Training in progress, epoch 11

logs/events.out.tfevents.1715675319.sphinx2

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:188f9906bfbe016e7a173b5dbc1718f3152df2970e223b5fc7fa2addefdabbbe
-size 68994
+oid sha256:2acc52a51e079256a9579908f4b199f5e1bc1ffe67916df7dc8ef207ecfd195d
+size 75324

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3c30fd658ecd8be527af455b8155b5a0c27d8cf22abe8ad1b2fe94f6826e0799
+oid sha256:52150a6ebc81dad193b3cae38fff2e4eb54f171b9235b36e52968c357caf53dd
 size 281715176

train_job_output.txt

@@ -368,4 +368,34 @@ command outputs:
 
 71%|███████   | 7575/10682 [1:09:53<27:49,  1.86it/s]
 71%|███████   | 7576/10682 [1:09:53<27:09,  1.91it/s]
 71%|███████   | 7577/10682 [1:09:54<26:41,  1.94it/s]
 71%|███████   | 7578/10682 [1:09:54<26:20,  1.96it/s]
 71%|███████   | 7579/10682 [1:09:55<26:07,  1.98it/s]
 71%|███████   | 7580/10682 [1:09:55<25:55,  1.99it/s]
 71%|███████   | 7581/10682 [1:09:56<25:48,  2.00it/s]
 71%|███████   | 7582/10682 [1:09:56<25:43,  2.01it/s]
 71%|███████   | 7583/10682 [1:09:57<25:40,  2.01it/s]
 71%|███████   | 7584/10682 [1:09:57<25:37,  2.02it/s]
 71%|███████   | 7585/10682 [1:09:58<25:34,  2.02it/s]
 71%|███████   | 7586/10682 [1:09:58<25:33,  2.02it/s]
 71%|███████   | 7587/10682 [1:09:59<25:30,  2.02it/s]
 71%|███████   | 7588/10682 [1:09:59<25:30,  2.02it/s]
 71%|███████   | 7589/10682 [1:10:00<25:28,  2.02it/s]
 71%|███████   | 7590/10682 [1:10:00<25:29,  2.02it/s]
 71%|███████   | 7591/10682 [1:10:01<25:28,  2.02it/s]
 71%|███████   | 7592/10682 [1:10:01<25:28,  2.02it/s]
 71%|███████   | 7593/10682 [1:10:02<25:27,  2.02it/s]
 71%|███████   | 7594/10682 [1:10:02<25:26,  2.02it/s]
 71%|███████   | 7595/10682 [1:10:03<25:24,  2.03it/s]
 71%|███████   | 7596/10682 [1:10:03<25:24,  2.02it/s]
 71%|███████   | 7597/10682 [1:10:04<25:24,  2.02it/s]
 71%|███████   | 7598/10682 [1:10:04<25:21,  2.03it/s]
 71%|███████   | 7599/10682 [1:10:05<25:23,  2.02it/s]
 71%|███████   | 7600/10682 [1:10:05<25:20,  2.03it/s]{'loss': 2.9745, 'grad_norm': 0.24363452196121216, 'learning_rate': 0.00023289281143540065, 'epoch': 9.96}
 
                                                      
 71%|███████   | 7600/10682 [1:10:05<25:20,  2.03it/s]
 71%|███████   | 7601/10682 [1:10:06<25:24,  2.02it/s]
 71%|███████   | 7602/10682 [1:10:06<25:23,  2.02it/s]
 71%|███████   | 7603/10682 [1:10:07<25:23,  2.02it/s]
 71%|███████   | 7604/10682 [1:10:07<25:23,  2.02it/s]
 71%|███████   | 7605/10682 [1:10:08<25:22,  2.02it/s]
 71%|███████   | 7606/10682 [1:10:08<25:22,  2.02it/s]
 71%|███████   | 7607/10682 [1:10:09<25:20,  2.02it/s]
 71%|███████   | 7608/10682 [1:10:09<25:20,  2.02it/s]
 71%|███████   | 7609/10682 [1:10:10<25:19,  2.02it/s]
 71%|███████   | 7610/10682 [1:10:10<25:20,  2.02it/s]
 71%|███████▏  | 7611/10682 [1:10:11<25:19,  2.02it/s]
 71%|███████▏  | 7612/10682 [1:10:11<25:17,  2.02it/s]
 71%|███████▏  | 7613/10682 [1:10:12<25:17,  2.02it/s]
 71%|███████▏  | 7614/10682 [1:10:12<25:15,  2.02it/s]
 71%|███████▏  | 7615/10682 [1:10:13<25:15,  2.02it/s]
 71%|███████▏  | 7616/10682 [1:10:13<25:14,  2.02it/s]
 71%|███████▏  | 7617/10682 [1:10:14<25:15,  2.02it/s]
 71%|███████▏  | 7618/10682 [1:10:14<25:13,  2.02it/s]
 71%|███████▏  | 7619/10682 [1:10:15<25:14,  2.02it/s]
 71%|███████▏  | 7620/10682 [1:10:15<25:12,  2.02it/s]
 71%|███████▏  | 7621/10682 [1:10:16<25:13,  2.02it/s]
 71%|███████▏  | 7622/10682 [1:10:16<25:12,  2.02it/s]
 71%|███████▏  | 7623/10682 [1:10:17<25:13,  2.02it/s]
 71%|███████▏  | 7624/10682 [1:10:17<25:11,  2.02it/s]
 71%|███████▏  | 7625/10682 [1:10:18<25:12,  2.02it/s]
                                                      {'loss': 2.9725, 'grad_norm': 0.23881086707115173, 'learning_rate': 0.00022944844326774121, 'epoch': 9.99}
 
 71%|███████▏  | 7625/10682 [1:10:18<25:12,  2.02it/s]
 71%|███████▏  | 7626/10682 [1:10:18<25:11,  2.02it/s]
 71%|███████▏  | 7627/10682 [1:10:19<25:11,  2.02it/s]
 71%|███████▏  | 7628/10682 [1:10:19<25:10,  2.02it/s]
 71%|███████▏  | 7629/10682 [1:10:20<25:10,  2.02it/s]
 71%|███████▏  | 7630/10682 [1:10:20<24:52,  2.05it/s]
 71%|███████▏  | 7631/10682 [1:10:32<3:25:31,  4.04s/it]
 71%|███████▏  | 7632/10682 [1:10:33<2:31:22,  2.98s/it]
 71%|███████▏  | 7633/10682 [1:10:33<1:53:26,  2.23s/it]
 71%|███████▏  | 7634/10682 [1:10:34<1:27:02,  1.71s/it]
 71%|███████▏  | 7635/10682 [1:10:34<1:08:32,  1.35s/it]
 71%|███████▏  | 7636/10682 [1:10:35<55:29,  1.09s/it]  
 71%|███████▏  | 7637/10682 [1:10:35<46:26,  1.09it/s]
 72%|███████▏  | 7638/10682 [1:10:36<40:07,  1.26it/s]
 72%|███████▏  | 7639/10682 [1:10:36<35:36,  1.42it/s]
 72%|███████▏  | 7640/10682 [1:10:37<32:30,  1.56it/s]
 72%|███████▏  | 7641/10682 [1:10:37<30:15,  1.67it/s]
 72%|███████▏  | 7642/10682 [1:10:38<28:43,  1.76it/s]
 72%|███████▏  | 7643/10682 [1:10:38<27:37,  1.83it/s]
 72%|███████▏  | 7644/10682 [1:10:39<26:52,  1.88it/s]
 72%|███████▏  | 7645/10682 [1:10:39<26:22,  1.92it/s]
 72%|███████▏  | 7646/10682 [1:10:40<25:58,  1.95it/s]
 72%|███████▏  | 7647/10682 [1:10:40<25:44,  1.96it/s]
 72%|███████▏  | 7648/10682 [1:10:41<25:34,  1.98it/s]
 72%|███████▏  | 7649/10682 [1:10:41<25:23,  1.99it/s]
 72%|███████▏  | 7650/10682 [1:10:42<25:18,  2.00it/s]{'loss': 2.8655, 'grad_norm': 0.2524799108505249, 'learning_rate': 0.00022602213475715589, 'epoch': 10.03}
-
                                                      
 72%|███████▏  | 7650/10682 [1:10:42<25:18,  2.00it/s]
 72%|███████▏  | 7651/10682 [1:10:42<25:20,  1.99it/s]
 72%|███████▏  | 7652/10682 [1:10:43<25:13,  2.00it/s]
 72%|███████▏  | 7653/10682 [1:10:43<27:12,  1.86it/s]
 72%|███████▏  | 7654/10682 [1:10:44<26:30,  1.90it/s]
 72%|███████▏  | 7655/10682 [1:10:44<26:02,  1.94it/s]
+
                                                      
 72%|███████▏  | 7650/10682 [1:10:42<25:18,  2.00it/s]
 72%|███████▏  | 7651/10682 [1:10:42<25:20,  1.99it/s]
 72%|███████▏  | 7652/10682 [1:10:43<25:13,  2.00it/s]
 72%|███████▏  | 7653/10682 [1:10:43<27:12,  1.86it/s]
 72%|███████▏  | 7654/10682 [1:10:44<26:30,  1.90it/s]
 72%|███████▏  | 7655/10682 [1:10:44<26:02,  1.94it/s]
 72%|███████▏  | 7656/10682 [1:10:45<25:42,  1.96it/s]
 72%|███████▏  | 7657/10682 [1:10:45<25:27,  1.98it/s]
 72%|███████▏  | 7658/10682 [1:10:46<25:19,  1.99it/s]
 72%|███████▏  | 7659/10682 [1:10:46<25:11,  2.00it/s]
 72%|███████▏  | 7660/10682 [1:10:47<25:06,  2.01it/s]
 72%|███████▏  | 7661/10682 [1:10:47<25:03,  2.01it/s]
 72%|███████▏  | 7662/10682 [1:10:48<24:59,  2.01it/s]
 72%|███████▏  | 7663/10682 [1:10:48<24:57,  2.02it/s]
 72%|███████▏  | 7664/10682 [1:10:49<24:55,  2.02it/s]
 72%|███████▏  | 7665/10682 [1:10:49<24:52,  2.02it/s]
 72%|███████▏  | 7666/10682 [1:10:50<24:51,  2.02it/s]
 72%|███████▏  | 7667/10682 [1:10:50<24:50,  2.02it/s]
 72%|███████▏  | 7668/10682 [1:10:51<24:51,  2.02it/s]
 72%|███████▏  | 7669/10682 [1:10:51<24:50,  2.02it/s]
 72%|███████▏  | 7670/10682 [1:10:52<24:52,  2.02it/s]
 72%|███████▏  | 7671/10682 [1:10:52<24:51,  2.02it/s]
 72%|███████▏  | 7672/10682 [1:10:53<24:50,  2.02it/s]
 72%|███████▏  | 7673/10682 [1:10:53<24:49,  2.02it/s]
 72%|███████▏  | 7674/10682 [1:10:54<24:49,  2.02it/s]
 72%|███████▏  | 7675/10682 [1:10:54<24:47,  2.02it/s]
                                                      
+
 72%|███████▏  | 7675/10682 [1:10:54<24:47,  2.02it/s]
 72%|███████▏  | 7676/10682 [1:10:55<24:50,  2.02it/s]
 72%|███████▏  | 7677/10682 [1:10:55<24:48,  2.02it/s]
 72%|███████▏  | 7678/10682 [1:10:56<24:49,  2.02it/s]
 72%|███████▏  | 7679/10682 [1:10:56<24:47,  2.02it/s]
 72%|███████▏  | 7680/10682 [1:10:57<24:47,  2.02it/s]
 72%|███████▏  | 7681/10682 [1:10:57<24:44,  2.02it/s]
 72%|███████▏  | 7682/10682 [1:10:58<24:46,  2.02it/s]
 72%|███████▏  | 7683/10682 [1:10:58<24:46,  2.02it/s]
 72%|███████▏  | 7684/10682 [1:10:59<24:46,  2.02it/s]
 72%|███████▏  | 7685/10682 [1:10:59<24:47,  2.01it/s]
 72%|███████▏  | 7686/10682 [1:11:00<24:45,  2.02it/s]
 72%|███████▏  | 7687/10682 [1:11:00<24:43,  2.02it/s]
 72%|███████▏  | 7688/10682 [1:11:01<24:42,  2.02it/s]
 72%|███████▏  | 7689/10682 [1:11:01<24:42,  2.02it/s]
 72%|███████▏  | 7690/10682 [1:11:02<24:41,  2.02it/s]
 72%|███████▏  | 7691/10682 [1:11:02<24:40,  2.02it/s]
 72%|███████▏  | 7692/10682 [1:11:03<24:40,  2.02it/s]
 72%|███████▏  | 7693/10682 [1:11:03<24:39,  2.02it/s]
 72%|███████▏  | 7694/10682 [1:11:04<24:38,  2.02it/s]
 72%|███████▏  | 7695/10682 [1:11:04<24:36,  2.02it/s]
 72%|███████▏  | 7696/10682 [1:11:05<24:36,  2.02it/s]
 72%|███████▏  | 7697/10682 [1:11:05<24:36,  2.02it/s]
 72%|███████▏  | 7698/10682 [1:11:06<24:35,  2.02it/s]
 72%|███████▏  | 7699/10682 [1:11:06<24:33,  2.02it/s]
 72%|███████▏  | 7700/10682 [1:11:07<24:33,  2.02it/s]
                                                      
+
 72%|███████▏  | 7700/10682 [1:11:07<24:33,  2.02it/s]
 72%|███████▏  | 7701/10682 [1:11:07<24:34,  2.02it/s]
 72%|███████▏  | 7702/10682 [1:11:08<24:34,  2.02it/s]
 72%|███████▏  | 7703/10682 [1:11:08<24:32,  2.02it/s]
 72%|███████▏  | 7704/10682 [1:11:09<24:34,  2.02it/s]
 72%|███████▏  | 7705/10682 [1:11:09<24:33,  2.02it/s]
 72%|███████▏  | 7706/10682 [1:11:10<24:33,  2.02it/s]
 72%|███████▏  | 7707/10682 [1:11:10<24:30,  2.02it/s]
 72%|███████▏  | 7708/10682 [1:11:11<24:30,  2.02it/s]
 72%|███████▏  | 7709/10682 [1:11:11<24:28,  2.02it/s]
 72%|███████▏  | 7710/10682 [1:11:12<24:28,  2.02it/s]
 72%|███████▏  | 7711/10682 [1:11:12<24:28,  2.02it/s]
 72%|███████▏  | 7712/10682 [1:11:13<24:27,  2.02it/s]
 72%|███████▏  | 7713/10682 [1:11:13<24:28,  2.02it/s]
 72%|███████▏  | 7714/10682 [1:11:14<24:25,  2.02it/s]
 72%|███████▏  | 7715/10682 [1:11:14<24:27,  2.02it/s]
 72%|███████▏  | 7716/10682 [1:11:15<24:27,  2.02it/s]
 72%|███████▏  | 7717/10682 [1:11:15<24:28,  2.02it/s]
 72%|███████▏  | 7718/10682 [1:11:16<24:27,  2.02it/s]
 72%|███████▏  | 7719/10682 [1:11:16<24:28,  2.02it/s]
 72%|███████▏  | 7720/10682 [1:11:17<24:27,  2.02it/s]
 72%|███████▏  | 7721/10682 [1:11:17<24:26,  2.02it/s]
 72%|███████▏  | 7722/10682 [1:11:18<24:24,  2.02it/s]
 72%|███████▏  | 7723/10682 [1:11:18<24:22,  2.02it/s]
 72%|███████▏  | 7724/10682 [1:11:19<24:22,  2.02it/s]
 72%|███████▏  | 7725/10682 [1:11:19<24:20,  2.02it/s]
                                                      
+
 72%|███████▏  | 7725/10682 [1:11:19<24:20,  2.02it/s]
 72%|███████▏  | 7726/10682 [1:11:20<24:23,  2.02it/s]
 72%|███████▏  | 7727/10682 [1:11:20<24:21,  2.02it/s]
 72%|███████▏  | 7728/10682 [1:11:21<24:21,  2.02it/s]
 72%|███████▏  | 7729/10682 [1:11:21<24:19,  2.02it/s]
 72%|███████▏  | 7730/10682 [1:11:22<24:21,  2.02it/s]
 72%|███████▏  | 7731/10682 [1:11:22<24:20,  2.02it/s]
 72%|███████▏  | 7732/10682 [1:11:23<24:20,  2.02it/s]
 72%|███████▏  | 7733/10682 [1:11:23<24:17,  2.02it/s]
 72%|███████▏  | 7734/10682 [1:11:24<24:17,  2.02it/s]
 72%|███████▏  | 7735/10682 [1:11:24<24:16,  2.02it/s]
 72%|███████▏  | 7736/10682 [1:11:24<24:17,  2.02it/s]
 72%|███████▏  | 7737/10682 [1:11:25<24:15,  2.02it/s]
 72%|███████▏  | 7738/10682 [1:11:25<24:16,  2.02it/s]
 72%|███████▏  | 7739/10682 [1:11:26<24:15,  2.02it/s]
 72%|███████▏  | 7740/10682 [1:11:26<24:16,  2.02it/s]
 72%|███████▏  | 7741/10682 [1:11:27<24:15,  2.02it/s]
 72%|███████▏  | 7742/10682 [1:11:27<24:15,  2.02it/s]
 72%|███████▏  | 7743/10682 [1:11:28<24:13,  2.02it/s]
 72%|███████▏  | 7744/10682 [1:11:28<24:14,  2.02it/s]
 73%|███████▎  | 7745/10682 [1:11:29<24:13,  2.02it/s]
 73%|███████▎  | 7746/10682 [1:11:29<24:13,  2.02it/s]
 73%|███████▎  | 7747/10682 [1:11:30<24:12,  2.02it/s]
 73%|███████▎  | 7748/10682 [1:11:30<24:12,  2.02it/s]
 73%|███████▎  | 7749/10682 [1:11:31<24:10,  2.02it/s]
 73%|███████▎  | 7750/10682 [1:11:31<24:11,  2.02it/s]
                                                      {'loss': 2.8612, 'grad_norm': 0.24724432826042175, 'learning_rate': 0.00021250205309110155, 'epoch': 10.16}
+
 73%|███████▎  | 7750/10682 [1:11:31<24:11,  2.02it/s]
 73%|███████▎  | 7751/10682 [1:11:32<24:12,  2.02it/s]
 73%|███████▎  | 7752/10682 [1:11:32<24:10,  2.02it/s]
 73%|███████▎  | 7753/10682 [1:11:33<24:10,  2.02it/s]
 73%|███████▎  | 7754/10682 [1:11:33<24:08,  2.02it/s]
 73%|███████▎  | 7755/10682 [1:11:34<24:07,  2.02it/s]
 73%|███████▎  | 7756/10682 [1:11:34<24:07,  2.02it/s]
 73%|███████▎  | 7757/10682 [1:11:35<24:06,  2.02it/s]
 73%|███████▎  | 7758/10682 [1:11:35<24:06,  2.02it/s]
 73%|███████▎  | 7759/10682 [1:11:36<24:05,  2.02it/s]
 73%|███████▎  | 7760/10682 [1:11:36<24:04,  2.02it/s]
 73%|███████▎  | 7761/10682 [1:11:37<24:04,  2.02it/s]
 73%|███████▎  | 7762/10682 [1:11:37<24:02,  2.02it/s]
 73%|███████▎  | 7763/10682 [1:11:38<24:03,  2.02it/s]
 73%|███████▎  | 7764/10682 [1:11:38<24:03,  2.02it/s]
 73%|███████▎  | 7765/10682 [1:11:39<24:03,  2.02it/s]
 73%|███████▎  | 7766/10682 [1:11:39<24:02,  2.02it/s]
 73%|███████▎  | 7767/10682 [1:11:40<24:02,  2.02it/s]
 73%|███████▎  | 7768/10682 [1:11:40<24:01,  2.02it/s]
 73%|███████▎  | 7769/10682 [1:11:41<24:00,  2.02it/s]
 73%|███████▎  | 7770/10682 [1:11:41<23:59,  2.02it/s]
 73%|███████▎  | 7771/10682 [1:11:42<23:59,  2.02it/s]
 73%|███████▎  | 7772/10682 [1:11:42<23:58,  2.02it/s]
 73%|███████▎  | 7773/10682 [1:11:43<23:57,  2.02it/s]
 73%|███████▎  | 7774/10682 [1:11:43<23:59,  2.02it/s]
 73%|███████▎  | 7775/10682 [1:11:44<23:58,  2.02it/s]{'loss': 2.8622, 'grad_norm': 0.2527807056903839, 'learning_rate': 0.00020916944887928359, 'epoch': 10.19}
+
                                                      
 73%|███████▎  | 7775/10682 [1:11:44<23:58,  2.02it/s]
 73%|███████▎  | 7776/10682 [1:11:44<23:59,  2.02it/s]
 73%|███████▎  | 7777/10682 [1:11:45<24:00,  2.02it/s]
 73%|███████▎  | 7778/10682 [1:11:45<23:58,  2.02it/s]
 73%|███████▎  | 7779/10682 [1:11:46<23:59,  2.02it/s]
 73%|███████▎  | 7780/10682 [1:11:46<23:58,  2.02it/s]
 73%|███████▎  | 7781/10682 [1:11:47<23:58,  2.02it/s]
 73%|███████▎  | 7782/10682 [1:11:47<23:56,  2.02it/s]
 73%|███████▎  | 7783/10682 [1:11:48<23:55,  2.02it/s]
 73%|███████▎  | 7784/10682 [1:11:48<23:54,  2.02it/s]
 73%|███████▎  | 7785/10682 [1:11:49<23:54,  2.02it/s]
 73%|███████▎  | 7786/10682 [1:11:49<23:52,  2.02it/s]
 73%|███████▎  | 7787/10682 [1:11:50<23:52,  2.02it/s]
 73%|███████▎  | 7788/10682 [1:11:50<23:51,  2.02it/s]
 73%|███████▎  | 7789/10682 [1:11:51<23:50,  2.02it/s]
 73%|███████▎  | 7790/10682 [1:11:51<23:54,  2.02it/s]
 73%|███████▎  | 7791/10682 [1:11:52<23:51,  2.02it/s]
 73%|███████▎  | 7792/10682 [1:11:52<23:50,  2.02it/s]
 73%|███████▎  | 7793/10682 [1:11:53<23:49,  2.02it/s]
 73%|███████▎  | 7794/10682 [1:11:53<23:48,  2.02it/s]
 73%|███████▎  | 7795/10682 [1:11:54<23:48,  2.02it/s]
 73%|███████▎  | 7796/10682 [1:11:54<23:46,  2.02it/s]
 73%|███████▎  | 7797/10682 [1:11:55<23:46,  2.02it/s]
 73%|███████▎  | 7798/10682 [1:11:55<23:47,  2.02it/s]
 73%|███████▎  | 7799/10682 [1:11:56<23:47,  2.02it/s]
 73%|███████▎  | 7800/10682 [1:11:56<23:45,  2.02it/s]
                                                      {'loss': 2.8737, 'grad_norm': 0.24947026371955872, 'learning_rate': 0.00020585625797294927, 'epoch': 10.22}
+
 73%|███████▎  | 7800/10682 [1:11:56<23:45,  2.02it/s]
 73%|███████▎  | 7801/10682 [1:11:57<23:48,  2.02it/s]
 73%|███████▎  | 7802/10682 [1:11:57<23:47,  2.02it/s]
 73%|███████▎  | 7803/10682 [1:11:58<23:46,  2.02it/s]
 73%|███████▎  | 7804/10682 [1:11:58<23:43,  2.02it/s]
 73%|███████▎  | 7805/10682 [1:11:59<23:45,  2.02it/s]
 73%|███████▎  | 7806/10682 [1:11:59<23:41,  2.02it/s]
 73%|███████▎  | 7807/10682 [1:12:00<23:42,  2.02it/s]
 73%|███████▎  | 7808/10682 [1:12:00<23:40,  2.02it/s]
 73%|███████▎  | 7809/10682 [1:12:01<23:40,  2.02it/s]
 73%|███████▎  | 7810/10682 [1:12:01<23:40,  2.02it/s]
 73%|███████▎  | 7811/10682 [1:12:02<23:40,  2.02it/s]
 73%|███████▎  | 7812/10682 [1:12:02<23:38,  2.02it/s]
 73%|███████▎  | 7813/10682 [1:12:03<23:39,  2.02it/s]
 73%|███████▎  | 7814/10682 [1:12:03<23:37,  2.02it/s]
 73%|███████▎  | 7815/10682 [1:12:04<23:37,  2.02it/s]
 73%|███████▎  | 7816/10682 [1:12:04<23:37,  2.02it/s]
 73%|███████▎  | 7817/10682 [1:12:05<23:35,  2.02it/s]
 73%|███████▎  | 7818/10682 [1:12:05<23:35,  2.02it/s]
 73%|███████▎  | 7819/10682 [1:12:06<23:35,  2.02it/s]
 73%|███████▎  | 7820/10682 [1:12:06<23:35,  2.02it/s]
 73%|███████▎  | 7821/10682 [1:12:07<23:36,  2.02it/s]
 73%|███████▎  | 7822/10682 [1:12:07<23:35,  2.02it/s]
 73%|███████▎  | 7823/10682 [1:12:08<23:34,  2.02it/s]
 73%|███████▎  | 7824/10682 [1:12:08<23:34,  2.02it/s]
 73%|███████▎  | 7825/10682 [1:12:09<23:32,  2.02it/s]
                                                      {'loss': 2.8657, 'grad_norm': 0.2530911862850189, 'learning_rate': 0.00020256270153176371, 'epoch': 10.26}
+
 73%|███████▎  | 7825/10682 [1:12:09<23:32,  2.02it/s]
 73%|███████▎  | 7826/10682 [1:12:09<23:33,  2.02it/s]
 73%|███████▎  | 7827/10682 [1:12:10<23:31,  2.02it/s]
 73%|███████▎  | 7828/10682 [1:12:10<23:32,  2.02it/s]
 73%|███████▎  | 7829/10682 [1:12:11<23:30,  2.02it/s]
 73%|███████▎  | 7830/10682 [1:12:11<23:30,  2.02it/s]
 73%|███████▎  | 7831/10682 [1:12:11<23:29,  2.02it/s]
 73%|███████▎  | 7832/10682 [1:12:12<23:31,  2.02it/s]
 73%|███████▎  | 7833/10682 [1:12:12<23:30,  2.02it/s]
 73%|███████▎  | 7834/10682 [1:12:13<23:29,  2.02it/s]
 73%|███████▎  | 7835/10682 [1:12:13<23:29,  2.02it/s]
 73%|███████▎  | 7836/10682 [1:12:14<23:27,  2.02it/s]
 73%|███████▎  | 7837/10682 [1:12:14<23:27,  2.02it/s]
 73%|███████▎  | 7838/10682 [1:12:15<23:26,  2.02it/s]
 73%|███████▎  | 7839/10682 [1:12:15<23:25,  2.02it/s]
 73%|███████▎  | 7840/10682 [1:12:16<23:25,  2.02it/s]
 73%|███████▎  | 7841/10682 [1:12:16<23:23,  2.02it/s]
 73%|███████▎  | 7842/10682 [1:12:17<23:24,  2.02it/s]
 73%|███████▎  | 7843/10682 [1:12:17<23:22,  2.02it/s]
 73%|███████▎  | 7844/10682 [1:12:18<23:24,  2.02it/s]
 73%|███████▎  | 7845/10682 [1:12:18<23:23,  2.02it/s]
 73%|███████▎  | 7846/10682 [1:12:19<23:23,  2.02it/s]
 73%|███████▎  | 7847/10682 [1:12:19<23:22,  2.02it/s]
 73%|███████▎  | 7848/10682 [1:12:20<23:23,  2.02it/s]
 73%|███████▎  | 7849/10682 [1:12:20<23:21,  2.02it/s]
 73%|███████▎  | 7850/10682 [1:12:21<23:21,  2.02it/s]
                                                      
+
 73%|███████▎  | 7850/10682 [1:12:21<23:21,  2.02it/s]
 73%|███████▎  | 7851/10682 [1:12:21<23:22,  2.02it/s]
 74%|███████▎  | 7852/10682 [1:12:22<23:19,  2.02it/s]
 74%|███████▎  | 7853/10682 [1:12:22<23:20,  2.02it/s]
 74%|███████▎  | 7854/10682 [1:12:23<23:18,  2.02it/s]
 74%|███████▎  | 7855/10682 [1:12:23<23:18,  2.02it/s]
 74%|███████▎  | 7856/10682 [1:12:24<23:17,  2.02it/s]
 74%|███████▎  | 7857/10682 [1:12:24<23:18,  2.02it/s]
 74%|███████▎  | 7858/10682 [1:12:25<23:17,  2.02it/s]
 74%|███████▎  | 7859/10682 [1:12:25<23:16,  2.02it/s]
 74%|███████▎  | 7860/10682 [1:12:26<23:16,  2.02it/s]
 74%|███████▎  | 7861/10682 [1:12:26<23:14,  2.02it/s]
 74%|███████▎  | 7862/10682 [1:12:27<23:15,  2.02it/s]
 74%|███████▎  | 7863/10682 [1:12:27<23:14,  2.02it/s]
 74%|███████▎  | 7864/10682 [1:12:28<23:16,  2.02it/s]
 74%|███████▎  | 7865/10682 [1:12:28<23:13,  2.02it/s]
 74%|███████▎  | 7866/10682 [1:12:29<23:13,  2.02it/s]
 74%|███████▎  | 7867/10682 [1:12:29<23:12,  2.02it/s]
 74%|███████▎  | 7868/10682 [1:12:30<23:13,  2.02it/s]
 74%|███████▎  | 7869/10682 [1:12:30<23:11,  2.02it/s]
 74%|███████▎  | 7870/10682 [1:12:31<23:11,  2.02it/s]
 74%|███████▎  | 7871/10682 [1:12:31<23:10,  2.02it/s]
 74%|███████▎  | 7872/10682 [1:12:32<23:10,  2.02it/s]
 74%|███████▎  | 7873/10682 [1:12:32<23:09,  2.02it/s]
 74%|███████▎  | 7874/10682 [1:12:33<23:10,  2.02it/s]
 74%|███████▎  | 7875/10682 [1:12:33<23:09,  2.02it/s]
                                                      {'loss': 2.8799, 'grad_norm': 0.26851731538772583, 'learning_rate': 0.00019603537011569566, 'epoch': 10.32}
+
 74%|███████▎  | 7875/10682 [1:12:33<23:09,  2.02it/s]
 74%|███████▎  | 7876/10682 [1:12:34<23:09,  2.02it/s]
 74%|███████▎  | 7877/10682 [1:12:34<23:09,  2.02it/s]
 74%|███████▍  | 7878/10682 [1:12:35<23:07,  2.02it/s]
 74%|███████▍  | 7879/10682 [1:12:35<23:08,  2.02it/s]
 74%|███████▍  | 7880/10682 [1:12:36<23:06,  2.02it/s]
 74%|███████▍  | 7881/10682 [1:12:36<23:04,  2.02it/s]
 74%|███████▍  | 7882/10682 [1:12:37<23:03,  2.02it/s]
 74%|███████▍  | 7883/10682 [1:12:37<23:03,  2.02it/s]
 74%|███████▍  | 7884/10682 [1:12:38<23:03,  2.02it/s]
 74%|███████▍  | 7885/10682 [1:12:38<23:04,  2.02it/s]
 74%|███████▍  | 7886/10682 [1:12:39<23:02,  2.02it/s]
 74%|███████▍  | 7887/10682 [1:12:39<23:02,  2.02it/s]
 74%|███████▍  | 7888/10682 [1:12:40<23:01,  2.02it/s]
 74%|███████▍  | 7889/10682 [1:12:40<23:02,  2.02it/s]
 74%|███████▍  | 7890/10682 [1:12:41<23:01,  2.02it/s]
 74%|███████▍  | 7891/10682 [1:12:41<23:01,  2.02it/s]
 74%|███████▍  | 7892/10682 [1:12:42<22:59,  2.02it/s]
 74%|███████▍  | 7893/10682 [1:12:42<23:00,  2.02it/s]
 74%|███████▍  | 7894/10682 [1:12:43<22:58,  2.02it/s]
 74%|███████▍  | 7895/10682 [1:12:43<22:59,  2.02it/s]
 74%|███████▍  | 7896/10682 [1:12:44<22:56,  2.02it/s]
 74%|███████▍  | 7897/10682 [1:12:44<22:57,  2.02it/s]
 74%|███████▍  | 7898/10682 [1:12:45<22:55,  2.02it/s]
 74%|███████▍  | 7899/10682 [1:12:45<22:56,  2.02it/s]
 74%|███████▍  | 7900/10682 [1:12:46<22:55,  2.02it/s]
                                                      {'loss': 2.8842, 'grad_norm': 0.24814572930335999, 'learning_rate': 0.0001928020308484042, 'epoch': 10.35}
+
 74%|███████▍  | 7900/10682 [1:12:46<22:55,  2.02it/s]
 74%|███████▍  | 7901/10682 [1:12:46<22:57,  2.02it/s]
 74%|███████▍  | 7902/10682 [1:12:47<22:54,  2.02it/s]
 74%|███████▍  | 7903/10682 [1:12:47<22:56,  2.02it/s]
 74%|███████▍  | 7904/10682 [1:12:48<22:53,  2.02it/s]
 74%|███████▍  | 7905/10682 [1:12:48<22:54,  2.02it/s]
 74%|███████▍  | 7906/10682 [1:12:49<22:53,  2.02it/s]
 74%|███████▍  | 7907/10682 [1:12:49<22:53,  2.02it/s]
 74%|███████▍  | 7908/10682 [1:12:50<22:51,  2.02it/s]
 74%|███████▍  | 7909/10682 [1:12:50<22:53,  2.02it/s]
 74%|███████▍  | 7910/10682 [1:12:51<22:51,  2.02it/s]
 74%|███████▍  | 7911/10682 [1:12:51<22:51,  2.02it/s]
 74%|███████▍  | 7912/10682 [1:12:52<22:49,  2.02it/s]
 74%|███████▍  | 7913/10682 [1:12:52<22:50,  2.02it/s]
 74%|███████▍  | 7914/10682 [1:12:53<22:48,  2.02it/s]
 74%|███████▍  | 7915/10682 [1:12:53<22:48,  2.02it/s]
 74%|███████▍  | 7916/10682 [1:12:54<22:47,  2.02it/s]
 74%|███████▍  | 7917/10682 [1:12:54<22:48,  2.02it/s]
 74%|███████▍  | 7918/10682 [1:12:55<22:46,  2.02it/s]
 74%|███████▍  | 7919/10682 [1:12:55<22:48,  2.02it/s]
 74%|███████▍  | 7920/10682 [1:12:56<22:45,  2.02it/s]
 74%|███████▍  | 7921/10682 [1:12:56<22:45,  2.02it/s]
 74%|███████▍  | 7922/10682 [1:12:57<22:44,  2.02it/s]
 74%|███████▍  | 7923/10682 [1:12:57<22:45,  2.02it/s]
 74%|███████▍  | 7924/10682 [1:12:58<22:45,  2.02it/s]
 74%|███████▍  | 7925/10682 [1:12:58<22:45,  2.02it/s]
                                                      {'loss': 2.8778, 'grad_norm': 0.24965329468250275, 'learning_rate': 0.00018958919743235897, 'epoch': 10.39}
+
 74%|███████▍  | 7925/10682 [1:12:58<22:45,  2.02it/s]
 74%|███████▍  | 7926/10682 [1:12:59<22:47,  2.02it/s]
 74%|███████▍  | 7927/10682 [1:12:59<22:44,  2.02it/s]
 74%|███████▍  | 7928/10682 [1:12:59<22:44,  2.02it/s]
 74%|███████▍  | 7929/10682 [1:13:00<22:43,  2.02it/s]
 74%|███████▍  | 7930/10682 [1:13:00<22:42,  2.02it/s]
 74%|███████▍  | 7931/10682 [1:13:01<22:41,  2.02it/s]
 74%|███████▍  | 7932/10682 [1:13:01<22:40,  2.02it/s]
 74%|███████▍  | 7933/10682 [1:13:02<22:39,  2.02it/s]
 74%|███████▍  | 7934/10682 [1:13:02<22:38,  2.02it/s]
 74%|███████▍  | 7935/10682 [1:13:03<22:39,  2.02it/s]
 74%|███████▍  | 7936/10682 [1:13:03<22:38,  2.02it/s]
 74%|███████▍  | 7937/10682 [1:13:04<22:39,  2.02it/s]
 74%|███████▍  | 7938/10682 [1:13:04<22:37,  2.02it/s]
 74%|███████▍  | 7939/10682 [1:13:05<22:37,  2.02it/s]
 74%|███████▍  | 7940/10682 [1:13:05<22:36,  2.02it/s]
 74%|███████▍  | 7941/10682 [1:13:06<22:37,  2.02it/s]
 74%|███████▍  | 7942/10682 [1:13:06<22:35,  2.02it/s]
 74%|███████▍  | 7943/10682 [1:13:07<22:36,  2.02it/s]
 74%|███████▍  | 7944/10682 [1:13:07<22:35,  2.02it/s]
 74%|███████▍  | 7945/10682 [1:13:08<22:35,  2.02it/s]
 74%|███████▍  | 7946/10682 [1:13:08<22:34,  2.02it/s]
 74%|███████▍  | 7947/10682 [1:13:09<22:33,  2.02it/s]
 74%|███████▍  | 7948/10682 [1:13:09<22:32,  2.02it/s]
 74%|███████▍  | 7949/10682 [1:13:10<22:32,  2.02it/s]
 74%|███████▍  | 7950/10682 [1:13:10<22:30,  2.02it/s]
                                                      {'loss': 2.8803, 'grad_norm': 0.24981389939785004, 'learning_rate': 0.0001863970843282357, 'epoch': 10.42}
+
 74%|███████▍  | 7950/10682 [1:13:10<22:30,  2.02it/s]
 74%|███████▍  | 7951/10682 [1:13:11<22:33,  2.02it/s]
 74%|███████▍  | 7952/10682 [1:13:11<22:31,  2.02it/s]
 74%|███████▍  | 7953/10682 [1:13:12<22:32,  2.02it/s]
 74%|███████▍  | 7954/10682 [1:13:12<22:31,  2.02it/s]
 74%|███████▍  | 7955/10682 [1:13:13<22:31,  2.02it/s]
 74%|███████▍  | 7956/10682 [1:13:13<22:29,  2.02it/s]
 74%|███████▍  | 7957/10682 [1:13:14<22:29,  2.02it/s]
 74%|███████▍  | 7958/10682 [1:13:14<22:27,  2.02it/s]
 75%|███████▍  | 7959/10682 [1:13:15<22:27,  2.02it/s]
 75%|███████▍  | 7960/10682 [1:13:15<22:25,  2.02it/s]
 75%|███████▍  | 7961/10682 [1:13:16<22:27,  2.02it/s]
 75%|███████▍  | 7962/10682 [1:13:16<22:27,  2.02it/s]
 75%|███████▍  | 7963/10682 [1:13:17<22:26,  2.02it/s]
 75%|███████▍  | 7964/10682 [1:13:17<22:26,  2.02it/s]
 75%|███████▍  | 7965/10682 [1:13:18<22:24,  2.02it/s]
 75%|███████▍  | 7966/10682 [1:13:18<22:23,  2.02it/s]
 75%|███████▍  | 7967/10682 [1:13:19<22:23,  2.02it/s]
 75%|███████▍  | 7968/10682 [1:13:19<22:24,  2.02it/s]
 75%|███████▍  | 7969/10682 [1:13:20<22:22,  2.02it/s]
 75%|███████▍  | 7970/10682 [1:13:20<22:21,  2.02it/s]
 75%|███████▍  | 7971/10682 [1:13:21<22:20,  2.02it/s]
 75%|███████▍  | 7972/10682 [1:13:21<22:20,  2.02it/s]
 75%|███████▍  | 7973/10682 [1:13:22<22:20,  2.02it/s]
 75%|███████▍  | 7974/10682 [1:13:22<22:21,  2.02it/s]
 75%|███████▍  | 7975/10682 [1:13:23<22:19,  2.02it/s]
                                                      
+
 75%|███████▍  | 7975/10682 [1:13:23<22:19,  2.02it/s]
 75%|███████▍  | 7976/10682 [1:13:23<22:23,  2.01it/s]
 75%|███████▍  | 7977/10682 [1:13:24<22:20,  2.02it/s]
 75%|███████▍  | 7978/10682 [1:13:24<22:20,  2.02it/s]
 75%|███████▍  | 7979/10682 [1:13:25<22:18,  2.02it/s]
 75%|███████▍  | 7980/10682 [1:13:25<22:19,  2.02it/s]
 75%|███████▍  | 7981/10682 [1:13:26<22:17,  2.02it/s]
 75%|███████▍  | 7982/10682 [1:13:26<22:17,  2.02it/s]
 75%|███████▍  | 7983/10682 [1:13:27<22:15,  2.02it/s]
 75%|███████▍  | 7984/10682 [1:13:27<22:15,  2.02it/s]
 75%|███████▍  | 7985/10682 [1:13:28<22:14,  2.02it/s]
 75%|███████▍  | 7986/10682 [1:13:28<22:14,  2.02it/s]
 75%|███████▍  | 7987/10682 [1:13:29<22:14,  2.02it/s]
 75%|███████▍  | 7988/10682 [1:13:29<22:13,  2.02it/s]
 75%|███████▍  | 7989/10682 [1:13:30<22:13,  2.02it/s]
 75%|███████▍  | 7990/10682 [1:13:30<22:12,  2.02it/s]
 75%|███████▍  | 7991/10682 [1:13:31<22:13,  2.02it/s]
 75%|███████▍  | 7992/10682 [1:13:31<22:11,  2.02it/s]
 75%|███████▍  | 7993/10682 [1:13:32<22:11,  2.02it/s]
 75%|███████▍  | 7994/10682 [1:13:32<22:09,  2.02it/s]
 75%|███████▍  | 7995/10682 [1:13:33<22:10,  2.02it/s]
 75%|███████▍  | 7996/10682 [1:13:33<22:10,  2.02it/s]
 75%|███████▍  | 7997/10682 [1:13:34<22:10,  2.02it/s]
 75%|███████▍  | 7998/10682 [1:13:34<22:08,  2.02it/s]
 75%|███████▍  | 7999/10682 [1:13:35<22:08,  2.02it/s]
 75%|███████▍  | 8000/10682 [1:13:35<22:06,  2.02it/s]
                                                      {'loss': 2.8918, 'grad_norm': 0.24982097744941711, 'learning_rate': 0.00018007586996870206, 'epoch': 10.48}
+
 75%|███████▍  | 8000/10682 [1:13:35<22:06,  2.02it/s]
 75%|███████▍  | 8001/10682 [1:13:36<22:09,  2.02it/s]
 75%|███████▍  | 8002/10682 [1:13:36<22:07,  2.02it/s]
 75%|███████▍  | 8003/10682 [1:13:37<22:06,  2.02it/s]
 75%|███████▍  | 8004/10682 [1:13:37<22:05,  2.02it/s]
 75%|███████▍  | 8005/10682 [1:13:38<22:05,  2.02it/s]
 75%|███████▍  | 8006/10682 [1:13:38<22:04,  2.02it/s]
 75%|███████▍  | 8007/10682 [1:13:39<22:04,  2.02it/s]
 75%|███████▍  | 8008/10682 [1:13:39<22:03,  2.02it/s]
 75%|███████▍  | 8009/10682 [1:13:40<22:03,  2.02it/s]
 75%|███████▍  | 8010/10682 [1:13:40<22:01,  2.02it/s]
 75%|███████▍  | 8011/10682 [1:13:41<22:01,  2.02it/s]
 75%|███████▌  | 8012/10682 [1:13:41<22:00,  2.02it/s]
 75%|███████▌  | 8013/10682 [1:13:42<22:01,  2.02it/s]
 75%|███████▌  | 8014/10682 [1:13:42<21:59,  2.02it/s]
 75%|███████▌  | 8015/10682 [1:13:43<22:00,  2.02it/s]
 75%|███████▌  | 8016/10682 [1:13:43<21:58,  2.02it/s]
 75%|███████▌  | 8017/10682 [1:13:44<21:58,  2.02it/s]
 75%|███████▌  | 8018/10682 [1:13:44<21:56,  2.02it/s]
 75%|███████▌  | 8019/10682 [1:13:45<21:58,  2.02it/s]
 75%|███████▌  | 8020/10682 [1:13:45<21:55,  2.02it/s]
 75%|███████▌  | 8021/10682 [1:13:46<21:56,  2.02it/s]
 75%|███████▌  | 8022/10682 [1:13:46<21:55,  2.02it/s]
 75%|███████▌  | 8023/10682 [1:13:47<21:55,  2.02it/s]
 75%|███████▌  | 8024/10682 [1:13:47<21:54,  2.02it/s]
 75%|███████▌  | 8025/10682 [1:13:48<21:54,  2.02it/s]{'loss': 2.9015, 'grad_norm': 0.2518491744995117, 'learning_rate': 0.00017694719066230924, 'epoch': 10.52}
                                                      
+
 75%|███████▌  | 8025/10682 [1:13:48<21:54,  2.02it/s]
 75%|███████▌  | 8026/10682 [1:13:48<21:54,  2.02it/s]
 75%|███████▌  | 8027/10682 [1:13:48<21:54,  2.02it/s]
 75%|███████▌  | 8028/10682 [1:13:49<21:51,  2.02it/s]
 75%|███████▌  | 8029/10682 [1:13:49<21:52,  2.02it/s]
 75%|███████▌  | 8030/10682 [1:13:50<21:50,  2.02it/s]
 75%|███████▌  | 8031/10682 [1:13:50<21:51,  2.02it/s]
 75%|███████▌  | 8032/10682 [1:13:51<21:51,  2.02it/s]
 75%|███████▌  | 8033/10682 [1:13:51<21:51,  2.02it/s]
 75%|███████▌  | 8034/10682 [1:13:52<21:50,  2.02it/s]
 75%|███████▌  | 8035/10682 [1:13:52<21:50,  2.02it/s]
 75%|███████▌  | 8036/10682 [1:13:53<21:48,  2.02it/s]
 75%|███████▌  | 8037/10682 [1:13:53<21:48,  2.02it/s]
 75%|███████▌  | 8038/10682 [1:13:54<21:46,  2.02it/s]
 75%|███████▌  | 8039/10682 [1:13:54<21:47,  2.02it/s]
 75%|███████▌  | 8040/10682 [1:13:55<21:46,  2.02it/s]
 75%|███████▌  | 8041/10682 [1:13:55<21:46,  2.02it/s]
 75%|███████▌  | 8042/10682 [1:13:56<21:45,  2.02it/s]
 75%|███████▌  | 8043/10682 [1:13:56<21:46,  2.02it/s]
 75%|███████▌  | 8044/10682 [1:13:57<21:43,  2.02it/s]
 75%|███████▌  | 8045/10682 [1:13:57<21:44,  2.02it/s]
 75%|███████▌  | 8046/10682 [1:13:58<21:43,  2.02it/s]
 75%|███████▌  | 8047/10682 [1:13:58<21:43,  2.02it/s]
 75%|███████▌  | 8048/10682 [1:13:59<21:42,  2.02it/s]
 75%|███████▌  | 8049/10682 [1:13:59<21:42,  2.02it/s]
 75%|███████▌  | 8050/10682 [1:14:00<21:40,  2.02it/s]
                                                      {'loss': 2.8929, 'grad_norm': 0.24705103039741516, 'learning_rate': 0.00017384007553770858, 'epoch': 10.55}
+
 75%|███████▌  | 8050/10682 [1:14:00<21:40,  2.02it/s]
 75%|███████▌  | 8051/10682 [1:14:00<21:42,  2.02it/s]
 75%|███████▌  | 8052/10682 [1:14:01<21:40,  2.02it/s]
 75%|███████▌  | 8053/10682 [1:14:01<21:41,  2.02it/s]
 75%|███████▌  | 8054/10682 [1:14:02<21:39,  2.02it/s]
 75%|███████▌  | 8055/10682 [1:14:02<21:40,  2.02it/s]
 75%|███████▌  | 8056/10682 [1:14:03<21:39,  2.02it/s]
 75%|███████▌  | 8057/10682 [1:14:03<21:39,  2.02it/s]
 75%|███████▌  | 8058/10682 [1:14:04<21:38,  2.02it/s]
 75%|███████▌  | 8059/10682 [1:14:04<21:38,  2.02it/s]
 75%|███████▌  | 8060/10682 [1:14:05<21:36,  2.02it/s]
 75%|███████▌  | 8061/10682 [1:14:05<21:36,  2.02it/s]
 75%|███████▌  | 8062/10682 [1:14:06<21:36,  2.02it/s]
 75%|███████▌  | 8063/10682 [1:14:06<21:35,  2.02it/s]
 75%|███████▌  | 8064/10682 [1:14:07<21:34,  2.02it/s]
 76%|███████▌  | 8065/10682 [1:14:07<21:35,  2.02it/s]
 76%|███████▌  | 8066/10682 [1:14:08<21:34,  2.02it/s]
 76%|███████▌  | 8067/10682 [1:14:08<21:34,  2.02it/s]
 76%|███████▌  | 8068/10682 [1:14:09<21:32,  2.02it/s]
 76%|███████▌  | 8069/10682 [1:14:09<21:32,  2.02it/s]
 76%|███████▌  | 8070/10682 [1:14:10<21:31,  2.02it/s]
 76%|███████▌  | 8071/10682 [1:14:10<21:32,  2.02it/s]
 76%|███████▌  | 8072/10682 [1:14:11<21:32,  2.02it/s]
 76%|███████▌  | 8073/10682 [1:14:11<21:33,  2.02it/s]
 76%|███████▌  | 8074/10682 [1:14:12<21:31,  2.02it/s]
 76%|███████▌  | 8075/10682 [1:14:12<21:31,  2.02it/s]
                                                      {'loss': 2.8997, 'grad_norm': 0.2514490485191345, 'learning_rate': 0.00017075473199874692, 'epoch': 10.58}
+
 76%|███████▌  | 8075/10682 [1:14:12<21:31,  2.02it/s]
 76%|███████▌  | 8076/10682 [1:14:13<21:33,  2.01it/s]
 76%|███████▌  | 8077/10682 [1:14:13<21:32,  2.02it/s]
 76%|███████▌  | 8078/10682 [1:14:14<21:31,  2.02it/s]
 76%|███████▌  | 8079/10682 [1:14:14<21:28,  2.02it/s]
 76%|███████▌  | 8080/10682 [1:14:15<21:29,  2.02it/s]
 76%|███████▌  | 8081/10682 [1:14:15<21:27,  2.02it/s]
 76%|███████▌  | 8082/10682 [1:14:16<21:27,  2.02it/s]
 76%|███████▌  | 8083/10682 [1:14:16<21:26,  2.02it/s]
 76%|███████▌  | 8084/10682 [1:14:17<21:25,  2.02it/s]
 76%|███████▌  | 8085/10682 [1:14:17<21:25,  2.02it/s]
 76%|███████▌  | 8086/10682 [1:14:18<21:23,  2.02it/s]
 76%|███████▌  | 8087/10682 [1:14:18<21:24,  2.02it/s]
 76%|███████▌  | 8088/10682 [1:14:19<21:22,  2.02it/s]
 76%|███████▌  | 8089/10682 [1:14:19<21:22,  2.02it/s]
 76%|███████▌  | 8090/10682 [1:14:20<21:21,  2.02it/s]
 76%|███████▌  | 8091/10682 [1:14:20<21:22,  2.02it/s]
 76%|███████▌  | 8092/10682 [1:14:21<21:21,  2.02it/s]
 76%|███████▌  | 8093/10682 [1:14:21<21:33,  2.00it/s]
 76%|███████▌  | 8094/10682 [1:14:22<21:28,  2.01it/s]
 76%|███████▌  | 8095/10682 [1:14:22<21:25,  2.01it/s]
 76%|███████▌  | 8096/10682 [1:14:23<21:23,  2.02it/s]
 76%|███████▌  | 8097/10682 [1:14:23<21:21,  2.02it/s]
 76%|███████▌  | 8098/10682 [1:14:24<21:19,  2.02it/s]
 76%|███████▌  | 8099/10682 [1:14:24<21:18,  2.02it/s]
 76%|███████▌  | 8100/10682 [1:14:25<21:18,  2.02it/s]
                                                      
+
 76%|███████▌  | 8100/10682 [1:14:25<21:18,  2.02it/s]
 76%|███████▌  | 8101/10682 [1:14:25<21:19,  2.02it/s]
 76%|███████▌  | 8102/10682 [1:14:26<21:17,  2.02it/s]
 76%|███████▌  | 8103/10682 [1:14:26<21:17,  2.02it/s]
 76%|███████▌  | 8104/10682 [1:14:27<21:15,  2.02it/s]
 76%|███████▌  | 8105/10682 [1:14:27<21:14,  2.02it/s]
 76%|███████▌  | 8106/10682 [1:14:28<21:14,  2.02it/s]
 76%|███████▌  | 8107/10682 [1:14:28<21:14,  2.02it/s]
 76%|███████▌  | 8108/10682 [1:14:29<21:13,  2.02it/s]
 76%|███████▌  | 8109/10682 [1:14:29<21:12,  2.02it/s]
 76%|███████▌  | 8110/10682 [1:14:30<21:11,  2.02it/s]
 76%|███████▌  | 8111/10682 [1:14:30<21:11,  2.02it/s]
 76%|███████▌  | 8112/10682 [1:14:31<21:11,  2.02it/s]
 76%|███████▌  | 8113/10682 [1:14:31<21:12,  2.02it/s]
 76%|███████▌  | 8114/10682 [1:14:32<21:12,  2.02it/s]
 76%|███████▌  | 8115/10682 [1:14:32<21:10,  2.02it/s]
 76%|███████▌  | 8116/10682 [1:14:33<21:10,  2.02it/s]
 76%|███████▌  | 8117/10682 [1:14:33<21:08,  2.02it/s]
 76%|███████▌  | 8118/10682 [1:14:34<21:09,  2.02it/s]
 76%|███████▌  | 8119/10682 [1:14:34<21:07,  2.02it/s]
 76%|███████▌  | 8120/10682 [1:14:35<21:07,  2.02it/s]
 76%|███████▌  | 8121/10682 [1:14:35<21:06,  2.02it/s]
 76%|███████▌  | 8122/10682 [1:14:36<21:05,  2.02it/s]
 76%|███████▌  | 8123/10682 [1:14:36<21:05,  2.02it/s]
 76%|███████▌  | 8124/10682 [1:14:37<21:05,  2.02it/s]
 76%|███████▌  | 8125/10682 [1:14:37<21:05,  2.02it/s]{'loss': 2.8945, 'grad_norm': 0.24637724459171295, 'learning_rate': 0.0001646501820129766, 'epoch': 10.65}
+
                                                      
 76%|███████▌  | 8125/10682 [1:14:37<21:05,  2.02it/s]
 76%|███████▌  | 8126/10682 [1:14:37<21:05,  2.02it/s]
 76%|███████▌  | 8127/10682 [1:14:38<21:05,  2.02it/s]
 76%|███████▌  | 8128/10682 [1:14:38<21:04,  2.02it/s]
 76%|███████▌  | 8129/10682 [1:14:39<21:04,  2.02it/s]
 76%|███████▌  | 8130/10682 [1:14:39<21:02,  2.02it/s]
 76%|███████▌  | 8131/10682 [1:14:40<21:02,  2.02it/s]
 76%|███████▌  | 8132/10682 [1:14:40<21:00,  2.02it/s]
 76%|███████▌  | 8133/10682 [1:14:41<20:59,  2.02it/s]
 76%|███████▌  | 8134/10682 [1:14:41<20:59,  2.02it/s]
 76%|███████▌  | 8135/10682 [1:14:42<20:58,  2.02it/s]
 76%|███████▌  | 8136/10682 [1:14:42<20:59,  2.02it/s]
 76%|███████▌  | 8137/10682 [1:14:43<20:58,  2.02it/s]
 76%|███████▌  | 8138/10682 [1:14:43<20:59,  2.02it/s]
 76%|███████▌  | 8139/10682 [1:14:44<20:58,  2.02it/s]
 76%|███████▌  | 8140/10682 [1:14:44<20:58,  2.02it/s]
 76%|███████▌  | 8141/10682 [1:14:45<20:58,  2.02it/s]
 76%|███████▌  | 8142/10682 [1:14:45<20:57,  2.02it/s]
 76%|███████▌  | 8143/10682 [1:14:46<20:55,  2.02it/s]
 76%|██��████▌  | 8144/10682 [1:14:46<20:55,  2.02it/s]
 76%|███████▌  | 8145/10682 [1:14:47<20:55,  2.02it/s]
 76%|███████▋  | 8146/10682 [1:14:47<20:54,  2.02it/s]
 76%|███████▋  | 8147/10682 [1:14:48<20:53,  2.02it/s]
 76%|███████▋  | 8148/10682 [1:14:48<20:54,  2.02it/s]
 76%|███████▋  | 8149/10682 [1:14:49<20:54,  2.02it/s]
 76%|███████▋  | 8150/10682 [1:14:49<20:52,  2.02it/s]{'loss': 2.8927, 'grad_norm': 0.25224441289901733, 'learning_rate': 0.00016163138305256598, 'epoch': 10.68}
                                                      
+
 76%|███████▋  | 8150/10682 [1:14:49<20:52,  2.02it/s]
 76%|███████▋  | 8151/10682 [1:14:50<20:54,  2.02it/s]
 76%|███████▋  | 8152/10682 [1:14:50<20:52,  2.02it/s]
 76%|███████▋  | 8153/10682 [1:14:51<20:52,  2.02it/s]
 76%|███████▋  | 8154/10682 [1:14:51<20:51,  2.02it/s]
 76%|███████▋  | 8155/10682 [1:14:52<20:51,  2.02it/s]
 76%|███████▋  | 8156/10682 [1:14:52<20:50,  2.02it/s]
 76%|███████▋  | 8157/10682 [1:14:53<20:49,  2.02it/s]
 76%|███████▋  | 8158/10682 [1:14:53<20:48,  2.02it/s]
 76%|███████▋  | 8159/10682 [1:14:54<20:49,  2.02it/s]
 76%|███████▋  | 8160/10682 [1:14:54<20:47,  2.02it/s]
 76%|███████▋  | 8161/10682 [1:14:55<20:48,  2.02it/s]
 76%|███████▋  | 8162/10682 [1:14:55<20:46,  2.02it/s]
 76%|███████▋  | 8163/10682 [1:14:56<20:46,  2.02it/s]
 76%|███████▋  | 8164/10682 [1:14:56<20:45,  2.02it/s]
 76%|███████▋  | 8165/10682 [1:14:57<20:45,  2.02it/s]
 76%|███████▋  | 8166/10682 [1:14:57<20:44,  2.02it/s]
 76%|███████▋  | 8167/10682 [1:14:58<20:44,  2.02it/s]
 76%|███████▋  | 8168/10682 [1:14:58<20:43,  2.02it/s]
 76%|███████▋  | 8169/10682 [1:14:59<20:43,  2.02it/s]
 76%|███████▋  | 8170/10682 [1:14:59<20:42,  2.02it/s]
 76%|███████▋  | 8171/10682 [1:15:00<20:42,  2.02it/s]
 77%|███████▋  | 8172/10682 [1:15:00<20:40,  2.02it/s]
 77%|███████▋  | 8173/10682 [1:15:01<20:41,  2.02it/s]
 77%|███████▋  | 8174/10682 [1:15:01<20:39,  2.02it/s]
 77%|███████▋  | 8175/10682 [1:15:02<20:39,  2.02it/s]
                                                      
+
 77%|███████▋  | 8175/10682 [1:15:02<20:39,  2.02it/s]
 77%|███████▋  | 8176/10682 [1:15:02<20:41,  2.02it/s]
 77%|███████▋  | 8177/10682 [1:15:03<20:40,  2.02it/s]
 77%|███████▋  | 8178/10682 [1:15:03<20:39,  2.02it/s]
 77%|███████▋  | 8179/10682 [1:15:04<20:38,  2.02it/s]
 77%|███████▋  | 8180/10682 [1:15:04<20:38,  2.02it/s]
 77%|███████▋  | 8181/10682 [1:15:05<20:37,  2.02it/s]
 77%|███████▋  | 8182/10682 [1:15:05<20:36,  2.02it/s]
 77%|███████▋  | 8183/10682 [1:15:06<20:37,  2.02it/s]
 77%|███████▋  | 8184/10682 [1:15:06<20:35,  2.02it/s]
 77%|███████▋  | 8185/10682 [1:15:07<20:35,  2.02it/s]
 77%|███████▋  | 8186/10682 [1:15:07<20:36,  2.02it/s]
 77%|███████▋  | 8187/10682 [1:15:08<20:36,  2.02it/s]
 77%|███████▋  | 8188/10682 [1:15:08<20:36,  2.02it/s]
 77%|███████▋  | 8189/10682 [1:15:09<20:33,  2.02it/s]
 77%|███████▋  | 8190/10682 [1:15:09<20:33,  2.02it/s]
 77%|███████▋  | 8191/10682 [1:15:10<20:31,  2.02it/s]
 77%|███████▋  | 8192/10682 [1:15:10<20:31,  2.02it/s]
 77%|███████▋  | 8193/10682 [1:15:11<20:30,  2.02it/s]
 77%|███████▋  | 8194/10682 [1:15:11<20:31,  2.02it/s]
 77%|███████▋  | 8195/10682 [1:15:12<20:29,  2.02it/s]
 77%|███████▋  | 8196/10682 [1:15:12<20:29,  2.02it/s]
 77%|███████▋  | 8197/10682 [1:15:13<20:28,  2.02it/s]
 77%|███████▋  | 8198/10682 [1:15:13<20:28,  2.02it/s]
 77%|███████▋  | 8199/10682 [1:15:14<20:27,  2.02it/s]
 77%|███████▋  | 8200/10682 [1:15:14<20:27,  2.02it/s]
                                                      
+
 77%|███████▋  | 8200/10682 [1:15:14<20:27,  2.02it/s]
 77%|███████▋  | 8201/10682 [1:15:15<20:27,  2.02it/s]
 77%|███████▋  | 8202/10682 [1:15:15<20:25,  2.02it/s]
 77%|███████▋  | 8203/10682 [1:15:16<20:26,  2.02it/s]
 77%|███████▋  | 8204/10682 [1:15:16<20:25,  2.02it/s]
 77%|███████▋  | 8205/10682 [1:15:17<22:12,  1.86it/s]
 77%|██████��▋  | 8206/10682 [1:15:17<21:43,  1.90it/s]
 77%|███████▋  | 8207/10682 [1:15:18<21:18,  1.94it/s]
 77%|███████▋  | 8208/10682 [1:15:18<21:01,  1.96it/s]
 77%|███████▋  | 8209/10682 [1:15:19<20:49,  1.98it/s]
 77%|███████▋  | 8210/10682 [1:15:19<20:41,  1.99it/s]
 77%|███████▋  | 8211/10682 [1:15:20<20:35,  2.00it/s]
 77%|███████▋  | 8212/10682 [1:15:20<20:30,  2.01it/s]
 77%|███████▋  | 8213/10682 [1:15:21<20:27,  2.01it/s]
 77%|███████▋  | 8214/10682 [1:15:21<20:37,  1.99it/s]
 77%|███████▋  | 8215/10682 [1:15:22<20:32,  2.00it/s]
 77%|███████▋  | 8216/10682 [1:15:22<20:28,  2.01it/s]
 77%|███████▋  | 8217/10682 [1:15:23<20:26,  2.01it/s]
 77%|███████▋  | 8218/10682 [1:15:23<20:24,  2.01it/s]
 77%|███████▋  | 8219/10682 [1:15:24<20:23,  2.01it/s]
 77%|███████▋  | 8220/10682 [1:15:24<20:20,  2.02it/s]
 77%|███████▋  | 8221/10682 [1:15:25<20:20,  2.02it/s]
 77%|███████▋  | 8222/10682 [1:15:25<20:18,  2.02it/s]
 77%|███████▋  | 8223/10682 [1:15:26<20:17,  2.02it/s]
 77%|███████▋  | 8224/10682 [1:15:26<20:16,  2.02it/s]
 77%|███████▋  | 8225/10682 [1:15:27<20:16,  2.02it/s]{'loss': 2.9031, 'grad_norm': 0.24478811025619507, 'learning_rate': 0.0001527113038415231, 'epoch': 10.78}
+
                                                      
 77%|███████▋  | 8225/10682 [1:15:27<20:16,  2.02it/s]
 77%|███████▋  | 8226/10682 [1:15:27<20:17,  2.02it/s]
 77%|███████▋  | 8227/10682 [1:15:28<20:16,  2.02it/s]
 77%|███████▋  | 8228/10682 [1:15:28<20:15,  2.02it/s]
 77%|███████▋  | 8229/10682 [1:15:29<20:16,  2.02it/s]
 77%|███████▋  | 8230/10682 [1:15:29<20:15,  2.02it/s]
 77%|███████▋  | 8231/10682 [1:15:30<20:14,  2.02it/s]
 77%|███████▋  | 8232/10682 [1:15:30<20:12,  2.02it/s]
 77%|███████▋  | 8233/10682 [1:15:31<20:12,  2.02it/s]
 77%|███████▋  | 8234/10682 [1:15:31<20:10,  2.02it/s]
 77%|███████▋  | 8235/10682 [1:15:32<20:10,  2.02it/s]
 77%|███████▋  | 8236/10682 [1:15:32<20:09,  2.02it/s]
 77%|███████▋  | 8237/10682 [1:15:33<20:09,  2.02it/s]
 77%|███████▋  | 8238/10682 [1:15:33<20:08,  2.02it/s]
 77%|███████▋  | 8239/10682 [1:15:34<20:08,  2.02it/s]
 77%|███████▋  | 8240/10682 [1:15:34<20:07,  2.02it/s]
 77%|███████▋  | 8241/10682 [1:15:35<20:07,  2.02it/s]
 77%|███████▋  | 8242/10682 [1:15:35<20:05,  2.02it/s]
 77%|███████▋  | 8243/10682 [1:15:36<20:05,  2.02it/s]
 77%|███████▋  | 8244/10682 [1:15:36<20:04,  2.02it/s]
 77%|███████▋  | 8245/10682 [1:15:37<20:05,  2.02it/s]
 77%|███████▋  | 8246/10682 [1:15:37<20:04,  2.02it/s]
 77%|███████▋  | 8247/10682 [1:15:38<20:05,  2.02it/s]
 77%|███████▋  | 8248/10682 [1:15:38<20:03,  2.02it/s]
 77%|███████▋  | 8249/10682 [1:15:39<20:04,  2.02it/s]
 77%|███████▋  | 8250/10682 [1:15:39<20:03,  2.02it/s]{'loss': 2.8918, 'grad_norm': 0.24946050345897675, 'learning_rate': 0.00014978404491439802, 'epoch': 10.81}
                                                      
+
 77%|███████▋  | 8250/10682 [1:15:39<20:03,  2.02it/s]
 77%|███████▋  | 8251/10682 [1:15:40<20:04,  2.02it/s]
 77%|███████▋  | 8252/10682 [1:15:40<20:03,  2.02it/s]
 77%|███████▋  | 8253/10682 [1:15:41<20:03,  2.02it/s]
 77%|███████▋  | 8254/10682 [1:15:41<20:01,  2.02it/s]
 77%|███████▋  | 8255/10682 [1:15:41<20:02,  2.02it/s]
 77%|███████▋  | 8256/10682 [1:15:42<20:01,  2.02it/s]
 77%|███████▋  | 8257/10682 [1:15:42<20:01,  2.02it/s]
 77%|███████▋  | 8258/10682 [1:15:43<20:00,  2.02it/s]
 77%|███████▋  | 8259/10682 [1:15:43<20:00,  2.02it/s]
 77%|███████▋  | 8260/10682 [1:15:44<19:59,  2.02it/s]
 77%|███████▋  | 8261/10682 [1:15:44<19:59,  2.02it/s]
 77%|███████▋  | 8262/10682 [1:15:45<19:57,  2.02it/s]
 77%|███████▋  | 8263/10682 [1:15:45<19:57,  2.02it/s]
 77%|███████▋  | 8264/10682 [1:15:46<19:57,  2.02it/s]
 77%|███████▋  | 8265/10682 [1:15:46<19:56,  2.02it/s]
 77%|███████▋  | 8266/10682 [1:15:47<19:56,  2.02it/s]
 77%|███████▋  | 8267/10682 [1:15:47<19:54,  2.02it/s]
 77%|███████▋  | 8268/10682 [1:15:48<19:55,  2.02it/s]
 77%|███████▋  | 8269/10682 [1:15:48<19:52,  2.02it/s]
 77%|███████▋  | 8270/10682 [1:15:49<19:52,  2.02it/s]
 77%|███████▋  | 8271/10682 [1:15:49<19:51,  2.02it/s]
 77%|███████▋  | 8272/10682 [1:15:50<19:52,  2.02it/s]
 77%|███████▋  | 8273/10682 [1:15:50<19:51,  2.02it/s]
 77%|███████▋  | 8274/10682 [1:15:51<19:51,  2.02it/s]
 77%|███████▋  | 8275/10682 [1:15:51<19:50,  2.02it/s]
                                                      
+
 77%|███████▋  | 8275/10682 [1:15:51<19:50,  2.02it/s]
 77%|███████▋  | 8276/10682 [1:15:52<19:52,  2.02it/s]
 77%|███████▋  | 8277/10682 [1:15:52<19:51,  2.02it/s]
 77%|███████▋  | 8278/10682 [1:15:53<19:50,  2.02it/s]
 78%|███████▊  | 8279/10682 [1:15:53<19:50,  2.02it/s]
 78%|███████▊  | 8280/10682 [1:15:54<19:49,  2.02it/s]
 78%|███████▊  | 8281/10682 [1:15:54<19:48,  2.02it/s]
 78%|███████▊  | 8282/10682 [1:15:55<19:48,  2.02it/s]
 78%|███████▊  | 8283/10682 [1:15:55<19:47,  2.02it/s]
 78%|███████▊  | 8284/10682 [1:15:56<19:46,  2.02it/s]
 78%|███████▊  | 8285/10682 [1:15:56<19:46,  2.02it/s]
 78%|███████▊  | 8286/10682 [1:15:57<19:46,  2.02it/s]
 78%|███████▊  | 8287/10682 [1:15:57<19:47,  2.02it/s]
 78%|███████▊  | 8288/10682 [1:15:58<19:46,  2.02it/s]
 78%|███████▊  | 8289/10682 [1:15:58<19:47,  2.02it/s]
 78%|███████▊  | 8290/10682 [1:15:59<21:33,  1.85it/s]
 78%|███████▊  | 8291/10682 [1:15:59<20:58,  1.90it/s]
 78%|███████▊  | 8292/10682 [1:16:00<20:36,  1.93it/s]
 78%|███████▊  | 8293/10682 [1:16:00<20:18,  1.96it/s]
 78%|███████▊  | 8294/10682 [1:16:01<20:08,  1.98it/s]
 78%|███████▊  | 8295/10682 [1:16:01<19:59,  1.99it/s]
 78%|███████▊  | 8296/10682 [1:16:02<19:54,  2.00it/s]
 78%|███████▊  | 8297/10682 [1:16:02<19:49,  2.01it/s]
 78%|███████▊  | 8298/10682 [1:16:03<19:46,  2.01it/s]
 78%|███████▊  | 8299/10682 [1:16:03<19:42,  2.02it/s]
 78%|███████▊  | 8300/10682 [1:16:04<19:42,  2.01it/s]
                                                      {'loss': 2.8984, 'grad_norm': 0.24762478470802307, 'learning_rate': 0.00014399985296581835, 'epoch': 10.88}
+
 78%|███████▊  | 8300/10682 [1:16:04<19:42,  2.01it/s]
 78%|███████▊  | 8301/10682 [1:16:04<19:41,  2.01it/s]
 78%|███████▊  | 8302/10682 [1:16:05<19:40,  2.02it/s]
 78%|███████▊  | 8303/10682 [1:16:05<19:38,  2.02it/s]
 78%|███████▊  | 8304/10682 [1:16:06<19:38,  2.02it/s]
 78%|███████▊  | 8305/10682 [1:16:06<19:37,  2.02it/s]
 78%|███████▊  | 8306/10682 [1:16:07<19:36,  2.02it/s]
 78%|███████▊  | 8307/10682 [1:16:07<19:35,  2.02it/s]
 78%|███████▊  | 8308/10682 [1:16:08<19:35,  2.02it/s]
 78%|███████▊  | 8309/10682 [1:16:08<19:34,  2.02it/s]
 78%|███████▊  | 8310/10682 [1:16:09<19:34,  2.02it/s]
 78%|███████▊  | 8311/10682 [1:16:09<19:34,  2.02it/s]
 78%|███████▊  | 8312/10682 [1:16:10<19:33,  2.02it/s]
 78%|███████▊  | 8313/10682 [1:16:10<19:33,  2.02it/s]
 78%|███████▊  | 8314/10682 [1:16:11<19:31,  2.02it/s]
 78%|███████▊  | 8315/10682 [1:16:11<19:32,  2.02it/s]
 78%|███████▊  | 8316/10682 [1:16:12<19:29,  2.02it/s]
 78%|███████▊  | 8317/10682 [1:16:12<19:29,  2.02it/s]
 78%|███████▊  | 8318/10682 [1:16:13<19:28,  2.02it/s]
 78%|███████▊  | 8319/10682 [1:16:13<19:28,  2.02it/s]
 78%|███████▊  | 8320/10682 [1:16:14<19:28,  2.02it/s]
 78%|███████▊  | 8321/10682 [1:16:14<19:28,  2.02it/s]
 78%|███████▊  | 8322/10682 [1:16:15<19:28,  2.02it/s]
 78%|███████▊  | 8323/10682 [1:16:15<19:28,  2.02it/s]
 78%|███████▊  | 8324/10682 [1:16:16<19:26,  2.02it/s]
 78%|███████▊  | 8325/10682 [1:16:16<19:26,  2.02it/s]
                                                      
+
 78%|███████▊  | 8325/10682 [1:16:16<19:26,  2.02it/s]
 78%|███████▊  | 8326/10682 [1:16:17<19:26,  2.02it/s]
 78%|███████▊  | 8327/10682 [1:16:17<19:26,  2.02it/s]
 78%|███████▊  | 8328/10682 [1:16:18<19:25,  2.02it/s]
 78%|███████▊  | 8329/10682 [1:16:18<19:24,  2.02it/s]
 78%|███████▊  | 8330/10682 [1:16:19<19:22,  2.02it/s]
 78%|███████▊  | 8331/10682 [1:16:19<19:23,  2.02it/s]
 78%|███████▊  | 8332/10682 [1:16:20<19:22,  2.02it/s]
 78%|███████▊  | 8333/10682 [1:16:20<19:23,  2.02it/s]
 78%|███████▊  | 8334/10682 [1:16:21<19:24,  2.02it/s]
 78%|███████▊  | 8335/10682 [1:16:21<19:23,  2.02it/s]
 78%|███████▊  | 8336/10682 [1:16:22<19:22,  2.02it/s]
 78%|███████▊  | 8337/10682 [1:16:22<19:20,  2.02it/s]
 78%|███████▊  | 8338/10682 [1:16:23<19:21,  2.02it/s]
 78%|███████▊  | 8339/10682 [1:16:23<19:19,  2.02it/s]
 78%|███████▊  | 8340/10682 [1:16:24<19:20,  2.02it/s]
 78%|███████▊  | 8341/10682 [1:16:24<19:18,  2.02it/s]
 78%|███████▊  | 8342/10682 [1:16:25<19:19,  2.02it/s]
 78%|███████▊  | 8343/10682 [1:16:25<19:18,  2.02it/s]
 78%|███████▊  | 8344/10682 [1:16:26<19:17,  2.02it/s]
 78%|███████▊  | 8345/10682 [1:16:26<19:17,  2.02it/s]
 78%|███████▊  | 8346/10682 [1:16:27<19:17,  2.02it/s]
 78%|███████▊  | 8347/10682 [1:16:27<19:16,  2.02it/s]
 78%|███████▊  | 8348/10682 [1:16:28<19:17,  2.02it/s]
 78%|███████▊  | 8349/10682 [1:16:28<19:16,  2.02it/s]
 78%|███████▊  | 8350/10682 [1:16:29<19:15,  2.02it/s]
                                                      {'loss': 2.9034, 'grad_norm': 0.2513554096221924, 'learning_rate': 0.00013831071326327282, 'epoch': 10.94}
+
 78%|███████▊  | 8350/10682 [1:16:29<19:15,  2.02it/s]
 78%|███████▊  | 8351/10682 [1:16:29<19:16,  2.02it/s]
 78%|███████▊  | 8352/10682 [1:16:30<19:14,  2.02it/s]
 78%|███████▊  | 8353/10682 [1:16:30<19:14,  2.02it/s]
 78%|███████▊  | 8354/10682 [1:16:31<19:12,  2.02it/s]
 78%|███████▊  | 8355/10682 [1:16:31<19:13,  2.02it/s]
 78%|███████▊  | 8356/10682 [1:16:32<19:12,  2.02it/s]
 78%|███████▊  | 8357/10682 [1:16:32<19:11,  2.02it/s]
 78%|███████▊  | 8358/10682 [1:16:33<19:10,  2.02it/s]
 78%|███████▊  | 8359/10682 [1:16:33<19:09,  2.02it/s]
 78%|███████▊  | 8360/10682 [1:16:34<19:08,  2.02it/s]
 78%|███████▊  | 8361/10682 [1:16:34<19:08,  2.02it/s]
 78%|███████▊  | 8362/10682 [1:16:35<19:07,  2.02it/s]
 78%|███████▊  | 8363/10682 [1:16:35<19:06,  2.02it/s]
 78%|███████▊  | 8364/10682 [1:16:36<19:06,  2.02it/s]
 78%|███████▊  | 8365/10682 [1:16:36<19:05,  2.02it/s]
 78%|███████▊  | 8366/10682 [1:16:37<19:05,  2.02it/s]
 78%|███████▊  | 8367/10682 [1:16:37<19:04,  2.02it/s]
 78%|███████▊  | 8368/10682 [1:16:38<19:05,  2.02it/s]
 78%|███████▊  | 8369/10682 [1:16:38<19:03,  2.02it/s]
 78%|███████▊  | 8370/10682 [1:16:39<19:04,  2.02it/s]
 78%|███████▊  | 8371/10682 [1:16:39<19:02,  2.02it/s]
 78%|███████▊  | 8372/10682 [1:16:40<19:03,  2.02it/s]
 78%|███████▊  | 8373/10682 [1:16:40<19:02,  2.02it/s]
 78%|███████▊  | 8374/10682 [1:16:41<19:02,  2.02it/s]
 78%|███████▊  | 8375/10682 [1:16:41<19:01,  2.02it/s]
                                                      
+
 78%|███████▊  | 8375/10682 [1:16:41<19:01,  2.02it/s]
 78%|███████▊  | 8376/10682 [1:16:42<19:02,  2.02it/s]
 78%|███████▊  | 8377/10682 [1:16:42<19:01,  2.02it/s]
 78%|███████▊  | 8378/10682 [1:16:43<19:00,  2.02it/s]
 78%|███████▊  | 8379/10682 [1:16:43<18:58,  2.02it/s]
 78%|███████▊  | 8380/10682 [1:16:44<18:59,  2.02it/s]
 78%|███████▊  | 8381/10682 [1:16:44<18:58,  2.02it/s]
 78%|███████▊  | 8382/10682 [1:16:45<18:59,  2.02it/s]
 78%|███████▊  | 8383/10682 [1:16:45<18:56,  2.02it/s]
 78%|███████▊  | 8384/10682 [1:16:46<18:57,  2.02it/s]
 78%|███████▊  | 8385/10682 [1:16:46<18:56,  2.02it/s]
 79%|███████▊  | 8386/10682 [1:16:47<18:56,  2.02it/s]
 79%|███████▊  | 8387/10682 [1:16:47<18:55,  2.02it/s]
 79%|███████▊  | 8388/10682 [1:16:47<18:56,  2.02it/s]
 79%|███████▊  | 8389/10682 [1:16:48<18:55,  2.02it/s]
 79%|███████▊  | 8390/10682 [1:16:48<18:55,  2.02it/s]
 79%|███████▊  | 8391/10682 [1:16:49<18:53,  2.02it/s]
 79%|███████▊  | 8392/10682 [1:16:49<18:54,  2.02it/s]
 79%|███████▊  | 8393/10682 [1:16:50<18:40,  2.04it/s]
 79%|███████▊  | 8394/10682 [1:17:04<2:55:03,  4.59s/it]
 79%|███████▊  | 8395/10682 [1:17:05<2:08:08,  3.36s/it]
 79%|███████▊  | 8396/10682 [1:17:05<1:35:23,  2.50s/it]
 79%|███████▊  | 8397/10682 [1:17:06<1:12:24,  1.90s/it]
 79%|███████▊  | 8398/10682 [1:17:06<56:24,  1.48s/it]  
 79%|███████▊  | 8399/10682 [1:17:07<45:10,  1.19s/it]
 79%|███████▊  | 8400/10682 [1:17:07<37:22,  1.02it/s]
                                                      
+
 79%|███████▊  | 8400/10682 [1:17:07<37:22,  1.02it/s]
 79%|███████▊  | 8401/10682 [1:17:08<31:48,  1.20it/s]
 79%|███████▊  | 8402/10682 [1:17:08<27:51,  1.36it/s]
 79%|███████▊  | 8403/10682 [1:17:09<25:08,  1.51it/s]
 79%|███████▊  | 8404/10682 [1:17:09<23:11,  1.64it/s]
 79%|███████▊  | 8405/10682 [1:17:10<21:51,  1.74it/s]
 79%|███████▊  | 8406/10682 [1:17:10<20:54,  1.81it/s]
 79%|███████▊  | 8407/10682 [1:17:11<20:17,  1.87it/s]
 79%|███████▊  | 8408/10682 [1:17:11<19:51,  1.91it/s]
 79%|███████▊  | 8409/10682 [1:17:12<19:30,  1.94it/s]
 79%|███████▊  | 8410/10682 [1:17:12<19:19,  1.96it/s]
 79%|███████▊  | 8411/10682 [1:17:13<19:06,  1.98it/s]
 79%|███████▊  | 8412/10682 [1:17:13<18:59,  1.99it/s]
 79%|███████▉  | 8413/10682 [1:17:14<18:53,  2.00it/s]
 79%|███████▉  | 8414/10682 [1:17:14<18:50,  2.01it/s]
 79%|███████▉  | 8415/10682 [1:17:15<18:49,  2.01it/s]
 79%|███████▉  | 8416/10682 [1:17:15<18:46,  2.01it/s]
 79%|███████▉  | 8417/10682 [1:17:16<18:43,  2.02it/s]
 79%|███████▉  | 8418/10682 [1:17:16<18:42,  2.02it/s]
 79%|███████▉  | 8419/10682 [1:17:17<18:39,  2.02it/s]
 79%|███████▉  | 8420/10682 [1:17:17<18:39,  2.02it/s]