 [1:27:56<03:04,  2.02it/s]
 97%|█████████▋| 10310/10682 [1:27:57<03:04,  2.02it/s]
 97%|█████████▋| 10311/10682 [1:27:57<03:03,  2.02it/s]
 97%|█████████▋| 10312/10682 [1:27:58<03:03,  2.02it/s]
 97%|█████████▋| 10313/10682 [1:27:58<03:03,  2.01it/s]
 97%|█████████▋| 10314/10682 [1:27:59<03:02,  2.02it/s]
 97%|█████████▋| 10315/10682 [1:27:59<03:02,  2.02it/s]
 97%|█████████▋| 10316/10682 [1:28:00<03:01,  2.02it/s]
 97%|█████████▋| 10317/10682 [1:28:00<03:00,  2.02it/s]
 97%|█████████▋| 10318/10682 [1:28:01<03:00,  2.02it/s]
 97%|█████████▋| 10319/10682 [1:28:01<02:59,  2.02it/s]
 97%|█████████▋| 10320/10682 [1:28:02<02:59,  2.02it/s]
 97%|█████████▋| 10321/10682 [1:28:02<02:58,  2.02it/s]
 97%|█████████▋| 10322/10682 [1:28:03<02:58,  2.02it/s]
 97%|█████████▋| 10323/10682 [1:28:03<02:58,  2.01it/s]
 97%|█████████▋| 10324/10682 [1:28:04<02:58,  2.01it/s]
 97%|█████████▋| 10325/10682 [1:28:04<02:57,  2.01it/s]{'loss': 2.4181, 'grad_norm': 0.2648935317993164, 'learning_rate': 3.3991130994299734e-06, 'epoch': 13.52}
+
                                                       
 97%|█████████▋| 10325/10682 [1:28:04<02:57,  2.01it/s]
 97%|█████████▋| 10326/10682 [1:28:05<02:56,  2.01it/s]
 97%|█████████▋| 10327/10682 [1:28:05<02:56,  2.02it/s]
 97%|█████████▋| 10328/10682 [1:28:06<02:55,  2.02it/s]
 97%|█████████▋| 10329/10682 [1:28:06<02:54,  2.02it/s]
 97%|█████████▋| 10330/10682 [1:28:06<02:54,  2.02it/s]
 97%|█████████▋| 10331/10682 [1:28:07<02:53,  2.02it/s]
 97%|█████████▋| 10332/10682 [1:28:07<02:53,  2.02it/s]
 97%|█████████▋| 10333/10682 [1:28:08<02:52,  2.02it/s]
 97%|█████████▋| 10334/10682 [1:28:08<02:52,  2.02it/s]
 97%|█████████▋| 10335/10682 [1:28:09<02:51,  2.02it/s]
 97%|█████████▋| 10336/10682 [1:28:09<02:51,  2.02it/s]
 97%|█████████▋| 10337/10682 [1:28:10<02:50,  2.02it/s]
 97%|█████████▋| 10338/10682 [1:28:10<02:50,  2.02it/s]
 97%|█████████▋| 10339/10682 [1:28:11<02:49,  2.02it/s]
 97%|█████████▋| 10340/10682 [1:28:11<02:49,  2.02it/s]
 97%|█████████▋| 10341/10682 [1:28:12<02:48,  2.03it/s]
 97%|█████████▋| 10342/10682 [1:28:12<02:48,  2.02it/s]
 97%|█████████▋| 10343/10682 [1:28:13<02:47,  2.02it/s]
 97%|█████████▋| 10344/10682 [1:28:13<02:47,  2.02it/s]
 97%|█████████▋| 10345/10682 [1:28:14<02:46,  2.02it/s]
 97%|█████████▋| 10346/10682 [1:28:14<02:46,  2.02it/s]
 97%|█████████▋| 10347/10682 [1:28:15<02:45,  2.02it/s]
 97%|█████████▋| 10348/10682 [1:28:15<02:45,  2.02it/s]
 97%|█████████▋| 10349/10682 [1:28:16<02:44,  2.02it/s]
 97%|█████████▋| 10350/10682 [1:28:16<02:44,  2.02it/s]
                                                       
+
 97%|█████████▋| 10350/10682 [1:28:16<02:44,  2.02it/s]
 97%|█████████▋| 10351/10682 [1:28:17<02:43,  2.02it/s]
 97%|█████████▋| 10352/10682 [1:28:17<02:43,  2.02it/s]
 97%|█████████▋| 10353/10682 [1:28:18<02:42,  2.02it/s]
 97%|█████████▋| 10354/10682 [1:28:18<02:42,  2.02it/s]
 97%|█████████▋| 10355/10682 [1:28:19<02:41,  2.02it/s]
 97%|█████████▋| 10356/10682 [1:28:19<02:41,  2.02it/s]
 97%|█████████▋| 10357/10682 [1:28:20<02:40,  2.02it/s]
 97%|█████████▋| 10358/10682 [1:28:20<02:39,  2.03it/s]
 97%|█████████▋| 10359/10682 [1:28:21<02:39,  2.02it/s]
 97%|█████████▋| 10360/10682 [1:28:21<02:38,  2.03it/s]
 97%|█████████▋| 10361/10682 [1:28:22<02:38,  2.02it/s]
 97%|█████████▋| 10362/10682 [1:28:22<02:37,  2.03it/s]
 97%|█████████▋| 10363/10682 [1:28:23<02:37,  2.02it/s]
 97%|█████████▋| 10364/10682 [1:28:23<02:37,  2.02it/s]
 97%|█████████▋| 10365/10682 [1:28:24<02:36,  2.02it/s]
 97%|█████████▋| 10366/10682 [1:28:24<02:36,  2.03it/s]
 97%|█████████▋| 10367/10682 [1:28:25<02:36,  2.02it/s]
 97%|█████████▋| 10368/10682 [1:28:25<02:35,  2.02it/s]
 97%|█████████▋| 10369/10682 [1:28:26<02:34,  2.02it/s]
 97%|█████████▋| 10370/10682 [1:28:26<02:34,  2.02it/s]
 97%|█████████▋| 10371/10682 [1:28:27<02:33,  2.02it/s]
 97%|█████████▋| 10372/10682 [1:28:27<02:33,  2.02it/s]
 97%|█████████▋| 10373/10682 [1:28:28<02:32,  2.02it/s]
 97%|█████████▋| 10374/10682 [1:28:28<02:32,  2.02it/s]
 97%|█████████▋| 10375/10682 [1:28:29<02:31,  2.02it/s]{'loss': 2.4033, 'grad_norm': 0.2658427059650421, 'learning_rate': 2.5143995351817882e-06, 'epoch': 13.59}
+
                                                       
 97%|█████████▋| 10375/10682 [1:28:29<02:31,  2.02it/s]
 97%|█████████▋| 10376/10682 [1:28:29<02:31,  2.02it/s]
 97%|█████████▋| 10377/10682 [1:28:30<02:30,  2.02it/s]
 97%|█████████▋| 10378/10682 [1:28:30<02:30,  2.02it/s]
 97%|█████████▋| 10379/10682 [1:28:31<02:29,  2.02it/s]
 97%|█████████▋| 10380/10682 [1:28:31<02:29,  2.02it/s]
 97%|█████████▋| 10381/10682 [1:28:32<02:28,  2.02it/s]
 97%|█████████▋| 10382/10682 [1:28:32<02:28,  2.02it/s]
 97%|█████████▋| 10383/10682 [1:28:33<02:27,  2.02it/s]
 97%|█████████▋| 10384/10682 [1:28:33<02:27,  2.02it/s]
 97%|█████████▋| 10385/10682 [1:28:34<02:26,  2.02it/s]
 97%|█████████▋| 10386/10682 [1:28:34<02:26,  2.02it/s]
 97%|█████████▋| 10387/10682 [1:28:35<02:25,  2.02it/s]
 97%|█████████▋| 10388/10682 [1:28:35<02:25,  2.02it/s]
 97%|█████████▋| 10389/10682 [1:28:36<02:25,  2.02it/s]
 97%|█████████▋| 10390/10682 [1:28:36<02:24,  2.02it/s]
 97%|█████████▋| 10391/10682 [1:28:37<02:24,  2.02it/s]
 97%|█████████▋| 10392/10682 [1:28:37<02:23,  2.02it/s]
 97%|█████████▋| 10393/10682 [1:28:38<02:23,  2.01it/s]
 97%|█████████▋| 10394/10682 [1:28:38<02:22,  2.02it/s]
 97%|█████████▋| 10395/10682 [1:28:39<02:22,  2.02it/s]
 97%|█████████▋| 10396/10682 [1:28:39<02:21,  2.02it/s]
 97%|█████████▋| 10397/10682 [1:28:40<02:21,  2.02it/s]
 97%|█████████▋| 10398/10682 [1:28:40<02:20,  2.02it/s]
 97%|█████████▋| 10399/10682 [1:28:41<02:20,  2.02it/s]
 97%|█████████▋| 10400/10682 [1:28:41<02:19,  2.02it/s]{'loss': 2.4157, 'grad_norm': 0.26090914011001587, 'learning_rate': 2.1218402281655835e-06, 'epoch': 13.62}
+
                                                       
 97%|█████████▋| 10400/10682 [1:28:41<02:19,  2.02it/s]
 97%|█████████▋| 10401/10682 [1:28:42<02:19,  2.02it/s]
 97%|█████████▋| 10402/10682 [1:28:42<02:18,  2.02it/s]
 97%|█████████▋| 10403/10682 [1:28:43<02:18,  2.02it/s]
 97%|█████████▋| 10404/10682 [1:28:43<02:17,  2.02it/s]
 97%|█████████▋| 10405/10682 [1:28:44<02:16,  2.02it/s]
 97%|█████████▋| 10406/10682 [1:28:44<02:16,  2.02it/s]
 97%|█████████▋| 10407/10682 [1:28:45<02:15,  2.02it/s]
 97%|█████████▋| 10408/10682 [1:28:45<02:15,  2.02it/s]
 97%|█████████▋| 10409/10682 [1:28:46<02:14,  2.02it/s]
 97%|█████████▋| 10410/10682 [1:28:46<02:14,  2.02it/s]
 97%|█████████▋| 10411/10682 [1:28:47<02:13,  2.02it/s]
 97%|█████████▋| 10412/10682 [1:28:47<02:13,  2.02it/s]
 97%|█████████▋| 10413/10682 [1:28:48<02:13,  2.02it/s]
 97%|█████████▋| 10414/10682 [1:28:48<02:12,  2.02it/s]
 98%|█████████▊| 10415/10682 [1:28:49<02:11,  2.02it/s]
 98%|█████████▊| 10416/10682 [1:28:49<02:11,  2.02it/s]
 98%|█████████▊| 10417/10682 [1:28:50<02:10,  2.02it/s]
 98%|█████████▊| 10418/10682 [1:28:50<02:10,  2.02it/s]
 98%|█████████▊| 10419/10682 [1:28:51<02:09,  2.02it/s]
 98%|█████████▊| 10420/10682 [1:28:51<02:09,  2.02it/s]
 98%|█████████▊| 10421/10682 [1:28:52<02:09,  2.02it/s]
 98%|█████████▊| 10422/10682 [1:28:52<02:08,  2.02it/s]
 98%|█████████▊| 10423/10682 [1:28:52<02:08,  2.02it/s]
 98%|█████████▊| 10424/10682 [1:28:53<02:07,  2.02it/s]
 98%|█████████▊| 10425/10682 [1:28:53<02:06,  2.02it/s]{'loss': 2.4027, 'grad_norm': 0.26506245136260986, 'learning_rate': 1.7625149152127318e-06, 'epoch': 13.65}
                                                       
+
 98%|█████████▊| 10425/10682 [1:28:53<02:06,  2.02it/s]
 98%|█████████▊| 10426/10682 [1:28:54<02:06,  2.02it/s]
 98%|█████████▊| 10427/10682 [1:28:54<02:06,  2.02it/s]
 98%|█████████▊| 10428/10682 [1:28:55<02:05,  2.02it/s]
 98%|█████████▊| 10429/10682 [1:28:55<02:05,  2.02it/s]
 98%|█████████▊| 10430/10682 [1:28:56<02:04,  2.02it/s]
 98%|█████████▊| 10431/10682 [1:28:56<02:04,  2.02it/s]
 98%|█████████▊| 10432/10682 [1:28:57<02:03,  2.02it/s]
 98%|█████████▊| 10433/10682 [1:28:57<02:03,  2.02it/s]
 98%|█████████▊| 10434/10682 [1:28:58<02:02,  2.02it/s]
 98%|█████████▊| 10435/10682 [1:28:58<02:02,  2.02it/s]
 98%|█████████▊| 10436/10682 [1:28:59<02:01,  2.02it/s]
 98%|█████████▊| 10437/10682 [1:28:59<02:01,  2.02it/s]
 98%|█████████▊| 10438/10682 [1:29:00<02:00,  2.02it/s]
 98%|█████████▊| 10439/10682 [1:29:00<02:00,  2.02it/s]
 98%|█████████▊| 10440/10682 [1:29:01<01:59,  2.02it/s]
 98%|█████████▊| 10441/10682 [1:29:01<01:59,  2.02it/s]
 98%|█████████▊| 10442/10682 [1:29:02<01:58,  2.02it/s]
 98%|█████████▊| 10443/10682 [1:29:02<01:58,  2.02it/s]
 98%|█████████▊| 10444/10682 [1:29:03<01:57,  2.02it/s]
 98%|█████████▊| 10445/10682 [1:29:03<01:57,  2.02it/s]
 98%|█████████▊| 10446/10682 [1:29:04<01:56,  2.02it/s]
 98%|█████████▊| 10447/10682 [1:29:04<01:56,  2.02it/s]
 98%|█████████▊| 10448/10682 [1:29:05<01:55,  2.02it/s]
 98%|█████████▊| 10449/10682 [1:29:05<01:55,  2.02it/s]
 98%|█████████▊| 10450/10682 [1:29:06<01:54,  2.02it/s]{'loss': 2.4052, 'grad_norm': 0.2620916962623596, 'learning_rate': 1.4364475817401635e-06, 'epoch': 13.69}
+
                                                       
 98%|█████████▊| 10450/10682 [1:29:06<01:54,  2.02it/s]
 98%|█████████▊| 10451/10682 [1:29:06<01:54,  2.02it/s]
 98%|█████████▊| 10452/10682 [1:29:07<01:53,  2.02it/s]
 98%|█████████▊| 10453/10682 [1:29:07<01:53,  2.02it/s]
 98%|█████████▊| 10454/10682 [1:29:08<01:52,  2.02it/s]
 98%|█████████▊| 10455/10682 [1:29:08<01:52,  2.02it/s]
 98%|█████████▊| 10456/10682 [1:29:09<01:51,  2.02it/s]
 98%|█████████▊| 10457/10682 [1:29:09<01:51,  2.02it/s]
 98%|█████████▊| 10458/10682 [1:29:10<01:50,  2.02it/s]
 98%|█████████▊| 10459/10682 [1:29:10<01:50,  2.02it/s]
 98%|█████████▊| 10460/10682 [1:29:11<01:49,  2.02it/s]
 98%|█████████▊| 10461/10682 [1:29:11<01:49,  2.02it/s]
 98%|█████████▊| 10462/10682 [1:29:12<01:48,  2.02it/s]
 98%|█████████▊| 10463/10682 [1:29:12<01:48,  2.03it/s]
 98%|█████████▊| 10464/10682 [1:29:13<01:47,  2.02it/s]
 98%|█████████▊| 10465/10682 [1:29:13<01:47,  2.02it/s]
 98%|█████████▊| 10466/10682 [1:29:14<01:46,  2.02it/s]
 98%|█████████▊| 10467/10682 [1:29:14<01:46,  2.02it/s]
 98%|█████████▊| 10468/10682 [1:29:15<01:45,  2.02it/s]
 98%|█████████▊| 10469/10682 [1:29:15<01:45,  2.02it/s]
 98%|█████████▊| 10470/10682 [1:29:16<01:44,  2.02it/s]
 98%|█████████▊| 10471/10682 [1:29:16<01:44,  2.02it/s]
 98%|█████████▊| 10472/10682 [1:29:17<01:44,  2.02it/s]
 98%|█████████▊| 10473/10682 [1:29:17<01:43,  2.02it/s]
 98%|█████████▊| 10474/10682 [1:29:18<01:43,  2.02it/s]
 98%|█████████▊| 10475/10682 [1:29:18<01:42,  2.02it/s]{'loss': 2.4143, 'grad_norm': 0.26500821113586426, 'learning_rate': 1.143659993153079e-06, 'epoch': 13.72}
+
                                                       
 98%|█████████▊| 10475/10682 [1:29:18<01:42,  2.02it/s]
 98%|█████████▊| 10476/10682 [1:29:19<01:42,  2.01it/s]
 98%|█████████▊| 10477/10682 [1:29:19<01:41,  2.01it/s]
 98%|█████████▊| 10478/10682 [1:29:20<01:41,  2.02it/s]
 98%|█████████▊| 10479/10682 [1:29:20<01:40,  2.02it/s]
 98%|█████████▊| 10480/10682 [1:29:21<01:39,  2.02it/s]
 98%|█████████▊| 10481/10682 [1:29:21<01:39,  2.02it/s]
 98%|█████████▊| 10482/10682 [1:29:22<01:39,  2.02it/s]
 98%|█████████▊| 10483/10682 [1:29:22<01:38,  2.02it/s]
 98%|█████████▊| 10484/10682 [1:29:23<01:38,  2.02it/s]
 98%|█████████▊| 10485/10682 [1:29:23<01:37,  2.02it/s]
 98%|█████████▊| 10486/10682 [1:29:24<01:37,  2.02it/s]
 98%|█████████▊| 10487/10682 [1:29:24<01:36,  2.02it/s]
 98%|█████████▊| 10488/10682 [1:29:25<01:36,  2.02it/s]
 98%|█████████▊| 10489/10682 [1:29:25<01:35,  2.02it/s]
 98%|█████████▊| 10490/10682 [1:29:26<01:35,  2.02it/s]
 98%|█████████▊| 10491/10682 [1:29:26<01:34,  2.02it/s]
 98%|█████████▊| 10492/10682 [1:29:27<01:34,  2.02it/s]
 98%|█████████▊| 10493/10682 [1:29:27<01:33,  2.02it/s]
 98%|█████████▊| 10494/10682 [1:29:28<01:33,  2.02it/s]
 98%|█████████▊| 10495/10682 [1:29:28<01:32,  2.02it/s]
 98%|█████████▊| 10496/10682 [1:29:29<01:31,  2.02it/s]
 98%|█████████▊| 10497/10682 [1:29:29<01:31,  2.02it/s]
 98%|█████████▊| 10498/10682 [1:29:30<01:30,  2.02it/s]
 98%|█████████▊| 10499/10682 [1:29:30<01:30,  2.02it/s]
 98%|█████████▊| 10500/10682 [1:29:31<01:29,  2.02it/s]{'loss': 2.4106, 'grad_norm': 0.26642704010009766, 'learning_rate': 8.841716933915555e-07, 'epoch': 13.75}
+
                                                       
 98%|█████████▊| 10500/10682 [1:29:31<01:29,  2.02it/s]
 98%|█████████▊| 10501/10682 [1:29:31<01:29,  2.02it/s]
 98%|█████████▊| 10502/10682 [1:29:32<01:29,  2.02it/s]
 98%|█████████▊| 10503/10682 [1:29:32<01:28,  2.02it/s]
 98%|█████████▊| 10504/10682 [1:29:33<01:28,  2.02it/s]
 98%|█████████▊| 10505/10682 [1:29:33<01:27,  2.02it/s]
 98%|█████████▊| 10506/10682 [1:29:34<01:27,  2.02it/s]
 98%|█████████▊| 10507/10682 [1:29:34<01:26,  2.02it/s]
 98%|█████████▊| 10508/10682 [1:29:35<01:26,  2.02it/s]
 98%|█████████▊| 10509/10682 [1:29:35<01:25,  2.02it/s]
 98%|█████████▊| 10510/10682 [1:29:36<01:25,  2.02it/s]
 98%|█████████▊| 10511/10682 [1:29:36<01:24,  2.02it/s]
 98%|█████████▊| 10512/10682 [1:29:37<01:24,  2.02it/s]
 98%|█████████▊| 10513/10682 [1:29:37<01:23,  2.02it/s]
 98%|█████████▊| 10514/10682 [1:29:38<01:23,  2.02it/s]
 98%|█████████▊| 10515/10682 [1:29:38<01:22,  2.02it/s]
 98%|█████████▊| 10516/10682 [1:29:39<01:22,  2.02it/s]
 98%|█████████▊| 10517/10682 [1:29:39<01:21,  2.02it/s]
 98%|█████████▊| 10518/10682 [1:29:40<01:21,  2.02it/s]
 98%|█████████▊| 10519/10682 [1:29:40<01:20,  2.02it/s]
 98%|█████████▊| 10520/10682 [1:29:40<01:20,  2.02it/s]
 98%|█████████▊| 10521/10682 [1:29:41<01:19,  2.02it/s]
 99%|█████████▊| 10522/10682 [1:29:41<01:19,  2.02it/s]
 99%|█████████▊| 10523/10682 [1:29:42<01:18,  2.02it/s]
 99%|█████████▊| 10524/10682 [1:29:42<01:18,  2.02it/s]
 99%|█████████▊| 10525/10682 [1:29:43<01:17,  2.02it/s]{'loss': 2.4203, 'grad_norm': 0.2669544219970703, 'learning_rate': 6.580000036264244e-07, 'epoch': 13.79}
+
                                                       
 99%|█████████▊| 10525/10682 [1:29:43<01:17,  2.02it/s]
 99%|█████████▊| 10526/10682 [1:29:43<01:17,  2.02it/s]
 99%|█████████▊| 10527/10682 [1:29:44<01:16,  2.02it/s]
 99%|█████████▊| 10528/10682 [1:29:44<01:16,  2.02it/s]
 99%|█████████▊| 10529/10682 [1:29:45<01:15,  2.02it/s]
 99%|█████████▊| 10530/10682 [1:29:45<01:15,  2.02it/s]
 99%|█████████▊| 10531/10682 [1:29:46<01:14,  2.02it/s]
 99%|█████████▊| 10532/10682 [1:29:46<01:14,  2.02it/s]
 99%|█████████▊| 10533/10682 [1:29:47<01:13,  2.02it/s]
 99%|█████████▊| 10534/10682 [1:29:47<01:13,  2.02it/s]
 99%|█████████▊| 10535/10682 [1:29:48<01:12,  2.02it/s]
 99%|█████████▊| 10536/10682 [1:29:48<01:12,  2.02it/s]
 99%|█████████▊| 10537/10682 [1:29:49<01:11,  2.02it/s]
 99%|█████████▊| 10538/10682 [1:29:49<01:11,  2.02it/s]
 99%|█████████▊| 10539/10682 [1:29:50<01:10,  2.02it/s]
 99%|█████████▊| 10540/10682 [1:29:50<01:10,  2.02it/s]
 99%|█████████▊| 10541/10682 [1:29:51<01:09,  2.02it/s]
 99%|█████████▊| 10542/10682 [1:29:51<01:09,  2.02it/s]
 99%|█████████▊| 10543/10682 [1:29:52<01:08,  2.02it/s]
 99%|█████████▊| 10544/10682 [1:29:52<01:08,  2.02it/s]
 99%|█████████▊| 10545/10682 [1:29:53<01:07,  2.02it/s]
 99%|█████████▊| 10546/10682 [1:29:53<01:07,  2.02it/s]
 99%|█████████▊| 10547/10682 [1:29:54<01:06,  2.02it/s]
 99%|█████████▊| 10548/10682 [1:29:54<01:06,  2.02it/s]
 99%|█████████▉| 10549/10682 [1:29:55<01:05,  2.02it/s]
 99%|█████████▉| 10550/10682 [1:29:55<01:05,  2.02it/s]{'loss': 2.3904, 'grad_norm': 0.2659035623073578, 'learning_rate': 4.651600211027507e-07, 'epoch': 13.82}
+
                                                       
 99%|█████████▉| 10550/10682 [1:29:55<01:05,  2.02it/s]
 99%|█████████▉| 10551/10682 [1:29:56<01:05,  2.01it/s]
 99%|█████████▉| 10552/10682 [1:29:56<01:04,  2.02it/s]
 99%|█████████▉| 10553/10682 [1:29:57<01:03,  2.02it/s]
 99%|█████████▉| 10554/10682 [1:29:57<01:03,  2.02it/s]
 99%|█████████▉| 10555/10682 [1:29:58<01:02,  2.02it/s]
 99%|█████████▉| 10556/10682 [1:29:58<01:02,  2.02it/s]
 99%|█████████▉| 10557/10682 [1:29:59<01:01,  2.02it/s]
 99%|█████████▉| 10558/10682 [1:29:59<01:01,  2.02it/s]
 99%|█████████▉| 10559/10682 [1:30:00<01:00,  2.02it/s]
 99%|█████████▉| 10560/10682 [1:30:00<01:00,  2.02it/s]
 99%|█████████▉| 10561/10682 [1:30:01<00:59,  2.02it/s]
 99%|█████████▉| 10562/10682 [1:30:01<00:59,  2.02it/s]
 99%|█████████▉| 10563/10682 [1:30:02<00:58,  2.02it/s]
 99%|█████████▉| 10564/10682 [1:30:02<00:58,  2.02it/s]
 99%|█████████▉| 10565/10682 [1:30:03<00:57,  2.02it/s]
 99%|█████████▉| 10566/10682 [1:30:03<00:57,  2.02it/s]
 99%|█████████▉| 10567/10682 [1:30:04<00:56,  2.02it/s]
 99%|█████████▉| 10568/10682 [1:30:04<00:56,  2.02it/s]
 99%|█████████▉| 10569/10682 [1:30:05<00:55,  2.02it/s]
 99%|█████████▉| 10570/10682 [1:30:05<00:55,  2.02it/s]
 99%|█████████▉| 10571/10682 [1:30:06<00:54,  2.02it/s]
 99%|█████████▉| 10572/10682 [1:30:06<00:54,  2.02it/s]
 99%|█████████▉| 10573/10682 [1:30:07<00:53,  2.02it/s]
 99%|█████████▉| 10574/10682 [1:30:07<00:53,  2.02it/s]
 99%|█████████▉| 10575/10682 [1:30:08<00:52,  2.02it/s]{'loss': 2.413, 'grad_norm': 0.2652822732925415, 'learning_rate': 3.0566461813213986e-07, 'epoch': 13.85}
+
                                                       
 99%|█████████▉| 10575/10682 [1:30:08<00:52,  2.02it/s]
 99%|█████████▉| 10576/10682 [1:30:08<00:52,  2.02it/s]
 99%|█████████▉| 10577/10682 [1:30:09<00:51,  2.02it/s]
 99%|█████████▉| 10578/10682 [1:30:09<00:51,  2.02it/s]
 99%|█████████▉| 10579/10682 [1:30:10<00:50,  2.02it/s]
 99%|█████████▉| 10580/10682 [1:30:10<00:50,  2.02it/s]
 99%|█████████▉| 10581/10682 [1:30:11<00:50,  2.02it/s]
 99%|█████████▉| 10582/10682 [1:30:11<00:49,  2.02it/s]
 99%|█████████▉| 10583/10682 [1:30:12<00:48,  2.02it/s]
 99%|█████████▉| 10584/10682 [1:30:12<00:48,  2.02it/s]
 99%|█████████▉| 10585/10682 [1:30:13<00:48,  2.02it/s]
 99%|█████████▉| 10586/10682 [1:30:13<00:47,  2.02it/s]
 99%|█████████▉| 10587/10682 [1:30:14<00:47,  2.02it/s]
 99%|█████████▉| 10588/10682 [1:30:14<00:46,  2.02it/s]
 99%|█████████▉| 10589/10682 [1:30:15<00:46,  2.02it/s]
 99%|█████████▉| 10590/10682 [1:30:15<00:45,  2.02it/s]
 99%|█████████▉| 10591/10682 [1:30:16<00:45,  2.02it/s]
 99%|█████████▉| 10592/10682 [1:30:16<00:44,  2.02it/s]
 99%|█████████▉| 10593/10682 [1:30:17<00:44,  2.02it/s]
 99%|█████████▉| 10594/10682 [1:30:17<00:43,  2.02it/s]
 99%|█████████▉| 10595/10682 [1:30:18<00:43,  2.02it/s]
 99%|█████████▉| 10596/10682 [1:30:18<00:42,  2.02it/s]
 99%|█████████▉| 10597/10682 [1:30:19<00:42,  2.02it/s]
 99%|█████████▉| 10598/10682 [1:30:19<00:41,  2.02it/s]
 99%|█████████▉| 10599/10682 [1:30:20<00:41,  2.02it/s]
 99%|█████████▉| 10600/10682 [1:30:20<00:40,  2.02it/s]{'loss': 2.4186, 'grad_norm': 0.2603202164173126, 'learning_rate': 1.7952444123359167e-07, 'epoch': 13.88}
+
                                                       
 99%|█████████▉| 10600/10682 [1:30:20<00:40,  2.02it/s]
 99%|█████████▉| 10601/10682 [1:30:21<00:40,  2.02it/s]
 99%|█████████▉| 10602/10682 [1:30:21<00:39,  2.02it/s]
 99%|█████████▉| 10603/10682 [1:30:22<00:39,  2.02it/s]
 99%|█████████▉| 10604/10682 [1:30:22<00:38,  2.02it/s]
 99%|█████████▉| 10605/10682 [1:30:23<00:38,  2.02it/s]
 99%|█████████▉| 10606/10682 [1:30:23<00:37,  2.02it/s]
 99%|█████████▉| 10607/10682 [1:30:24<00:37,  2.02it/s]
 99%|█████████▉| 10608/10682 [1:30:24<00:36,  2.02it/s]
 99%|█████████▉| 10609/10682 [1:30:25<00:36,  2.02it/s]
 99%|█████████▉| 10610/10682 [1:30:25<00:35,  2.02it/s]
 99%|█████████▉| 10611/10682 [1:30:26<00:35,  2.02it/s]
 99%|█████████▉| 10612/10682 [1:30:26<00:34,  2.02it/s]
 99%|█████████▉| 10613/10682 [1:30:27<00:34,  2.02it/s]
 99%|█████████▉| 10614/10682 [1:30:27<00:33,  2.02it/s]
 99%|█████████▉| 10615/10682 [1:30:28<00:33,  2.02it/s]
 99%|█████████▉| 10616/10682 [1:30:28<00:32,  2.02it/s]
 99%|█████████▉| 10617/10682 [1:30:29<00:32,  2.02it/s]
 99%|█████████▉| 10618/10682 [1:30:29<00:31,  2.02it/s]
 99%|█████████▉| 10619/10682 [1:30:30<00:31,  2.02it/s]
 99%|█████████▉| 10620/10682 [1:30:30<00:30,  2.02it/s]
 99%|█████████▉| 10621/10682 [1:30:31<00:30,  2.02it/s]
 99%|█████████▉| 10622/10682 [1:30:31<00:29,  2.02it/s]
 99%|█████████▉| 10623/10682 [1:30:31<00:29,  2.02it/s]
 99%|█████████▉| 10624/10682 [1:30:32<00:28,  2.02it/s]
 99%|█████████▉| 10625/10682 [1:30:32<00:28,  2.02it/s]
                                                       
+
 99%|█████████▉| 10625/10682 [1:30:32<00:28,  2.02it/s]
 99%|█████████▉| 10626/10682 [1:30:33<00:27,  2.01it/s]
 99%|█████████▉| 10627/10682 [1:30:33<00:27,  2.01it/s]
 99%|█████████▉| 10628/10682 [1:30:34<00:26,  2.02it/s]
                                                       
+
+
                                                       
                                                       
+
+Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.