diff --git "a/train_job_output.txt" "b/train_job_output.txt"
--- "a/train_job_output.txt"
+++ "b/train_job_output.txt"
@@ -330,4 +330,65 @@ Retrying in 1s [Retry 1/5].
 
 57%|█████▋    | 6050/10682 [51:25<38:06,  2.03it/s]
 57%|█████▋    | 6051/10682 [51:25<38:08,  2.02it/s]
 57%|█████▋    | 6052/10682 [51:26<38:08,  2.02it/s]
 57%|█████▋    | 6053/10682 [51:26<38:07,  2.02it/s]
 57%|█████▋    | 6054/10682 [51:27<38:07,  2.02it/s]
 57%|█████▋    | 6055/10682 [51:27<38:04,  2.03it/s]
 57%|█████▋    | 6056/10682 [51:28<38:06,  2.02it/s]
 57%|█████▋    | 6057/10682 [51:28<38:03,  2.03it/s]
 57%|█████▋    | 6058/10682 [51:29<38:04,  2.02it/s]
 57%|█████▋    | 6059/10682 [51:29<38:02,  2.03it/s]
 57%|█████▋    | 6060/10682 [51:30<38:02,  2.02it/s]
 57%|█████▋    | 6061/10682 [51:30<38:02,  2.02it/s]
 57%|█████▋    | 6062/10682 [51:31<38:02,  2.02it/s]
 57%|█████▋    | 6063/10682 [51:31<38:02,  2.02it/s]
 57%|█████▋    | 6064/10682 [51:32<38:01,  2.02it/s]
 57%|█████▋    | 6065/10682 [51:32<38:01,  2.02it/s]
 57%|█████▋    | 6066/10682 [51:33<37:59,  2.02it/s]
 57%|█████▋    | 6067/10682 [51:33<38:00,  2.02it/s]
 57%|█████▋    | 6068/10682 [51:34<37:58,  2.03it/s]
 57%|█████▋    | 6069/10682 [51:34<37:59,  2.02it/s]
 57%|█████▋    | 6070/10682 [51:35<37:55,  2.03it/s]
 57%|█████▋    | 6071/10682 [51:35<37:57,  2.02it/s]
 57%|█████▋    | 6072/10682 [51:36<37:55,  2.03it/s]
 57%|█████▋    | 6073/10682 [51:36<37:57,  2.02it/s]
 57%|█████▋    | 6074/10682 [51:37<37:56,  2.02it/s]
 57%|█████▋    | 6075/10682 [51:37<37:55,  2.02it/s]
                                                    {'loss': 3.0987, 'grad_norm': 0.2673705220222473, 'learning_rate': 0.0004674241234809751, 'epoch': 7.96}
 
 57%|█████▋    | 6075/10682 [51:37<37:55,  2.02it/s]
 57%|█████▋    | 6076/10682 [51:38<37:55,  2.02it/s]
 57%|█████▋    | 6077/10682 [51:38<37:53,  2.03it/s]
 57%|█████▋    | 6078/10682 [51:39<37:54,  2.02it/s]
 57%|█████▋    | 6079/10682 [51:39<37:55,  2.02it/s]
 57%|█████▋    | 6080/10682 [51:40<37:54,  2.02it/s]
 57%|█████▋    | 6081/10682 [51:40<37:52,  2.02it/s]
 57%|█████▋    | 6082/10682 [51:41<37:54,  2.02it/s]
 57%|█████▋    | 6083/10682 [51:41<37:51,  2.03it/s]
 57%|█████▋    | 6084/10682 [51:42<37:52,  2.02it/s]
 57%|█████▋    | 6085/10682 [51:42<37:53,  2.02it/s]
 57%|█████▋    | 6086/10682 [51:43<37:53,  2.02it/s]
 57%|█████▋    | 6087/10682 [51:43<37:51,  2.02it/s]
 57%|█████▋    | 6088/10682 [51:44<37:51,  2.02it/s]
 57%|█████▋    | 6089/10682 [51:44<37:48,  2.02it/s]
 57%|█████▋    | 6090/10682 [51:45<37:47,  2.02it/s]
 57%|█████▋    | 6091/10682 [51:45<37:46,  2.03it/s]
 57%|█████▋    | 6092/10682 [51:46<37:46,  2.02it/s]
 57%|█████▋    | 6093/10682 [51:46<37:47,  2.02it/s]
 57%|█████▋    | 6094/10682 [51:47<37:45,  2.03it/s]
 57%|█████▋    | 6095/10682 [51:47<37:47,  2.02it/s]
 57%|█████▋    | 6096/10682 [51:48<37:44,  2.03it/s]
 57%|█████▋    | 6097/10682 [51:48<37:45,  2.02it/s]
 57%|█████▋    | 6098/10682 [51:49<37:42,  2.03it/s]
 57%|█████▋    | 6099/10682 [51:49<37:44,  2.02it/s]
 57%|█████▋    | 6100/10682 [51:50<37:43,  2.02it/s]
                                                    {'loss': 3.1022, 'grad_norm': 0.253995418548584, 'learning_rate': 0.0004633488518231793, 'epoch': 7.99}
 
 57%|█████▋    | 6100/10682 [51:50<37:43,  2.02it/s]
 57%|█████▋    | 6101/10682 [51:50<37:44,  2.02it/s]
 57%|█████▋    | 6102/10682 [51:51<37:44,  2.02it/s]
 57%|█████▋    | 6103/10682 [51:51<37:44,  2.02it/s]
 57%|█████▋    | 6104/10682 [51:52<37:43,  2.02it/s]
 57%|█████▋    | 6105/10682 [51:52<37:41,  2.02it/s]
 57%|█████▋    | 6106/10682 [51:53<37:41,  2.02it/s]
 57%|█████▋    | 6107/10682 [51:53<37:39,  2.02it/s]
 57%|█████▋    | 6108/10682 [51:54<37:15,  2.05it/s]
 57%|█████▋    | 6109/10682 [52:05<4:57:55,  3.91s/it]
 57%|█████▋    | 6110/10682 [52:06<3:39:59,  2.89s/it]
 57%|█████▋    | 6111/10682 [52:06<2:45:15,  2.17s/it]
 57%|█████▋    | 6112/10682 [52:07<2:06:55,  1.67s/it]
 57%|█████▋    | 6113/10682 [52:07<1:40:09,  1.32s/it]
 57%|█████▋    | 6114/10682 [52:08<1:21:22,  1.07s/it]
 57%|█████▋    | 6115/10682 [52:08<1:08:12,  1.12it/s]
 57%|█████▋    | 6116/10682 [52:09<59:00,  1.29it/s]  
 57%|█████▋    | 6117/10682 [52:09<52:32,  1.45it/s]
 57%|█████▋    | 6118/10682 [52:10<48:02,  1.58it/s]
 57%|█████▋    | 6119/10682 [52:10<44:54,  1.69it/s]
 57%|█████▋    | 6120/10682 [52:11<42:40,  1.78it/s]
 57%|█████▋    | 6121/10682 [52:11<41:08,  1.85it/s]
 57%|█████▋    | 6122/10682 [52:12<40:03,  1.90it/s]
 57%|█████▋    | 6123/10682 [52:12<39:18,  1.93it/s]
 57%|█████▋    | 6124/10682 [52:13<38:51,  1.96it/s]
 57%|█████▋    | 6125/10682 [52:13<38:27,  1.97it/s]
                                                    
{'loss': 2.9833, 'grad_norm': 0.2493210732936859, 'learning_rate': 0.00045927602667567333, 'epoch': 8.02}
-
 57%|█████▋    | 6125/10682 [52:13<38:27,  1.97it/s]
 57%|█████▋    | 6126/10682 [52:14<38:11,  1.99it/s]
 57%|█████▋    | 6127/10682 [52:14<37:58,  2.00it/s]
 57%|█████▋    | 6128/10682 [52:15<37:52,  2.00it/s]
 57%|█████▋    | 6129/10682 [52:15<37:48,  2.01it/s]
 57%|█████▋    | 6130/10682 [52:16<37:42,  2.01it/s]
 57%|█████▋    | 6131/10682 [52:16<37:36,  2.02it/s]
 57%|█████▋    | 6132/10682 [52:17<37:32,  2.02it/s]
\ No newline at end of file
+
 57%|█████▋    | 6125/10682 [52:13<38:27,  1.97it/s]
 57%|█████▋    | 6126/10682 [52:14<38:11,  1.99it/s]
 57%|█████▋    | 6127/10682 [52:14<37:58,  2.00it/s]
 57%|█████▋    | 6128/10682 [52:15<37:52,  2.00it/s]
 57%|█████▋    | 6129/10682 [52:15<37:48,  2.01it/s]
 57%|█████▋    | 6130/10682 [52:16<37:42,  2.01it/s]
 57%|█████▋    | 6131/10682 [52:16<37:36,  2.02it/s]
 57%|█████▋    | 6132/10682 [52:17<37:32,  2.02it/s]
 57%|█████▋    | 6133/10682 [52:17<37:28,  2.02it/s]
 57%|█████▋    | 6134/10682 [52:18<37:27,  2.02it/s]
 57%|█████▋    | 6135/10682 [52:18<37:25,  2.02it/s]
 57%|█████▋    | 6136/10682 [52:19<37:24,  2.03it/s]
 57%|█████▋    | 6137/10682 [52:19<37:20,  2.03it/s]
 57%|█████▋    | 6138/10682 [52:20<37:20,  2.03it/s]
 57%|█████▋    | 6139/10682 [52:20<37:20,  2.03it/s]
 57%|█████▋    | 6140/10682 [52:21<37:21,  2.03it/s]
 57%|█████▋    | 6141/10682 [52:21<37:20,  2.03it/s]
 57%|█████▋    | 6142/10682 [52:22<37:17,  2.03it/s]
 58%|█████▊    | 6143/10682 [52:22<37:18,  2.03it/s]
 58%|█████▊    | 6144/10682 [52:23<37:18,  2.03it/s]
 58%|█████▊    | 6145/10682 [52:23<37:21,  2.02it/s]
 58%|█████▊    | 6146/10682 [52:24<37:21,  2.02it/s]
 58%|█████▊    | 6147/10682 [52:24<37:21,  2.02it/s]
 58%|█████▊    | 6148/10682 [52:25<37:19,  2.02it/s]
 58%|█████▊    | 6149/10682 [52:25<37:18,  2.02it/s]
 58%|█████▊    | 6150/10682 [52:26<37:15,  2.03it/s]
                                                    {'loss': 2.9362, 'grad_norm': 0.2625344395637512, 'learning_rate': 0.00045520591990466365, 'epoch': 8.06}
+

 58%|█████▊    | 6150/10682 [52:26<37:15,  2.03it/s]
 58%|█████▊    | 6151/10682 [52:26<37:19,  2.02it/s]
 58%|█████▊    | 6152/10682 [52:27<37:16,  2.03it/s]
 58%|█████▊    | 6153/10682 [52:27<37:18,  2.02it/s]
 58%|█████▊    | 6154/10682 [52:28<37:15,  2.03it/s]
 58%|█████▊    | 6155/10682 [52:28<37:13,  2.03it/s]
 58%|█████▊    | 6156/10682 [52:29<37:12,  2.03it/s]
 58%|█████▊    | 6157/10682 [52:29<37:10,  2.03it/s]
 58%|█████▊    | 6158/10682 [52:30<37:11,  2.03it/s]
 58%|█████▊    | 6159/10682 [52:30<37:10,  2.03it/s]
 58%|█████▊    | 6160/10682 [52:31<37:12,  2.03it/s]
 58%|█████▊    | 6161/10682 [52:31<37:09,  2.03it/s]
 58%|█████▊    | 6162/10682 [52:32<37:11,  2.03it/s]
 58%|█████▊    | 6163/10682 [52:32<37:11,  2.03it/s]
 58%|█████▊    | 6164/10682 [52:33<37:11,  2.02it/s]
 58%|█████▊    | 6165/10682 [52:33<37:11,  2.02it/s]
 58%|█████▊    | 6166/10682 [52:34<37:12,  2.02it/s]
 58%|█████▊    | 6167/10682 [52:34<37:11,  2.02it/s]
 58%|█████▊    | 6168/10682 [52:35<37:09,  2.02it/s]
 58%|█████▊    | 6169/10682 [52:35<37:09,  2.02it/s]
 58%|█████▊    | 6170/10682 [52:36<37:05,  2.03it/s]
 58%|█████▊    | 6171/10682 [52:36<37:07,  2.03it/s]
 58%|█████▊    | 6172/10682 [52:37<37:04,  2.03it/s]
 58%|█████▊    | 6173/10682 [52:37<37:07,  2.02it/s]
 58%|█████▊    | 6174/10682 [52:38<37:06,  2.03it/s]
 58%|█████▊    | 6175/10682 [52:38<37:06,  2.02it/s]{'loss': 2.9549, 'grad_norm': 0.2687247395515442, 'learning_rate': 0.00045113880319490264, 'epoch': 8.09}
                                                    
+
 58%|█████▊    | 6175/10682 [52:38<37:06,  2.02it/s]
 58%|█████▊    | 6176/10682 [52:38<37:04,  2.03it/s]
 58%|█████▊    | 6177/10682 [52:39<37:03,  2.03it/s]
 58%|█████▊    | 6178/10682 [52:39<37:02,  2.03it/s]
 58%|█████▊    | 6179/10682 [52:40<37:00,  2.03it/s]
 58%|█████▊    | 6180/10682 [52:40<37:01,  2.03it/s]
 58%|█████▊    | 6181/10682 [52:41<37:01,  2.03it/s]
 58%|█████▊    | 6182/10682 [52:41<37:02,  2.03it/s]
 58%|█████▊    | 6183/10682 [52:42<37:03,  2.02it/s]
 58%|█████▊    | 6184/10682 [52:42<37:03,  2.02it/s]
 58%|█████▊    | 6185/10682 [52:43<37:02,  2.02it/s]
 58%|█████▊    | 6186/10682 [52:43<37:03,  2.02it/s]
 58%|█████▊    | 6187/10682 [52:44<37:01,  2.02it/s]
 58%|█████▊    | 6188/10682 [52:44<37:01,  2.02it/s]
 58%|█████▊    | 6189/10682 [52:45<36:56,  2.03it/s]
 58%|█████▊    | 6190/10682 [52:45<36:56,  2.03it/s]
 58%|█████▊    | 6191/10682 [52:46<36:55,  2.03it/s]
 58%|█████▊    | 6192/10682 [52:46<36:55,  2.03it/s]
 58%|█████▊    | 6193/10682 [52:47<36:55,  2.03it/s]
 58%|█████▊    | 6194/10682 [52:47<36:56,  2.02it/s]
 58%|█████▊    | 6195/10682 [52:48<36:56,  2.02it/s]
 58%|█████▊    | 6196/10682 [52:48<36:56,  2.02it/s]
 58%|█████▊    | 6197/10682 [52:49<36:57,  2.02it/s]
 58%|█████▊    | 6198/10682 [52:49<36:54,  2.02it/s]
 58%|█████▊    | 6199/10682 [52:50<36:54,  2.02it/s]
 58%|█████▊    | 6200/10682 [52:50<36:54,  2.02it/s]{'loss': 2.9595, 'grad_norm': 0.26444023847579956, 'learning_rate': 0.00044707494803155137, 'epoch': 8.12}
                                                    
+
 58%|█████▊    | 6200/10682 [52:50<36:54,  2.02it/s]
 58%|█████▊    | 6201/10682 [52:51<36:57,  2.02it/s]
 58%|█████▊    | 6202/10682 [52:51<36:56,  2.02it/s]
 58%|█████▊    | 6203/10682 [52:52<36:56,  2.02it/s]
 58%|█████▊    | 6204/10682 [52:52<36:55,  2.02it/s]
 58%|█████▊    | 6205/10682 [52:53<36:53,  2.02it/s]
 58%|█████▊    | 6206/10682 [52:53<36:53,  2.02it/s]
 58%|█████▊    | 6207/10682 [52:54<36:52,  2.02it/s]
 58%|█████▊    | 6208/10682 [52:54<36:51,  2.02it/s]
 58%|█████▊    | 6209/10682 [52:55<36:49,  2.02it/s]
 58%|█████▊    | 6210/10682 [52:55<36:49,  2.02it/s]
 58%|█████▊    | 6211/10682 [52:56<36:48,  2.02it/s]
 58%|█████▊    | 6212/10682 [52:56<36:48,  2.02it/s]
 58%|█████▊    | 6213/10682 [52:57<36:46,  2.03it/s]
 58%|█████▊    | 6214/10682 [52:57<36:45,  2.03it/s]
 58%|█████▊    | 6215/10682 [52:58<36:44,  2.03it/s]
 58%|█████▊    | 6216/10682 [52:58<36:44,  2.03it/s]
 58%|█████▊    | 6217/10682 [52:59<36:44,  2.03it/s]
 58%|█████▊    | 6218/10682 [52:59<36:44,  2.02it/s]
 58%|█████▊    | 6219/10682 [53:00<36:44,  2.02it/s]
 58%|█████▊    | 6220/10682 [53:00<36:44,  2.02it/s]
 58%|█████▊    | 6221/10682 [53:01<36:45,  2.02it/s]
 58%|█████▊    | 6222/10682 [53:01<36:44,  2.02it/s]
 58%|█████▊    | 6223/10682 [53:02<36:44,  2.02it/s]
 58%|█████▊    | 6224/10682 [53:02<36:42,  2.02it/s]
 58%|█████▊    | 6225/10682 [53:03<36:44,  2.02it/s]
                                                    {'loss': 2.9675, 'grad_norm': 0.2638459801673889, 'learning_rate': 0.00044301462568205915, 'epoch': 8.15}
+
 58%|█████▊    | 6225/10682 [53:03<36:44,  2.02it/s]
 58%|█████▊    | 6226/10682 [53:03<36:45,  2.02it/s]
 58%|█████▊    | 6227/10682 [53:04<36:42,  2.02it/s]
 58%|█████▊    | 6228/10682 [53:04<36:42,  2.02it/s]
 58%|█████▊    | 6229/10682 [53:05<36:41,  2.02it/s]
 58%|█████▊    | 6230/10682 [53:05<36:39,  2.02it/s]
 58%|█████▊    | 6231/10682 [53:06<36:38,  2.02it/s]
 58%|█████▊    | 6232/10682 [53:06<36:36,  2.03it/s]
 58%|█████▊    | 6233/10682 [53:07<36:38,  2.02it/s]
 58%|█████▊    | 6234/10682 [53:07<36:36,  2.02it/s]
 58%|█████▊    | 6235/10682 [53:08<36:36,  2.02it/s]
 58%|█████▊    | 6236/10682 [53:08<36:35,  2.02it/s]
 58%|█████▊    | 6237/10682 [53:09<36:35,  2.02it/s]
 58%|█████▊    | 6238/10682 [53:09<36:34,  2.02it/s]
 58%|█████▊    | 6239/10682 [53:10<36:32,  2.03it/s]
 58%|█████▊    | 6240/10682 [53:10<36:33,  2.02it/s]
 58%|█████▊    | 6241/10682 [53:11<36:31,  2.03it/s]
 58%|█████▊    | 6242/10682 [53:11<36:33,  2.02it/s]
 58%|█████▊    | 6243/10682 [53:12<36:31,  2.03it/s]
 58%|█████▊    | 6244/10682 [53:12<36:34,  2.02it/s]
 58%|█████▊    | 6245/10682 [53:13<36:31,  2.02it/s]
 58%|█████▊    | 6246/10682 [53:13<36:32,  2.02it/s]
 58%|█████▊    | 6247/10682 [53:14<36:30,  2.02it/s]
 58%|█████▊    | 6248/10682 [53:14<36:30,  2.02it/s]
 59%|█████▊    | 6249/10682 [53:15<36:29,  2.02it/s]
 59%|█████▊    | 6250/10682 [53:15<36:28,  2.03it/s]
                                                    {'loss': 2.9643, 'grad_norm': 0.2715643346309662, 'learning_rate': 0.0004389581071780553, 'epoch': 8.19}
+
 59%|█████▊    | 6250/10682 [53:15<36:28,  2.03it/s]
 59%|█████▊    | 6251/10682 [53:16<36:32,  2.02it/s]
 59%|█████▊    | 6252/10682 [53:16<36:31,  2.02it/s]
 59%|█████▊    | 6253/10682 [53:17<36:31,  2.02it/s]
 59%|█████▊    | 6254/10682 [53:17<36:29,  2.02it/s]
 59%|█████▊    | 6255/10682 [53:18<36:26,  2.02it/s]
 59%|█████▊    | 6256/10682 [53:18<36:27,  2.02it/s]
 59%|█████▊    | 6257/10682 [53:19<36:24,  2.03it/s]
 59%|█████▊    | 6258/10682 [53:19<36:25,  2.02it/s]
 59%|█████▊    | 6259/10682 [53:19<36:24,  2.02it/s]
 59%|█████▊    | 6260/10682 [53:20<36:21,  2.03it/s]
 59%|█████▊    | 6261/10682 [53:20<36:22,  2.03it/s]
 59%|█████▊    | 6262/10682 [53:21<36:20,  2.03it/s]
 59%|█████▊    | 6263/10682 [53:21<36:22,  2.02it/s]
 59%|█████▊    | 6264/10682 [53:22<36:20,  2.03it/s]
 59%|█████▊    | 6265/10682 [53:22<36:22,  2.02it/s]
 59%|█████▊    | 6266/10682 [53:23<36:20,  2.03it/s]
 59%|█████▊    | 6267/10682 [53:23<36:19,  2.03it/s]
 59%|█████▊    | 6268/10682 [53:24<36:19,  2.03it/s]
 59%|█████▊    | 6269/10682 [53:24<36:16,  2.03it/s]
 59%|█████▊    | 6270/10682 [53:25<36:18,  2.03it/s]
 59%|█████▊    | 6271/10682 [53:25<36:16,  2.03it/s]
 59%|█████▊    | 6272/10682 [53:26<36:17,  2.02it/s]
 59%|█████▊    | 6273/10682 [53:26<36:16,  2.03it/s]
 59%|█████▊    | 6274/10682 [53:27<36:18,  2.02it/s]
 59%|█████▊    | 6275/10682 [53:27<36:18,  2.02it/s]
                                                    {'loss': 2.979, 'grad_norm': 0.26490533351898193, 'learning_rate': 0.0004349056632972576, 'epoch': 8.22}
+

 59%|█████▊    | 6275/10682 [53:27<36:18,  2.02it/s]
 59%|█████▉    | 6276/10682 [53:28<36:20,  2.02it/s]
 59%|█████▉    | 6277/10682 [53:28<36:18,  2.02it/s]
 59%|█████▉    | 6278/10682 [53:29<36:16,  2.02it/s]
 59%|█████▉    | 6279/10682 [53:29<36:14,  2.03it/s]
 59%|█████▉    | 6280/10682 [53:30<36:14,  2.02it/s]
 59%|█████▉    | 6281/10682 [53:30<36:11,  2.03it/s]
 59%|█████▉    | 6282/10682 [53:31<36:11,  2.03it/s]
 59%|█████▉    | 6283/10682 [53:31<36:11,  2.03it/s]
 59%|█████▉    | 6284/10682 [53:32<36:10,  2.03it/s]
 59%|█████▉    | 6285/10682 [53:32<36:11,  2.02it/s]
 59%|█████▉    | 6286/10682 [53:33<36:07,  2.03it/s]
 59%|█████▉    | 6287/10682 [53:33<36:09,  2.03it/s]
 59%|█████▉    | 6288/10682 [53:34<36:08,  2.03it/s]
 59%|█████▉    | 6289/10682 [53:34<36:10,  2.02it/s]
 59%|█████▉    | 6290/10682 [53:35<36:08,  2.03it/s]
 59%|█████▉    | 6291/10682 [53:35<36:10,  2.02it/s]
 59%|█████▉    | 6292/10682 [53:36<36:09,  2.02it/s]
 59%|█████▉    | 6293/10682 [53:36<36:10,  2.02it/s]
 59%|█████▉    | 6294/10682 [53:37<36:05,  2.03it/s]
 59%|█████▉    | 6295/10682 [53:37<36:06,  2.03it/s]
 59%|█████▉    | 6296/10682 [53:38<39:14,  1.86it/s]
 59%|█████▉    | 6297/10682 [53:38<38:19,  1.91it/s]
 59%|█████▉    | 6298/10682 [53:39<37:37,  1.94it/s]
 59%|█████▉    | 6299/10682 [53:39<37:11,  1.96it/s]
 59%|█████▉    | 6300/10682 [53:40<36:49,  1.98it/s]{'loss': 2.9683, 'grad_norm': 0.2625379264354706, 'learning_rate': 0.00043085756454539745, 'epoch': 8.25}
+
                                                    

 59%|█████▉    | 6300/10682 [53:40<36:49,  1.98it/s]
 59%|█████▉    | 6301/10682 [53:40<36:37,  1.99it/s]
 59%|█████▉    | 6302/10682 [53:41<36:26,  2.00it/s]
 59%|█████▉    | 6303/10682 [53:41<36:19,  2.01it/s]
 59%|█████▉    | 6304/10682 [53:42<36:13,  2.01it/s]
 59%|█████▉    | 6305/10682 [53:42<36:10,  2.02it/s]
 59%|█████▉    | 6306/10682 [53:43<36:07,  2.02it/s]
 59%|█████▉    | 6307/10682 [53:43<36:03,  2.02it/s]
 59%|█████▉    | 6308/10682 [53:44<36:03,  2.02it/s]
 59%|█████▉    | 6309/10682 [53:44<35:59,  2.02it/s]
 59%|█████▉    | 6310/10682 [53:45<36:01,  2.02it/s]
 59%|█████▉    | 6311/10682 [53:45<35:58,  2.03it/s]
 59%|█████▉    | 6312/10682 [53:46<36:00,  2.02it/s]
 59%|█████▉    | 6313/10682 [53:46<35:58,  2.02it/s]
 59%|█████▉    | 6314/10682 [53:47<35:58,  2.02it/s]
 59%|█████▉    | 6315/10682 [53:47<35:56,  2.02it/s]
 59%|█████▉    | 6316/10682 [53:48<35:56,  2.02it/s]
 59%|█████▉    | 6317/10682 [53:48<35:55,  2.03it/s]
 59%|█████▉    | 6318/10682 [53:49<35:54,  2.03it/s]
 59%|█████▉    | 6319/10682 [53:49<35:55,  2.02it/s]
 59%|█████▉    | 6320/10682 [53:50<35:53,  2.03it/s]
 59%|█████▉    | 6321/10682 [53:50<35:54,  2.02it/s]
 59%|█████▉    | 6322/10682 [53:51<35:51,  2.03it/s]
 59%|█████▉    | 6323/10682 [53:51<35:54,  2.02it/s]
 59%|█████▉    | 6324/10682 [53:52<35:51,  2.03it/s]
 59%|█████▉    | 6325/10682 [53:52<35:51,  2.02it/s]
                                                    {'loss': 2.9921, 'grad_norm': 0.2598678171634674, 'learning_rate': 0.0004268140811381641, 'epoch': 8.28}
+
 59%|█████▉    | 6325/10682 [53:52<35:51,  2.02it/s]
 59%|█████▉    | 6326/10682 [53:53<35:52,  2.02it/s]
 59%|█████▉    | 6327/10682 [53:53<35:53,  2.02it/s]
 59%|█████▉    | 6328/10682 [53:54<35:50,  2.02it/s]
 59%|█████▉    | 6329/10682 [53:54<35:48,  2.03it/s]
 59%|█████▉    | 6330/10682 [53:55<35:49,  2.02it/s]
 59%|█████▉    | 6331/10682 [53:55<35:47,  2.03it/s]
 59%|█████▉    | 6332/10682 [53:56<35:48,  2.02it/s]
 59%|█████▉    | 6333/10682 [53:56<35:46,  2.03it/s]
 59%|█████▉    | 6334/10682 [53:57<35:47,  2.02it/s]
 59%|█████▉    | 6335/10682 [53:57<35:44,  2.03it/s]
 59%|█████▉    | 6336/10682 [53:58<35:43,  2.03it/s]
 59%|█████▉    | 6337/10682 [53:58<35:44,  2.03it/s]
 59%|█████▉    | 6338/10682 [53:59<35:44,  2.03it/s]
 59%|█████▉    | 6339/10682 [53:59<35:43,  2.03it/s]
 59%|█████▉    | 6340/10682 [54:00<35:41,  2.03it/s]
 59%|█████▉    | 6341/10682 [54:00<35:42,  2.03it/s]
 59%|█████▉    | 6342/10682 [54:01<35:40,  2.03it/s]
 59%|█████▉    | 6343/10682 [54:01<35:41,  2.03it/s]
 59%|█████▉    | 6344/10682 [54:02<35:39,  2.03it/s]
 59%|█████▉    | 6345/10682 [54:02<35:38,  2.03it/s]
 59%|█████▉    | 6346/10682 [54:03<35:39,  2.03it/s]
 59%|█████▉    | 6347/10682 [54:03<35:37,  2.03it/s]
 59%|█████▉    | 6348/10682 [54:04<35:37,  2.03it/s]
 59%|█████▉    | 6349/10682 [54:04<35:36,  2.03it/s]
 59%|█████▉    | 6350/10682 [54:05<35:42,  2.02it/s]{'loss': 2.9826, 'grad_norm': 0.2634151875972748, 'learning_rate': 0.0004227754829831662, 'epoch': 8.32}
                                                    
+
 59%|█████▉    | 6350/10682 [54:05<35:42,  2.02it/s]
 59%|█████▉    | 6351/10682 [54:05<35:41,  2.02it/s]
 59%|█████▉    | 6352/10682 [54:06<35:39,  2.02it/s]
 59%|█████▉    | 6353/10682 [54:06<35:38,  2.02it/s]
 59%|█████▉    | 6354/10682 [54:07<35:36,  2.03it/s]
 59%|█████▉    | 6355/10682 [54:07<35:38,  2.02it/s]
 60%|█████▉    | 6356/10682 [54:08<35:35,  2.03it/s]
 60%|█████▉    | 6357/10682 [54:08<35:35,  2.02it/s]
 60%|█████▉    | 6358/10682 [54:09<35:33,  2.03it/s]
 60%|█████▉    | 6359/10682 [54:09<35:33,  2.03it/s]
 60%|█████▉    | 6360/10682 [54:10<35:32,  2.03it/s]
 60%|█████▉    | 6361/10682 [54:10<35:32,  2.03it/s]
 60%|█████▉    | 6362/10682 [54:11<35:32,  2.03it/s]
 60%|█████▉    | 6363/10682 [54:11<35:30,  2.03it/s]
 60%|█████▉    | 6364/10682 [54:11<35:31,  2.03it/s]
 60%|█████▉    | 6365/10682 [54:12<35:32,  2.02it/s]
 60%|█████▉    | 6366/10682 [54:12<35:32,  2.02it/s]
 60%|█████▉    | 6367/10682 [54:13<35:30,  2.03it/s]
 60%|█████▉    | 6368/10682 [54:13<35:30,  2.02it/s]
 60%|█████▉    | 6369/10682 [54:14<35:28,  2.03it/s]
 60%|█████▉    | 6370/10682 [54:14<35:30,  2.02it/s]
 60%|█████▉    | 6371/10682 [54:15<35:29,  2.02it/s]
 60%|█████▉    | 6372/10682 [54:15<35:29,  2.02it/s]
 60%|█████▉    | 6373/10682 [54:16<35:30,  2.02it/s]
 60%|█████▉    | 6374/10682 [54:17<38:32,  1.86it/s]
 60%|█████▉    | 6375/10682 [54:17<37:36,  1.91it/s]{'loss': 2.9961, 'grad_norm': 0.2710336744785309, 'learning_rate': 0.0004187420396619156, 'epoch': 8.35}
                                                    
+
 60%|█████▉    | 6375/10682 [54:17<37:36,  1.91it/s]
 60%|█████▉    | 6376/10682 [54:18<36:59,  1.94it/s]
 60%|█████▉    | 6377/10682 [54:18<36:31,  1.96it/s]
 60%|█████▉    | 6378/10682 [54:19<36:13,  1.98it/s]
 60%|█████▉    | 6379/10682 [54:19<35:59,  1.99it/s]
 60%|█████▉    | 6380/10682 [54:20<35:50,  2.00it/s]
 60%|█████▉    | 6381/10682 [54:20<35:43,  2.01it/s]
 60%|█████▉    | 6382/10682 [54:21<35:40,  2.01it/s]
 60%|█████▉    | 6383/10682 [54:21<35:35,  2.01it/s]
 60%|█████▉    | 6384/10682 [54:22<35:31,  2.02it/s]
 60%|█████▉    | 6385/10682 [54:22<35:27,  2.02it/s]
 60%|█████▉    | 6386/10682 [54:23<35:28,  2.02it/s]
 60%|█████▉    | 6387/10682 [54:23<35:25,  2.02it/s]
 60%|█████▉    | 6388/10682 [54:23<35:24,  2.02it/s]
 60%|█████▉    | 6389/10682 [54:24<35:23,  2.02it/s]
 60%|█████▉    | 6390/10682 [54:24<35:20,  2.02it/s]
 60%|█████▉    | 6391/10682 [54:25<35:21,  2.02it/s]
 60%|█████▉    | 6392/10682 [54:25<35:20,  2.02it/s]
 60%|█████▉    | 6393/10682 [54:26<35:19,  2.02it/s]
 60%|█████▉    | 6394/10682 [54:26<35:19,  2.02it/s]
 60%|█████▉    | 6395/10682 [54:27<35:17,  2.02it/s]
 60%|█████▉    | 6396/10682 [54:27<35:21,  2.02it/s]
 60%|█████▉    | 6397/10682 [54:28<35:17,  2.02it/s]
 60%|█████▉    | 6398/10682 [54:28<35:17,  2.02it/s]
 60%|█████▉    | 6399/10682 [54:29<35:16,  2.02it/s]
 60%|█████▉    | 6400/10682 [54:29<35:15,  2.02it/s]
                                                    {'loss': 2.993, 'grad_norm': 0.2814801037311554, 'learning_rate': 0.00041471402041183287, 'epoch': 8.38}
+
 60%|█████▉    | 6400/10682 [54:29<35:15,  2.02it/s]
 60%|█████▉    | 6401/10682 [54:30<35:16,  2.02it/s]
 60%|█████▉    | 6402/10682 [54:30<35:17,  2.02it/s]
 60%|█████▉    | 6403/10682 [54:31<35:16,  2.02it/s]
 60%|█████▉    | 6404/10682 [54:31<35:14,  2.02it/s]
 60%|█████▉    | 6405/10682 [54:32<35:16,  2.02it/s]
 60%|█████▉    | 6406/10682 [54:32<35:14,  2.02it/s]
 60%|█████▉    | 6407/10682 [54:33<35:13,  2.02it/s]
 60%|█████▉    | 6408/10682 [54:33<35:11,  2.02it/s]
 60%|█████▉    | 6409/10682 [54:34<35:11,  2.02it/s]
 60%|██████    | 6410/10682 [54:34<35:10,  2.02it/s]
 60%|██████    | 6411/10682 [54:35<35:11,  2.02it/s]
 60%|██████    | 6412/10682 [54:35<35:07,  2.03it/s]
 60%|██████    | 6413/10682 [54:36<35:10,  2.02it/s]
 60%|██████    | 6414/10682 [54:36<35:09,  2.02it/s]
 60%|██████    | 6415/10682 [54:37<35:08,  2.02it/s]
 60%|██████    | 6416/10682 [54:37<35:08,  2.02it/s]
 60%|██████    | 6417/10682 [54:38<35:08,  2.02it/s]
 60%|██████    | 6418/10682 [54:38<35:07,  2.02it/s]
 60%|██████    | 6419/10682 [54:39<35:09,  2.02it/s]
 60%|██████    | 6420/10682 [54:39<35:05,  2.02it/s]
 60%|██████    | 6421/10682 [54:40<35:06,  2.02it/s]
 60%|██████    | 6422/10682 [54:40<35:04,  2.02it/s]
 60%|██████    | 6423/10682 [54:41<35:04,  2.02it/s]
 60%|██████    | 6424/10682 [54:41<35:04,  2.02it/s]
 60%|██████    | 6425/10682 [54:42<35:05,  2.02it/s]{'loss': 2.9891, 'grad_norm': 0.2660364508628845, 'learning_rate': 0.0004106916941082745, 'epoch': 8.42}
+
                                                    

 60%|██████    | 6425/10682 [54:42<35:05,  2.02it/s]
 60%|██████    | 6426/10682 [54:42<35:04,  2.02it/s]
 60%|██████    | 6427/10682 [54:43<35:05,  2.02it/s]
 60%|██████    | 6428/10682 [54:43<35:02,  2.02it/s]
 60%|██████    | 6429/10682 [54:44<35:04,  2.02it/s]
 60%|██████    | 6430/10682 [54:44<35:01,  2.02it/s]
 60%|██████    | 6431/10682 [54:45<35:01,  2.02it/s]
 60%|██████    | 6432/10682 [54:45<34:58,  2.03it/s]
 60%|██████    | 6433/10682 [54:46<34:57,  2.03it/s]
 60%|██████    | 6434/10682 [54:46<34:58,  2.02it/s]
 60%|██████    | 6435/10682 [54:47<34:56,  2.03it/s]
 60%|██████    | 6436/10682 [54:47<34:58,  2.02it/s]
 60%|██████    | 6437/10682 [54:48<34:54,  2.03it/s]
 60%|██████    | 6438/10682 [54:48<34:55,  2.03it/s]
 60%|██████    | 6439/10682 [54:49<34:56,  2.02it/s]
 60%|██████    | 6440/10682 [54:49<34:56,  2.02it/s]
 60%|██████    | 6441/10682 [54:50<34:55,  2.02it/s]
 60%|██████    | 6442/10682 [54:50<34:57,  2.02it/s]
 60%|██████    | 6443/10682 [54:51<34:55,  2.02it/s]
 60%|██████    | 6444/10682 [54:51<34:54,  2.02it/s]
 60%|██████    | 6445/10682 [54:52<34:51,  2.03it/s]
 60%|██████    | 6446/10682 [54:52<34:51,  2.03it/s]
 60%|██████    | 6447/10682 [54:53<34:51,  2.02it/s]
 60%|██████    | 6448/10682 [54:53<34:49,  2.03it/s]
 60%|██████    | 6449/10682 [54:54<34:50,  2.02it/s]
 60%|██████    | 6450/10682 [54:54<34:48,  2.03it/s]{'loss': 2.9995, 'grad_norm': 0.26787522435188293, 'learning_rate': 0.0004066753292465861, 'epoch': 8.45}
+
                                                    

 60%|██████    | 6450/10682 [54:54<34:48,  2.03it/s]
 60%|██████    | 6451/10682 [54:55<34:51,  2.02it/s]
 60%|██████    | 6452/10682 [54:55<34:47,  2.03it/s]
 60%|██████    | 6453/10682 [54:56<34:49,  2.02it/s]
 60%|██████    | 6454/10682 [54:56<34:45,  2.03it/s]
 60%|██████    | 6455/10682 [54:57<34:47,  2.03it/s]
 60%|██████    | 6456/10682 [54:57<34:45,  2.03it/s]
 60%|██████    | 6457/10682 [54:58<34:47,  2.02it/s]
 60%|██████    | 6458/10682 [54:58<34:43,  2.03it/s]
 60%|██████    | 6459/10682 [54:59<34:42,  2.03it/s]
 60%|██████    | 6460/10682 [54:59<34:42,  2.03it/s]
 60%|██████    | 6461/10682 [55:00<34:42,  2.03it/s]
 60%|██████    | 6462/10682 [55:00<34:43,  2.03it/s]
 61%|██████    | 6463/10682 [55:01<34:42,  2.03it/s]
 61%|██████    | 6464/10682 [55:01<34:45,  2.02it/s]
 61%|██████    | 6465/10682 [55:02<34:43,  2.02it/s]
 61%|██████    | 6466/10682 [55:02<34:44,  2.02it/s]
 61%|██████    | 6467/10682 [55:03<34:42,  2.02it/s]
 61%|██████    | 6468/10682 [55:03<34:42,  2.02it/s]
 61%|██████    | 6469/10682 [55:04<34:40,  2.02it/s]
 61%|██████    | 6470/10682 [55:04<34:41,  2.02it/s]
 61%|██████    | 6471/10682 [55:05<34:40,  2.02it/s]
 61%|██████    | 6472/10682 [55:05<34:37,  2.03it/s]
 61%|██████    | 6473/10682 [55:05<34:38,  2.02it/s]
 61%|██████    | 6474/10682 [55:06<34:37,  2.03it/s]
 61%|██████    | 6475/10682 [55:06<34:38,  2.02it/s]{'loss': 2.9967, 'grad_norm': 0.2651734948158264, 'learning_rate': 0.00040266519392417936, 'epoch': 8.48}
+
                                                    

 61%|██████    | 6475/10682 [55:06<34:38,  2.02it/s]
 61%|██████    | 6476/10682 [55:07<34:39,  2.02it/s]
 61%|██████    | 6477/10682 [55:07<34:39,  2.02it/s]
 61%|██████    | 6478/10682 [55:08<34:39,  2.02it/s]
 61%|██████    | 6479/10682 [55:08<34:39,  2.02it/s]
 61%|██████    | 6480/10682 [55:09<34:37,  2.02it/s]
 61%|██████    | 6481/10682 [55:09<34:39,  2.02it/s]
 61%|██████    | 6482/10682 [55:10<34:35,  2.02it/s]
 61%|██████    | 6483/10682 [55:10<34:35,  2.02it/s]
 61%|██████    | 6484/10682 [55:11<34:33,  2.02it/s]
 61%|██████    | 6485/10682 [55:11<34:34,  2.02it/s]
 61%|██████    | 6486/10682 [55:12<34:36,  2.02it/s]
 61%|██████    | 6487/10682 [55:12<34:35,  2.02it/s]
 61%|██████    | 6488/10682 [55:13<34:33,  2.02it/s]
 61%|██████    | 6489/10682 [55:13<34:33,  2.02it/s]
 61%|██████    | 6490/10682 [55:14<34:32,  2.02it/s]
 61%|██████    | 6491/10682 [55:14<34:32,  2.02it/s]
 61%|██████    | 6492/10682 [55:15<34:31,  2.02it/s]
 61%|██████    | 6493/10682 [55:15<34:29,  2.02it/s]
 61%|██████    | 6494/10682 [55:16<34:29,  2.02it/s]
 61%|██████    | 6495/10682 [55:16<34:27,  2.03it/s]
 61%|██████    | 6496/10682 [55:17<34:28,  2.02it/s]
 61%|██████    | 6497/10682 [55:17<34:26,  2.03it/s]
 61%|██████    | 6498/10682 [55:18<34:29,  2.02it/s]
 61%|██████    | 6499/10682 [55:18<34:26,  2.02it/s]
 61%|██████    | 6500/10682 [55:19<34:27,  2.02it/s]{'loss': 3.0018, 'grad_norm': 0.27361437678337097, 'learning_rate': 0.00039866155582263653, 'epoch': 8.51}
                                                    
+
 61%|██████    | 6500/10682 [55:19<34:27,  2.02it/s]
 61%|██████    | 6501/10682 [55:19<34:28,  2.02it/s]
 61%|██████    | 6502/10682 [55:20<34:27,  2.02it/s]
 61%|██████    | 6503/10682 [55:20<34:25,  2.02it/s]
 61%|██████    | 6504/10682 [55:21<34:26,  2.02it/s]
 61%|██████    | 6505/10682 [55:21<34:24,  2.02it/s]
 61%|██████    | 6506/10682 [55:22<34:25,  2.02it/s]
 61%|██████    | 6507/10682 [55:22<34:22,  2.02it/s]
 61%|██████    | 6508/10682 [55:23<34:23,  2.02it/s]
 61%|██████    | 6509/10682 [55:23<34:22,  2.02it/s]
 61%|██████    | 6510/10682 [55:24<34:23,  2.02it/s]
 61%|██████    | 6511/10682 [55:24<34:21,  2.02it/s]
 61%|██████    | 6512/10682 [55:25<34:21,  2.02it/s]
 61%|██████    | 6513/10682 [55:25<34:19,  2.02it/s]
 61%|██████    | 6514/10682 [55:26<34:20,  2.02it/s]
 61%|██████    | 6515/10682 [55:26<34:19,  2.02it/s]
 61%|██████    | 6516/10682 [55:27<34:17,  2.02it/s]
 61%|██████    | 6517/10682 [55:27<34:17,  2.02it/s]
 61%|██████    | 6518/10682 [55:28<34:15,  2.03it/s]
 61%|██████    | 6519/10682 [55:28<34:17,  2.02it/s]
 61%|██████    | 6520/10682 [55:29<34:14,  2.03it/s]
 61%|██████    | 6521/10682 [55:29<34:16,  2.02it/s]
 61%|██████    | 6522/10682 [55:30<34:14,  2.02it/s]
 61%|██████    | 6523/10682 [55:30<34:14,  2.02it/s]
 61%|██████    | 6524/10682 [55:31<34:12,  2.03it/s]
 61%|██████    | 6525/10682 [55:31<34:11,  2.03it/s]{'loss': 3.0049, 'grad_norm': 0.2760602831840515, 'learning_rate': 0.0003946646821898418, 'epoch': 8.55}
                                                    
+
 61%|██████    | 6525/10682 [55:31<34:11,  2.03it/s]
 61%|██████    | 6526/10682 [55:32<34:15,  2.02it/s]
 61%|██████    | 6527/10682 [55:32<34:14,  2.02it/s]
 61%|██████    | 6528/10682 [55:33<34:14,  2.02it/s]
 61%|██████    | 6529/10682 [55:33<34:12,  2.02it/s]
 61%|██████    | 6530/10682 [55:34<34:10,  2.02it/s]
 61%|██████    | 6531/10682 [55:34<34:10,  2.02it/s]
 61%|██████    | 6532/10682 [55:35<34:11,  2.02it/s]
 61%|██████    | 6533/10682 [55:35<34:12,  2.02it/s]
 61%|██████    | 6534/10682 [55:36<34:09,  2.02it/s]
 61%|██████    | 6535/10682 [55:36<34:10,  2.02it/s]
 61%|██████    | 6536/10682 [55:37<34:09,  2.02it/s]
 61%|██████    | 6537/10682 [55:37<34:08,  2.02it/s]
 61%|██████    | 6538/10682 [55:38<34:08,  2.02it/s]
 61%|██████    | 6539/10682 [55:38<34:09,  2.02it/s]
 61%|██████    | 6540/10682 [55:39<34:09,  2.02it/s]
 61%|██████    | 6541/10682 [55:39<34:07,  2.02it/s]
 61%|██████    | 6542/10682 [55:40<34:08,  2.02it/s]
 61%|██████▏   | 6543/10682 [55:40<34:05,  2.02it/s]
 61%|██████▏   | 6544/10682 [55:41<34:04,  2.02it/s]
 61%|██████▏   | 6545/10682 [55:41<34:05,  2.02it/s]
 61%|██████▏   | 6546/10682 [55:42<34:04,  2.02it/s]
 61%|██████▏   | 6547/10682 [55:42<34:07,  2.02it/s]
 61%|██████▏   | 6548/10682 [55:43<34:05,  2.02it/s]
 61%|██████▏   | 6549/10682 [55:43<34:06,  2.02it/s]
 61%|██████▏   | 6550/10682 [55:44<34:03,  2.02it/s]
                                                    
{'loss': 2.9936, 'grad_norm': 0.2589092552661896, 'learning_rate': 0.0003906748398221429, 'epoch': 8.58}
+
 61%|██████▏   | 6550/10682 [55:44<34:03,  2.02it/s]
 61%|██████▏   | 6551/10682 [55:44<34:08,  2.02it/s]
 61%|██████▏   | 6552/10682 [55:45<34:04,  2.02it/s]
 61%|██████▏   | 6553/10682 [55:45<34:05,  2.02it/s]
 61%|██████▏   | 6554/10682 [55:46<34:01,  2.02it/s]
 61%|██████▏   | 6555/10682 [55:46<34:03,  2.02it/s]
 61%|██████▏   | 6556/10682 [55:47<34:00,  2.02it/s]
 61%|██████▏   | 6557/10682 [55:47<34:01,  2.02it/s]
 61%|██████▏   | 6558/10682 [55:48<33:59,  2.02it/s]
 61%|██████▏   | 6559/10682 [55:48<33:59,  2.02it/s]
 61%|██████▏   | 6560/10682 [55:49<33:59,  2.02it/s]
 61%|██████▏   | 6561/10682 [55:49<34:00,  2.02it/s]
 61%|██████▏   | 6562/10682 [55:50<33:59,  2.02it/s]
 61%|██████▏   | 6563/10682 [55:50<33:58,  2.02it/s]
 61%|██████▏   | 6564/10682 [55:50<33:55,  2.02it/s]
 61%|██████▏   | 6565/10682 [55:51<33:55,  2.02it/s]
 61%|██████▏   | 6566/10682 [55:51<33:54,  2.02it/s]
 61%|██████▏   | 6567/10682 [55:52<33:55,  2.02it/s]
 61%|██████▏   | 6568/10682 [55:52<33:54,  2.02it/s]
 61%|██████▏   | 6569/10682 [55:53<33:53,  2.02it/s]
 62%|██████▏   | 6570/10682 [55:53<33:52,  2.02it/s]
 62%|██████▏   | 6571/10682 [55:54<33:53,  2.02it/s]
 62%|██████▏   | 6572/10682 [55:54<33:53,  2.02it/s]
 62%|██████▏   | 6573/10682 [55:55<33:51,  2.02it/s]
 62%|██████▏   | 6574/10682 [55:55<33:50,  2.02it/s]
 62%|██████▏   | 6575/10682 [55:56<33:51,  2.02it/s]{'loss': 3.0132, 'grad_norm': 0.26368701457977295, 'learning_rate': 0.00038669229504654146, 'epoch': 8.61}
+
                                                    

 62%|██████▏   | 6575/10682 [55:56<33:51,  2.02it/s]
 62%|██████▏   | 6576/10682 [55:56<33:52,  2.02it/s]
 62%|██████▏   | 6577/10682 [55:57<33:51,  2.02it/s]
 62%|██████▏   | 6578/10682 [55:57<33:52,  2.02it/s]
 62%|██████▏   | 6579/10682 [55:58<33:53,  2.02it/s]
 62%|██████▏   | 6580/10682 [55:58<33:53,  2.02it/s]
 62%|██████▏   | 6581/10682 [55:59<33:51,  2.02it/s]
 62%|██████▏   | 6582/10682 [55:59<33:50,  2.02it/s]
 62%|██████▏   | 6583/10682 [56:00<33:50,  2.02it/s]
 62%|██████▏   | 6584/10682 [56:00<33:49,  2.02it/s]
 62%|██████▏   | 6585/10682 [56:01<33:47,  2.02it/s]
 62%|██████▏   | 6586/10682 [56:01<33:48,  2.02it/s]
 62%|██████▏   | 6587/10682 [56:02<33:47,  2.02it/s]
 62%|██████▏   | 6588/10682 [56:02<33:47,  2.02it/s]
 62%|██████▏   | 6589/10682 [56:03<33:45,  2.02it/s]
 62%|██████▏   | 6590/10682 [56:03<33:45,  2.02it/s]
 62%|██████▏   | 6591/10682 [56:04<33:45,  2.02it/s]
 62%|██████▏   | 6592/10682 [56:04<33:44,  2.02it/s]
 62%|██████▏   | 6593/10682 [56:05<33:43,  2.02it/s]
 62%|██████▏   | 6594/10682 [56:05<33:42,  2.02it/s]
 62%|██████▏   | 6595/10682 [56:06<33:42,  2.02it/s]
 62%|���█████▏   | 6596/10682 [56:06<33:42,  2.02it/s]
 62%|██████▏   | 6597/10682 [56:07<33:42,  2.02it/s]
 62%|██████▏   | 6598/10682 [56:07<33:41,  2.02it/s]
 62%|██████▏   | 6599/10682 [56:08<33:42,  2.02it/s]
 62%|██████▏   | 6600/10682 [56:08<33:40,  2.02it/s]
                                                    {'loss': 3.0135, 'grad_norm': 0.2620540261268616, 'learning_rate': 0.00038271731370291583, 'epoch': 8.64}
+

 62%|██████▏   | 6600/10682 [56:08<33:40,  2.02it/s]
 62%|██████▏   | 6601/10682 [56:09<33:44,  2.02it/s]
 62%|██████▏   | 6602/10682 [56:09<33:43,  2.02it/s]
 62%|██████▏   | 6603/10682 [56:10<33:41,  2.02it/s]
 62%|██████▏   | 6604/10682 [56:10<33:40,  2.02it/s]
 62%|██████▏   | 6605/10682 [56:11<33:39,  2.02it/s]
 62%|██████▏   | 6606/10682 [56:11<33:39,  2.02it/s]
 62%|██████▏   | 6607/10682 [56:12<33:40,  2.02it/s]
 62%|██████▏   | 6608/10682 [56:12<33:39,  2.02it/s]
 62%|██████▏   | 6609/10682 [56:13<33:37,  2.02it/s]
 62%|██████▏   | 6610/10682 [56:13<33:38,  2.02it/s]
 62%|██████▏   | 6611/10682 [56:14<33:35,  2.02it/s]
 62%|██████▏   | 6612/10682 [56:14<33:35,  2.02it/s]
 62%|██████▏   | 6613/10682 [56:15<33:35,  2.02it/s]
 62%|██████▏   | 6614/10682 [56:15<33:37,  2.02it/s]
 62%|██████▏   | 6615/10682 [56:16<33:35,  2.02it/s]
 62%|██████▏   | 6616/10682 [56:16<33:36,  2.02it/s]
 62%|██████▏   | 6617/10682 [56:17<33:34,  2.02it/s]
 62%|██████▏   | 6618/10682 [56:17<33:34,  2.02it/s]
 62%|██████▏   | 6619/10682 [56:18<33:31,  2.02it/s]
 62%|██████▏   | 6620/10682 [56:18<33:31,  2.02it/s]
 62%|██████▏   | 6621/10682 [56:19<33:29,  2.02it/s]
 62%|██████▏   | 6622/10682 [56:19<33:31,  2.02it/s]
 62%|██████▏   | 6623/10682 [56:20<33:29,  2.02it/s]
 62%|██████▏   | 6624/10682 [56:20<33:28,  2.02it/s]
 62%|██████▏   | 6625/10682 [56:21<33:25,  2.02it/s]{'loss': 2.997, 'grad_norm': 0.27740538120269775, 'learning_rate': 0.0003787501611262762, 'epoch': 8.68}
+
                                                    

 62%|██████▏   | 6625/10682 [56:21<33:25,  2.02it/s]
 62%|██████▏   | 6626/10682 [56:21<33:29,  2.02it/s]
 62%|██████▏   | 6627/10682 [56:22<33:26,  2.02it/s]
 62%|██████▏   | 6628/10682 [56:22<33:26,  2.02it/s]
 62%|██████▏   | 6629/10682 [56:23<33:27,  2.02it/s]
 62%|██████▏   | 6630/10682 [56:23<33:26,  2.02it/s]
 62%|██████▏   | 6631/10682 [56:24<33:27,  2.02it/s]
 62%|██████▏   | 6632/10682 [56:24<33:27,  2.02it/s]
 62%|██████▏   | 6633/10682 [56:25<33:26,  2.02it/s]
 62%|██████▏   | 6634/10682 [56:25<33:25,  2.02it/s]
 62%|██████▏   | 6635/10682 [56:26<33:24,  2.02it/s]
 62%|██████▏   | 6636/10682 [56:26<33:23,  2.02it/s]
 62%|██████▏   | 6637/10682 [56:27<33:24,  2.02it/s]
 62%|██████▏   | 6638/10682 [56:27<33:24,  2.02it/s]
 62%|██████▏   | 6639/10682 [56:28<33:22,  2.02it/s]
 62%|██████▏   | 6640/10682 [56:28<33:20,  2.02it/s]
 62%|██████▏   | 6641/10682 [56:29<33:19,  2.02it/s]
 62%|██████▏   | 6642/10682 [56:29<33:20,  2.02it/s]
 62%|██████▏   | 6643/10682 [56:30<33:20,  2.02it/s]
 62%|██████▏   | 6644/10682 [56:30<33:18,  2.02it/s]
 62%|██████▏   | 6645/10682 [56:31<33:18,  2.02it/s]
 62%|██████▏   | 6646/10682 [56:31<33:18,  2.02it/s]
 62%|██████▏   | 6647/10682 [56:32<33:17,  2.02it/s]
 62%|██████▏   | 6648/10682 [56:32<33:18,  2.02it/s]
 62%|██████▏   | 6649/10682 [56:33<33:16,  2.02it/s]
 62%|██████▏   | 6650/10682 [56:33<33:17,  2.02it/s]{'loss': 3.0071, 'grad_norm': 0.26853495836257935, 'learning_rate': 0.0003747911021290519, 'epoch': 8.71}
+
                                                    

 62%|██████▏   | 6650/10682 [56:33<33:17,  2.02it/s]
 62%|██████▏   | 6651/10682 [56:34<33:16,  2.02it/s]
 62%|██████▏   | 6652/10682 [56:34<33:18,  2.02it/s]
 62%|██████▏   | 6653/10682 [56:35<33:16,  2.02it/s]
 62%|██████▏   | 6654/10682 [56:35<33:18,  2.02it/s]
 62%|██████▏   | 6655/10682 [56:36<33:15,  2.02it/s]
 62%|██████▏   | 6656/10682 [56:36<33:14,  2.02it/s]
 62%|██████▏   | 6657/10682 [56:37<33:12,  2.02it/s]
 62%|██████▏   | 6658/10682 [56:37<33:13,  2.02it/s]
 62%|██████▏   | 6659/10682 [56:38<33:10,  2.02it/s]
 62%|██████▏   | 6660/10682 [56:38<33:11,  2.02it/s]
 62%|██████▏   | 6661/10682 [56:39<33:10,  2.02it/s]
 62%|█████���▏   | 6662/10682 [56:39<33:12,  2.02it/s]
 62%|██████▏   | 6663/10682 [56:40<33:11,  2.02it/s]
 62%|██████▏   | 6664/10682 [56:40<33:09,  2.02it/s]
 62%|██████▏   | 6665/10682 [56:41<33:09,  2.02it/s]
 62%|██████▏   | 6666/10682 [56:41<33:08,  2.02it/s]
 62%|██████▏   | 6667/10682 [56:41<33:08,  2.02it/s]
 62%|██████▏   | 6668/10682 [56:42<33:11,  2.02it/s]
 62%|██████▏   | 6669/10682 [56:42<33:10,  2.02it/s]
 62%|██████▏   | 6670/10682 [56:43<33:08,  2.02it/s]
 62%|██████▏   | 6671/10682 [56:43<33:08,  2.02it/s]
 62%|██████▏   | 6672/10682 [56:44<33:06,  2.02it/s]
 62%|██████▏   | 6673/10682 [56:44<33:05,  2.02it/s]
 62%|██████▏   | 6674/10682 [56:45<33:04,  2.02it/s]
 62%|██████▏   | 6675/10682 [56:45<33:03,  2.02it/s]{'loss': 3.0143, 'grad_norm': 0.2528739869594574, 'learning_rate': 0.0003708404009834161, 'epoch': 8.74}
+
                                                    

 62%|██████▏   | 6675/10682 [56:45<33:03,  2.02it/s]
 62%|██████▏   | 6676/10682 [56:46<33:06,  2.02it/s]
 63%|██████▎   | 6677/10682 [56:46<33:04,  2.02it/s]
 63%|██████▎   | 6678/10682 [56:47<33:04,  2.02it/s]
 63%|██████▎   | 6679/10682 [56:47<33:02,  2.02it/s]
 63%|██████▎   | 6680/10682 [56:48<33:01,  2.02it/s]
 63%|██████▎   | 6681/10682 [56:48<33:00,  2.02it/s]
 63%|██████▎   | 6682/10682 [56:49<33:00,  2.02it/s]
 63%|██████▎   | 6683/10682 [56:49<32:59,  2.02it/s]
 63%|██████▎   | 6684/10682 [56:50<33:00,  2.02it/s]
 63%|██████▎   | 6685/10682 [56:50<33:01,  2.02it/s]
 63%|██████▎   | 6686/10682 [56:51<32:59,  2.02it/s]
 63%|██████▎   | 6687/10682 [56:51<32:59,  2.02it/s]
 63%|██████▎   | 6688/10682 [56:52<32:57,  2.02it/s]
 63%|██████▎   | 6689/10682 [56:52<32:56,  2.02it/s]
 63%|██████▎   | 6690/10682 [56:53<32:57,  2.02it/s]
 63%|██████▎   | 6691/10682 [56:53<32:57,  2.02it/s]
 63%|██████▎   | 6692/10682 [56:54<32:56,  2.02it/s]
 63%|██████▎   | 6693/10682 [56:54<32:55,  2.02it/s]
 63%|██████▎   | 6694/10682 [56:55<32:53,  2.02it/s]
 63%|██████▎   | 6695/10682 [56:55<32:52,  2.02it/s]
 63%|██████▎   | 6696/10682 [56:56<32:51,  2.02it/s]
 63%|██████▎   | 6697/10682 [56:56<32:51,  2.02it/s]
 63%|██████▎   | 6698/10682 [56:57<32:50,  2.02it/s]
 63%|██████▎   | 6699/10682 [56:57<32:52,  2.02it/s]
 63%|██████▎   | 6700/10682 [56:58<32:49,  2.02it/s]{'loss': 3.011, 'grad_norm': 0.26451829075813293, 'learning_rate': 0.0003668983214036445, 'epoch': 8.78}
+
                                                    

 63%|██████▎   | 6700/10682 [56:58<32:49,  2.02it/s]
 63%|██████▎   | 6701/10682 [56:58<32:52,  2.02it/s]
 63%|██████▎   | 6702/10682 [56:59<32:50,  2.02it/s]
 63%|██████▎   | 6703/10682 [56:59<32:49,  2.02it/s]
 63%|██████▎   | 6704/10682 [57:00<32:50,  2.02it/s]
 63%|██████▎   | 6705/10682 [57:00<32:48,  2.02it/s]
 63%|██████▎   | 6706/10682 [57:01<32:48,  2.02it/s]
 63%|██████▎   | 6707/10682 [57:01<32:46,  2.02it/s]
 63%|██████▎   | 6708/10682 [57:02<32:45,  2.02it/s]
 63%|██████▎   | 6709/10682 [57:02<32:45,  2.02it/s]
 63%|██████▎   | 6710/10682 [57:03<32:45,  2.02it/s]
 63%|██████▎   | 6711/10682 [57:03<32:45,  2.02it/s]
 63%|██████▎   | 6712/10682 [57:04<32:44,  2.02it/s]
 63%|██████▎   | 6713/10682 [57:04<32:44,  2.02it/s]
 63%|██████▎   | 6714/10682 [57:05<32:43,  2.02it/s]
 63%|██████▎   | 6715/10682 [57:05<32:42,  2.02it/s]
 63%|██████▎   | 6716/10682 [57:06<32:41,  2.02it/s]
 63%|██████▎   | 6717/10682 [57:06<32:40,  2.02it/s]
 63%|██████▎   | 6718/10682 [57:07<32:40,  2.02it/s]
 63%|██████▎   | 6719/10682 [57:07<32:40,  2.02it/s]
 63%|██████▎   | 6720/10682 [57:08<32:41,  2.02it/s]
 63%|██████▎   | 6721/10682 [57:08<32:39,  2.02it/s]
 63%|██████▎   | 6722/10682 [57:09<32:39,  2.02it/s]
 63%|██████▎   | 6723/10682 [57:09<32:38,  2.02it/s]
 63%|██████▎   | 6724/10682 [57:10<32:36,  2.02it/s]
 63%|██████▎   | 6725/10682 [57:10<32:36,  2.02it/s]
                                                    
{'loss': 3.0054, 'grad_norm': 0.2683119773864746, 'learning_rate': 0.00036296512652851323, 'epoch': 8.81}
+
 63%|██████▎   | 6725/10682 [57:10<32:36,  2.02it/s]
 63%|██████▎   | 6726/10682 [57:11<32:38,  2.02it/s]
 63%|██████▎   | 6727/10682 [57:11<32:36,  2.02it/s]
 63%|██████▎   | 6728/10682 [57:12<32:37,  2.02it/s]
 63%|██████▎   | 6729/10682 [57:12<32:36,  2.02it/s]
 63%|██████▎   | 6730/10682 [57:13<32:34,  2.02it/s]
 63%|██████▎   | 6731/10682 [57:13<32:33,  2.02it/s]
 63%|██████▎   | 6732/10682 [57:14<32:34,  2.02it/s]
 63%|██████▎   | 6733/10682 [57:14<32:32,  2.02it/s]
 63%|██████▎   | 6734/10682 [57:15<32:32,  2.02it/s]
 63%|██████▎   | 6735/10682 [57:15<32:30,  2.02it/s]
 63%|██████▎   | 6736/10682 [57:16<32:30,  2.02it/s]
 63%|██████▎   | 6737/10682 [57:16<32:29,  2.02it/s]
 63%|██████▎   | 6738/10682 [57:17<32:28,  2.02it/s]
 63%|██████▎   | 6739/10682 [57:17<32:27,  2.02it/s]
 63%|██████▎   | 6740/10682 [57:18<32:29,  2.02it/s]
 63%|██████▎   | 6741/10682 [57:18<32:28,  2.02it/s]
 63%|██████▎   | 6742/10682 [57:19<32:28,  2.02it/s]
 63%|██████▎   | 6743/10682 [57:19<32:27,  2.02it/s]
 63%|██████▎   | 6744/10682 [57:20<32:26,  2.02it/s]
 63%|██████▎   | 6745/10682 [57:20<32:27,  2.02it/s]
 63%|██████▎   | 6746/10682 [57:21<32:26,  2.02it/s]
 63%|██████▎   | 6747/10682 [57:21<32:26,  2.02it/s]
 63%|██████▎   | 6748/10682 [57:22<32:26,  2.02it/s]
 63%|██████▎   | 6749/10682 [57:22<32:25,  2.02it/s]
 63%|██████▎   | 6750/10682 [57:23<32:24,  2.02it/s]
                                                    
{'loss': 2.9963, 'grad_norm': 0.2591618299484253, 'learning_rate': 0.00035904107890373196, 'epoch': 8.84}
+
 63%|██████▎   | 6750/10682 [57:23<32:24,  2.02it/s]
 63%|██████▎   | 6751/10682 [57:23<32:26,  2.02it/s]
 63%|██████▎   | 6752/10682 [57:24<32:25,  2.02it/s]
 63%|██████▎   | 6753/10682 [57:24<32:25,  2.02it/s]
 63%|██████▎   | 6754/10682 [57:25<32:24,  2.02it/s]
 63%|██████▎   | 6755/10682 [57:25<32:25,  2.02it/s]
 63%|██████▎   | 6756/10682 [57:26<32:22,  2.02it/s]
 63%|██████▎   | 6757/10682 [57:26<32:22,  2.02it/s]
 63%|██████▎   | 6758/10682 [57:27<32:21,  2.02it/s]
 63%|██████▎   | 6759/10682 [57:27<32:21,  2.02it/s]
 63%|██████▎   | 6760/10682 [57:28<32:21,  2.02it/s]
 63%|██████▎   | 6761/10682 [57:28<32:18,  2.02it/s]
 63%|██████▎   | 6762/10682 [57:29<32:17,  2.02it/s]
 63%|██████▎   | 6763/10682 [57:29<32:17,  2.02it/s]
 63%|██████▎   | 6764/10682 [57:29<32:17,  2.02it/s]
 63%|██████▎   | 6765/10682 [57:30<32:17,  2.02it/s]
 63%|██████▎   | 6766/10682 [57:30<32:15,  2.02it/s]
 63%|██████▎   | 6767/10682 [57:31<32:15,  2.02it/s]
 63%|██████▎   | 6768/10682 [57:31<32:14,  2.02it/s]
 63%|██████▎   | 6769/10682 [57:32<32:12,  2.02it/s]
 63%|██████▎   | 6770/10682 [57:32<32:11,  2.03it/s]
 63%|██████▎   | 6771/10682 [57:33<32:11,  2.02it/s]
 63%|██████▎   | 6772/10682 [57:33<32:10,  2.03it/s]
 63%|██████▎   | 6773/10682 [57:34<32:10,  2.03it/s]
 63%|██████▎   | 6774/10682 [57:34<32:10,  2.02it/s]
 63%|██████▎   | 6775/10682 [57:35<32:10,  2.02it/s]{'loss': 3.0146, 'grad_norm': 0.2743259370326996, 'learning_rate': 0.00035512644046442116, 'epoch': 8.87}
+
                                                    

 63%|██████▎   | 6775/10682 [57:35<32:10,  2.02it/s]
 63%|██████▎   | 6776/10682 [57:35<32:11,  2.02it/s]
 63%|██████▎   | 6777/10682 [57:36<32:10,  2.02it/s]
 63%|██████▎   | 6778/10682 [57:36<32:09,  2.02it/s]
 63%|██████▎   | 6779/10682 [57:37<32:07,  2.02it/s]
 63%|██████▎   | 6780/10682 [57:37<32:07,  2.02it/s]
 63%|██████▎   | 6781/10682 [57:38<32:04,  2.03it/s]
 63%|██████▎   | 6782/10682 [57:38<32:03,  2.03it/s]
 63%|██████▎   | 6783/10682 [57:39<32:03,  2.03it/s]
 64%|██████▎   | 6784/10682 [57:39<32:06,  2.02it/s]
 64%|██████▎   | 6785/10682 [57:40<32:03,  2.03it/s]
 64%|██████▎   | 6786/10682 [57:40<32:05,  2.02it/s]
 64%|██████▎   | 6787/10682 [57:41<32:02,  2.03it/s]
 64%|██████▎   | 6788/10682 [57:41<32:01,  2.03it/s]
 64%|██████▎   | 6789/10682 [57:42<32:03,  2.02it/s]
 64%|██████▎   | 6790/10682 [57:42<32:02,  2.02it/s]
 64%|██████▎   | 6791/10682 [57:43<32:03,  2.02it/s]
 64%|██████▎   | 6792/10682 [57:43<32:01,  2.02it/s]
 64%|██████▎   | 6793/10682 [57:44<32:00,  2.02it/s]
 64%|██████▎   | 6794/10682 [57:44<32:00,  2.02it/s]
 64%|██████▎   | 6795/10682 [57:45<31:59,  2.02it/s]
 64%|██████▎   | 6796/10682 [57:45<31:59,  2.02it/s]
 64%|██████▎   | 6797/10682 [57:46<31:58,  2.02it/s]
 64%|██████▎   | 6798/10682 [57:46<31:58,  2.02it/s]
 64%|██████▎   | 6799/10682 [57:47<31:58,  2.02it/s]
 64%|██████▎   | 6800/10682 [57:47<31:57,  2.02it/s]{'loss': 3.013, 'grad_norm': 0.2601087689399719, 'learning_rate': 0.00035122147251762554, 'epoch': 8.91}
                                                    
+
 64%|██████▎   | 6800/10682 [57:47<31:57,  2.02it/s]
 64%|██████▎   | 6801/10682 [57:48<32:00,  2.02it/s]
 64%|██████▎   | 6802/10682 [57:48<31:58,  2.02it/s]
 64%|██████▎   | 6803/10682 [57:49<31:58,  2.02it/s]
 64%|██████▎   | 6804/10682 [57:49<31:57,  2.02it/s]
 64%|██████▎   | 6805/10682 [57:50<31:57,  2.02it/s]
 64%|██████▎   | 6806/10682 [57:50<31:54,  2.02it/s]
 64%|██████▎   | 6807/10682 [57:51<31:55,  2.02it/s]
 64%|██████▎   | 6808/10682 [57:51<31:54,  2.02it/s]
 64%|██████▎   | 6809/10682 [57:52<31:51,  2.03it/s]
 64%|██████▍   | 6810/10682 [57:52<31:53,  2.02it/s]
 64%|██████▍   | 6811/10682 [57:53<31:52,  2.02it/s]
 64%|██████▍   | 6812/10682 [57:53<31:53,  2.02it/s]
 64%|██████▍   | 6813/10682 [57:54<31:51,  2.02it/s]
 64%|██████▍   | 6814/10682 [57:54<31:52,  2.02it/s]
 64%|██████▍   | 6815/10682 [57:55<31:50,  2.02it/s]
 64%|██████▍   | 6816/10682 [57:55<31:50,  2.02it/s]
 64%|██████▍   | 6817/10682 [57:56<31:49,  2.02it/s]
 64%|██████▍   | 6818/10682 [57:56<31:48,  2.02it/s]
 64%|██████▍   | 6819/10682 [57:57<31:50,  2.02it/s]
 64%|██████▍   | 6820/10682 [57:57<31:50,  2.02it/s]
 64%|██████▍   | 6821/10682 [57:58<31:48,  2.02it/s]
 64%|██████▍   | 6822/10682 [57:58<31:47,  2.02it/s]
 64%|██████▍   | 6823/10682 [57:59<31:44,  2.03it/s]
 64%|██████▍   | 6824/10682 [57:59<31:46,  2.02it/s]
 64%|██████▍   | 6825/10682 [58:00<31:44,  2.02it/s]
                                                    
{'loss': 3.0206, 'grad_norm': 0.2635193169116974, 'learning_rate': 0.00034732643572487243, 'epoch': 8.94}
+
 64%|██████▍   | 6825/10682 [58:00<31:44,  2.02it/s]
 64%|██████▍   | 6826/10682 [58:00<31:48,  2.02it/s]
 64%|██████▍   | 6827/10682 [58:01<31:45,  2.02it/s]
 64%|██████▍   | 6828/10682 [58:01<31:46,  2.02it/s]
 64%|██████▍   | 6829/10682 [58:02<31:43,  2.02it/s]
 64%|██████▍   | 6830/10682 [58:02<31:46,  2.02it/s]
 64%|██████▍   | 6831/10682 [58:03<31:44,  2.02it/s]
 64%|██████▍   | 6832/10682 [58:03<31:43,  2.02it/s]
 64%|██████▍   | 6833/10682 [58:04<31:41,  2.02it/s]
 64%|██████▍   | 6834/10682 [58:04<31:41,  2.02it/s]
 64%|██████▍   | 6835/10682 [58:05<31:39,  2.02it/s]
 64%|██████▍   | 6836/10682 [58:05<31:39,  2.02it/s]
 64%|██████▍   | 6837/10682 [58:06<31:39,  2.02it/s]
 64%|██████▍   | 6838/10682 [58:06<31:39,  2.02it/s]
 64%|██████▍   | 6839/10682 [58:07<31:37,  2.03it/s]
 64%|██████▍   | 6840/10682 [58:07<31:35,  2.03it/s]
 64%|██████▍   | 6841/10682 [58:08<31:37,  2.02it/s]
 64%|██████▍   | 6842/10682 [58:08<31:36,  2.03it/s]
 64%|██████▍   | 6843/10682 [58:09<31:36,  2.02it/s]
 64%|██████▍   | 6844/10682 [58:09<31:34,  2.03it/s]
 64%|██████▍   | 6845/10682 [58:10<31:34,  2.02it/s]
 64%|██████▍   | 6846/10682 [58:10<31:34,  2.03it/s]
 64%|██████▍   | 6847/10682 [58:11<31:35,  2.02it/s]
 64%|██████▍   | 6848/10682 [58:11<31:34,  2.02it/s]
 64%|██████▍   | 6849/10682 [58:11<31:35,  2.02it/s]
 64%|██████▍   | 6850/10682 [58:12<31:37,  2.02it/s]
                                                    {'loss': 3.0199, 'grad_norm': 0.2645472586154938, 'learning_rate': 0.000343441590084772, 'epoch': 8.97}
+
 64%|██████▍   | 6850/10682 [58:12<31:37,  2.02it/s]
 64%|██████▍   | 6851/10682 [58:12<31:40,  2.02it/s]
 64%|██████▍   | 6852/10682 [58:13<31:36,  2.02it/s]
 64%|██████▍   | 6853/10682 [58:13<31:35,  2.02it/s]
 64%|██████▍   | 6854/10682 [58:14<31:32,  2.02it/s]
 64%|██████▍   | 6855/10682 [58:14<31:32,  2.02it/s]
 64%|██████▍   | 6856/10682 [58:15<31:31,  2.02it/s]
 64%|██████▍   | 6857/10682 [58:15<31:31,  2.02it/s]
 64%|██████▍   | 6858/10682 [58:16<31:28,  2.02it/s]
 64%|██████▍   | 6859/10682 [58:16<31:28,  2.02it/s]
 64%|██████▍   | 6860/10682 [58:17<31:27,  2.02it/s]
 64%|██████▍   | 6861/10682 [58:17<31:28,  2.02it/s]
 64%|██████▍   | 6862/10682 [58:18<31:26,  2.02it/s]
 64%|██████▍   | 6863/10682 [58:18<31:28,  2.02it/s]
 64%|██████▍   | 6864/10682 [58:19<31:28,  2.02it/s]
 64%|██████▍   | 6865/10682 [58:19<31:28,  2.02it/s]
 64%|██████▍   | 6866/10682 [58:20<31:27,  2.02it/s]
 64%|██████▍   | 6867/10682 [58:20<31:26,  2.02it/s]
 64%|██████▍   | 6868/10682 [58:21<31:25,  2.02it/s]
 64%|██████▍   | 6869/10682 [58:21<31:24,  2.02it/s]
 64%|██████▍   | 6870/10682 [58:22<31:23,  2.02it/s]
 64%|██████▍   | 6871/10682 [58:22<32:07,  1.98it/s]
 64%|██████▍   | 6872/10682 [58:35<4:15:57,  4.03s/it]
 64%|██████▍   | 6873/10682 [58:35<3:08:32,  2.97s/it]
 64%|██████▍   | 6874/10682 [58:36<2:21:20,  2.23s/it]
 64%|██████▍   | 6875/10682 [58:36<1:48:19,  1.71s/it]
                                                      
{'loss': 2.9851, 'grad_norm': 0.2988123893737793, 'learning_rate': 0.0003395671949156626, 'epoch': 9.0}
+
 64%|██████▍   | 6875/10682 [58:36<1:48:19,  1.71s/it]
 64%|██████▍   | 6876/10682 [58:37<1:25:12,  1.34s/it]
 64%|██████▍   | 6877/10682 [58:37<1:09:03,  1.09s/it]
 64%|██████▍   | 6878/10682 [58:38<57:41,  1.10it/s]  
 64%|██████▍   | 6879/10682 [58:38<49:49,  1.27it/s]
 64%|██████▍   | 6880/10682 [58:39<44:22,  1.43it/s]
 64%|██████▍   | 6881/10682 [58:39<40:27,  1.57it/s]
 64%|██████▍   | 6882/10682 [58:40<37:42,  1.68it/s]
 64%|██████▍   | 6883/10682 [58:40<35:45,  1.77it/s]
 64%|██████▍   | 6884/10682 [58:41<34:26,  1.84it/s]
 64%|██████▍   | 6885/10682 [58:41<33:30,  1.89it/s]
 64%|██████▍   | 6886/10682 [58:42<32:50,  1.93it/s]
 64%|██████▍   | 6887/10682 [58:42<32:20,  1.96it/s]
 64%|██████▍   | 6888/10682 [58:43<31:59,  1.98it/s]
 64%|██████▍   | 6889/10682 [58:43<31:47,  1.99it/s]
 65%|██████▍   | 6890/10682 [58:44<31:36,  2.00it/s]
 65%|██████▍   | 6891/10682 [58:44<31:28,  2.01it/s]
 65%|██████▍   | 6892/10682 [58:45<31:23,  2.01it/s]
 65%|██████▍   | 6893/10682 [58:45<31:19,  2.02it/s]
 65%|██████▍   | 6894/10682 [58:46<31:14,  2.02it/s]
 65%|██████▍   | 6895/10682 [58:46<31:14,  2.02it/s]
 65%|██████▍   | 6896/10682 [58:47<31:14,  2.02it/s]
 65%|██████▍   | 6897/10682 [58:47<31:13,  2.02it/s]
 65%|██████▍   | 6898/10682 [58:48<31:09,  2.02it/s]
 65%|██████▍   | 6899/10682 [58:48<31:14,  2.02it/s]
 65%|██████▍   | 6900/10682 [58:49<31:10,  2.02it/s]{'loss': 2.862, 'grad_norm': 0.2616669237613678, 'learning_rate': 0.0003357035088382997, 'epoch': 9.04}
+
                                                    

 65%|██████▍   | 6900/10682 [58:49<31:10,  2.02it/s]
 65%|██████▍   | 6901/10682 [58:49<31:10,  2.02it/s]
 65%|██████▍   | 6902/10682 [58:50<31:08,  2.02it/s]
 65%|██████▍   | 6903/10682 [58:50<31:08,  2.02it/s]
 65%|██████▍   | 6904/10682 [58:50<31:05,  2.03it/s]
 65%|██████▍   | 6905/10682 [58:51<31:03,  2.03it/s]
 65%|██████▍   | 6906/10682 [58:51<31:03,  2.03it/s]
 65%|██████▍   | 6907/10682 [58:52<31:01,  2.03it/s]
 65%|██████▍   | 6908/10682 [58:52<31:05,  2.02it/s]
 65%|██████▍   | 6909/10682 [58:53<31:03,  2.02it/s]
 65%|██████▍   | 6910/10682 [58:53<31:03,  2.02it/s]
 65%|██████▍   | 6911/10682 [58:54<31:02,  2.02it/s]
 65%|██████▍   | 6912/10682 [58:54<31:01,  2.03it/s]
 65%|██████▍   | 6913/10682 [58:55<31:02,  2.02it/s]
 65%|██████▍   | 6914/10682 [58:55<31:02,  2.02it/s]
 65%|██████▍   | 6915/10682 [58:56<31:02,  2.02it/s]
 65%|██████▍   | 6916/10682 [58:56<31:01,  2.02it/s]
 65%|██████▍   | 6917/10682 [58:57<31:00,  2.02it/s]
 65%|██████▍   | 6918/10682 [58:57<30:59,  2.02it/s]
 65%|██████▍   | 6919/10682 [58:58<30:57,  2.03it/s]
 65%|██████▍   | 6920/10682 [58:58<30:59,  2.02it/s]
 65%|██████▍   | 6921/10682 [58:59<30:56,  2.03it/s]
 65%|██████▍   | 6922/10682 [58:59<30:54,  2.03it/s]
 65%|██████▍   | 6923/10682 [59:00<30:56,  2.02it/s]
 65%|██████▍   | 6924/10682 [59:00<30:55,  2.03it/s]
 65%|██████▍   | 6925/10682 [59:01<30:54,  2.03it/s]{'loss': 2.8711, 'grad_norm': 0.26863160729408264, 'learning_rate': 0.0003318507897585944, 'epoch': 9.07}
+
                                                    

 65%|██████▍   | 6925/10682 [59:01<30:54,  2.03it/s]
 65%|██████▍   | 6926/10682 [59:02<33:41,  1.86it/s]
 65%|██████▍   | 6927/10682 [59:02<32:49,  1.91it/s]
 65%|██████▍   | 6928/10682 [59:02<32:13,  1.94it/s]
 65%|██████▍   | 6929/10682 [59:03<31:48,  1.97it/s]
 65%|██████▍   | 6930/10682 [59:03<31:31,  1.98it/s]
 65%|██████▍   | 6931/10682 [59:04<31:19,  2.00it/s]
 65%|██████▍   | 6932/10682 [59:04<31:12,  2.00it/s]
 65%|██████▍   | 6933/10682 [59:05<31:05,  2.01it/s]
 65%|██████▍   | 6934/10682 [59:05<31:01,  2.01it/s]
 65%|██████▍   | 6935/10682 [59:06<30:58,  2.02it/s]
 65%|██████▍   | 6936/10682 [59:06<30:53,  2.02it/s]
 65%|██████▍   | 6937/10682 [59:07<30:52,  2.02it/s]
 65%|██████▍   | 6938/10682 [59:07<30:50,  2.02it/s]
 65%|██████▍   | 6939/10682 [59:08<30:50,  2.02it/s]
 65%|██████▍   | 6940/10682 [59:08<30:48,  2.02it/s]
 65%|██████▍   | 6941/10682 [59:09<30:47,  2.02it/s]
 65%|██████▍   | 6942/10682 [59:09<30:47,  2.02it/s]
 65%|██████▍   | 6943/10682 [59:10<30:46,  2.02it/s]
 65%|██████▌   | 6944/10682 [59:10<30:48,  2.02it/s]
 65%|██████▌   | 6945/10682 [59:11<30:45,  2.02it/s]
 65%|██████▌   | 6946/10682 [59:11<30:44,  2.03it/s]
 65%|██████▌   | 6947/10682 [59:12<30:43,  2.03it/s]
 65%|██████▌   | 6948/10682 [59:12<30:44,  2.02it/s]
 65%|██████▌   | 6949/10682 [59:13<30:42,  2.03it/s]
 65%|██████▌   | 6950/10682 [59:13<30:43,  2.02it/s]
                                                    {'loss': 2.8819, 'grad_norm': 0.26469868421554565, 'learning_rate': 0.0003280092948503962, 'epoch': 9.1}
+
 65%|██████▌   | 6950/10682 [59:13<30:43,  2.02it/s]
 65%|██████▌   | 6951/10682 [59:14<30:44,  2.02it/s]
 65%|██████▌   | 6952/10682 [59:14<30:43,  2.02it/s]
 65%|██████▌   | 6953/10682 [59:15<30:41,  2.02it/s]
 65%|██████▌   | 6954/10682 [59:15<30:42,  2.02it/s]
 65%|██████▌   | 6955/10682 [59:16<30:43,  2.02it/s]
 65%|██████▌   | 6956/10682 [59:16<30:41,  2.02it/s]
 65%|██████▌   | 6957/10682 [59:17<30:39,  2.02it/s]
 65%|██████▌   | 6958/10682 [59:17<30:40,  2.02it/s]
 65%|██████▌   | 6959/10682 [59:18<30:38,  2.03it/s]
 65%|██████▌   | 6960/10682 [59:18<30:37,  2.03it/s]
 65%|██████▌   | 6961/10682 [59:19<30:36,  2.03it/s]
 65%|██████▌   | 6962/10682 [59:19<30:36,  2.03it/s]
 65%|██████▌   | 6963/10682 [59:20<30:36,  2.03it/s]
 65%|██████▌   | 6964/10682 [59:20<30:35,  2.03it/s]
 65%|██████▌   | 6965/10682 [59:21<30:35,  2.03it/s]
 65%|██████▌   | 6966/10682 [59:21<30:33,  2.03it/s]
 65%|██████▌   | 6967/10682 [59:22<30:35,  2.02it/s]
 65%|██████▌   | 6968/10682 [59:22<30:34,  2.02it/s]
 65%|██████▌   | 6969/10682 [59:23<30:34,  2.02it/s]
 65%|██████▌   | 6970/10682 [59:23<30:32,  2.03it/s]
 65%|██████▌   | 6971/10682 [59:24<30:32,  2.02it/s]
 65%|██████▌   | 6972/10682 [59:24<30:32,  2.02it/s]
 65%|██████▌   | 6973/10682 [59:25<30:32,  2.02it/s]
 65%|██████▌   | 6974/10682 [59:25<30:31,  2.02it/s]
 65%|██████▌   | 6975/10682 [59:26<30:33,  2.02it/s]
                                                    {'loss': 2.883, 'grad_norm': 0.2707022428512573, 'learning_rate': 0.00032417928053832723, 'epoch': 9.14}
+

 65%|██████▌   | 6975/10682 [59:26<30:33,  2.02it/s]
 65%|██████▌   | 6976/10682 [59:26<30:33,  2.02it/s]
 65%|██████▌   | 6977/10682 [59:27<30:32,  2.02it/s]
 65%|██████▌   | 6978/10682 [59:27<30:31,  2.02it/s]
 65%|██████▌   | 6979/10682 [59:28<30:32,  2.02it/s]
 65%|██████▌   | 6980/10682 [59:28<30:29,  2.02it/s]
 65%|██████▌   | 6981/10682 [59:29<30:28,  2.02it/s]
 65%|██████▌   | 6982/10682 [59:29<30:27,  2.02it/s]
 65%|██████▌   | 6983/10682 [59:30<30:25,  2.03it/s]
 65%|██████▌   | 6984/10682 [59:30<30:25,  2.03it/s]
 65%|██████▌   | 6985/10682 [59:31<30:24,  2.03it/s]
 65%|██████▌   | 6986/10682 [59:31<30:24,  2.03it/s]
 65%|██████▌   | 6987/10682 [59:32<30:23,  2.03it/s]
 65%|██████▌   | 6988/10682 [59:32<30:23,  2.03it/s]
 65%|██████▌   | 6989/10682 [59:33<30:22,  2.03it/s]
 65%|██████▌   | 6990/10682 [59:33<30:22,  2.03it/s]
 65%|██████▌   | 6991/10682 [59:34<30:24,  2.02it/s]
 65%|██████▌   | 6992/10682 [59:34<30:25,  2.02it/s]
 65%|██████▌   | 6993/10682 [59:35<30:24,  2.02it/s]
 65%|██████▌   | 6994/10682 [59:35<30:24,  2.02it/s]
 65%|██████▌   | 6995/10682 [59:36<30:22,  2.02it/s]
 65%|██████▌   | 6996/10682 [59:36<30:21,  2.02it/s]
 66%|██████▌   | 6997/10682 [59:37<30:20,  2.02it/s]
 66%|██████▌   | 6998/10682 [59:37<30:18,  2.03it/s]
 66%|██████▌   | 6999/10682 [59:38<30:20,  2.02it/s]
 66%|██████▌   | 7000/10682 [59:38<30:18,  2.02it/s]
                                                    {'loss': 2.879, 'grad_norm': 0.26636505126953125, 'learning_rate': 0.0003203610024806651, 'epoch': 9.17}
+
 66%|██████▌   | 7000/10682 [59:38<30:18,  2.02it/s]
 66%|██████▌   | 7001/10682 [59:39<30:20,  2.02it/s]
 66%|██████▌   | 7002/10682 [59:39<30:19,  2.02it/s]
 66%|██████▌   | 7003/10682 [59:40<30:18,  2.02it/s]
 66%|██████▌   | 7004/10682 [59:40<30:17,  2.02it/s]
 66%|██████▌   | 7005/10682 [59:41<30:15,  2.02it/s]
 66%|██████▌   | 7006/10682 [59:41<30:14,  2.03it/s]
 66%|██████▌   | 7007/10682 [59:42<30:14,  2.03it/s]
 66%|██████▌   | 7008/10682 [59:42<30:16,  2.02it/s]
 66%|██████▌   | 7009/10682 [59:43<30:17,  2.02it/s]
 66%|██████▌   | 7010/10682 [59:43<30:15,  2.02it/s]
 66%|██████▌   | 7011/10682 [59:44<32:51,  1.86it/s]
 66%|██████▌   | 7012/10682 [59:44<32:01,  1.91it/s]
 66%|██████▌   | 7013/10682 [59:45<31:29,  1.94it/s]
 66%|██████▌   | 7014/10682 [59:45<31:04,  1.97it/s]
 66%|██████▌   | 7015/10682 [59:46<30:48,  1.98it/s]
 66%|██████▌   | 7016/10682 [59:46<30:35,  2.00it/s]
 66%|██████▌   | 7017/10682 [59:47<30:25,  2.01it/s]
 66%|██████▌   | 7018/10682 [59:47<30:20,  2.01it/s]
 66%|██████▌   | 7019/10682 [59:48<30:16,  2.02it/s]
 66%|██████▌   | 7020/10682 [59:48<30:14,  2.02it/s]
 66%|██████▌   | 7021/10682 [59:49<30:10,  2.02it/s]
 66%|██████▌   | 7022/10682 [59:49<30:09,  2.02it/s]
 66%|██████▌   | 7023/10682 [59:50<30:07,  2.02it/s]
 66%|██████▌   | 7024/10682 [59:50<30:05,  2.03it/s]
 66%|██████▌   | 7025/10682 [59:51<30:05,  2.03it/s]
                                                    
{'loss': 2.889, 'grad_norm': 0.2684054970741272, 'learning_rate': 0.0003165547155522783, 'epoch': 9.2}
+
 66%|██████▌   | 7025/10682 [59:51<30:05,  2.03it/s]
 66%|██████▌   | 7026/10682 [59:51<30:08,  2.02it/s]
 66%|██████▌   | 7027/10682 [59:52<30:07,  2.02it/s]
 66%|██████▌   | 7028/10682 [59:52<30:06,  2.02it/s]
 66%|██████▌   | 7029/10682 [59:53<30:05,  2.02it/s]
 66%|██████▌   | 7030/10682 [59:53<30:03,  2.02it/s]
 66%|██████▌   | 7031/10682 [59:54<30:03,  2.02it/s]
 66%|██████▌   | 7032/10682 [59:54<30:00,  2.03it/s]
 66%|██████▌   | 7033/10682 [59:54<30:00,  2.03it/s]
 66%|██████▌   | 7034/10682 [59:55<30:00,  2.03it/s]
 66%|██████▌   | 7035/10682 [59:55<30:01,  2.02it/s]
 66%|██████▌   | 7036/10682 [59:56<29:59,  2.03it/s]
 66%|██████▌   | 7037/10682 [59:56<30:00,  2.02it/s]
 66%|██████▌   | 7038/10682 [59:57<29:59,  2.03it/s]
 66%|██████▌   | 7039/10682 [59:57<29:58,  2.03it/s]
 66%|██████▌   | 7040/10682 [59:58<29:59,  2.02it/s]
 66%|██████▌   | 7041/10682 [59:58<29:56,  2.03it/s]
 66%|██████▌   | 7042/10682 [59:59<29:58,  2.02it/s]
 66%|██████▌   | 7043/10682 [59:59<29:56,  2.03it/s]
 66%|██████▌   | 7044/10682 [1:00:00<29:57,  2.02it/s]
 66%|██████▌   | 7045/10682 [1:00:00<29:57,  2.02it/s]
 66%|██████▌   | 7046/10682 [1:00:01<29:55,  2.03it/s]
 66%|██████▌   | 7047/10682 [1:00:01<29:56,  2.02it/s]
 66%|██████▌   | 7048/10682 [1:00:02<29:54,  2.03it/s]
 66%|██████▌   | 7049/10682 [1:00:02<29:55,  2.02it/s]
 66%|██████▌   | 7050/10682 [1:00:03<29:53,  2.02it/s]
                                                      {'loss': 2.8916, 'grad_norm': 0.2807353138923645, 'learning_rate': 0.0003127606738276115, 'epoch': 9.23}
+
 66%|██████▌   | 7050/10682 [1:00:03<29:53,  2.02it/s]
 66%|██████▌   | 7051/10682 [1:00:03<29:56,  2.02it/s]
 66%|██████▌   | 7052/10682 [1:00:04<29:55,  2.02it/s]
 66%|██████▌   | 7053/10682 [1:00:04<29:53,  2.02it/s]
 66%|██████▌   | 7054/10682 [1:00:05<29:53,  2.02it/s]
 66%|██████▌   | 7055/10682 [1:00:05<29:52,  2.02it/s]
 66%|██████▌   | 7056/10682 [1:00:06<29:52,  2.02it/s]
 66%|██████▌   | 7057/10682 [1:00:06<29:52,  2.02it/s]
 66%|██████▌   | 7058/10682 [1:00:07<29:51,  2.02it/s]
 66%|██████▌   | 7059/10682 [1:00:07<29:49,  2.02it/s]
 66%|██████▌   | 7060/10682 [1:00:08<29:48,  2.03it/s]
 66%|██████▌   | 7061/10682 [1:00:08<29:52,  2.02it/s]
 66%|██████▌   | 7062/10682 [1:00:09<29:52,  2.02it/s]
 66%|██████▌   | 7063/10682 [1:00:09<29:50,  2.02it/s]
 66%|██████▌   | 7064/10682 [1:00:10<29:48,  2.02it/s]
 66%|██████▌   | 7065/10682 [1:00:10<29:46,  2.02it/s]
 66%|██████▌   | 7066/10682 [1:00:11<29:45,  2.02it/s]
 66%|██████▌   | 7067/10682 [1:00:11<29:46,  2.02it/s]
 66%|██████▌   | 7068/10682 [1:00:12<29:48,  2.02it/s]
 66%|██████▌   | 7069/10682 [1:00:12<29:46,  2.02it/s]
 66%|██████▌   | 7070/10682 [1:00:13<29:46,  2.02it/s]
 66%|██████▌   | 7071/10682 [1:00:13<29:45,  2.02it/s]
 66%|██████▌   | 7072/10682 [1:00:14<29:42,  2.02it/s]
 66%|██████▌   | 7073/10682 [1:00:14<29:45,  2.02it/s]
 66%|██████▌   | 7074/10682 [1:00:15<29:44,  2.02it/s]
 66%|██████▌   | 7075/10682 [1:00:15<29:43,  2.02it/s]
                                                      {'loss': 2.8971, 'grad_norm': 0.26620224118232727, 'learning_rate': 0.00030897913056372755, 'epoch': 9.27}
+

 66%|██████▌   | 7075/10682 [1:00:15<29:43,  2.02it/s]
 66%|██████▌   | 7076/10682 [1:00:16<29:43,  2.02it/s]
 66%|██████▋   | 7077/10682 [1:00:16<29:43,  2.02it/s]
 66%|██████▋   | 7078/10682 [1:00:17<29:41,  2.02it/s]
 66%|██████▋   | 7079/10682 [1:00:17<29:41,  2.02it/s]
 66%|██████▋   | 7080/10682 [1:00:18<29:39,  2.02it/s]
 66%|██████▋   | 7081/10682 [1:00:18<29:40,  2.02it/s]
 66%|██████▋   | 7082/10682 [1:00:19<29:39,  2.02it/s]
 66%|██████▋   | 7083/10682 [1:00:19<29:39,  2.02it/s]
 66%|██████▋   | 7084/10682 [1:00:20<29:39,  2.02it/s]
 66%|██████▋   | 7085/10682 [1:00:20<29:39,  2.02it/s]
 66%|██████▋   | 7086/10682 [1:00:21<29:37,  2.02it/s]
 66%|██████▋   | 7087/10682 [1:00:21<29:36,  2.02it/s]
 66%|██████▋   | 7088/10682 [1:00:22<29:36,  2.02it/s]
 66%|██████▋   | 7089/10682 [1:00:22<29:36,  2.02it/s]
 66%|██████▋   | 7090/10682 [1:00:23<29:35,  2.02it/s]
 66%|██████▋   | 7091/10682 [1:00:23<29:36,  2.02it/s]
 66%|██████▋   | 7092/10682 [1:00:24<29:35,  2.02it/s]
 66%|██████▋   | 7093/10682 [1:00:24<29:34,  2.02it/s]
 66%|██████▋   | 7094/10682 [1:00:25<29:34,  2.02it/s]
 66%|██████▋   | 7095/10682 [1:00:25<29:34,  2.02it/s]
 66%|██████▋   | 7096/10682 [1:00:26<29:33,  2.02it/s]
 66%|██████▋   | 7097/10682 [1:00:26<29:33,  2.02it/s]
 66%|██████▋   | 7098/10682 [1:00:27<29:32,  2.02it/s]
 66%|██████▋   | 7099/10682 [1:00:27<29:32,  2.02it/s]
 66%|██████▋   | 7100/10682 [1:00:28<29:30,  2.02it/s]
                                                      {'loss': 2.8974, 'grad_norm': 0.27408939599990845, 'learning_rate': 0.00030521033818340054, 'epoch': 9.3}
+

 66%|██████▋   | 7100/10682 [1:00:28<29:30,  2.02it/s]
 66%|██████▋   | 7101/10682 [1:00:28<29:30,  2.02it/s]
 66%|██████▋   | 7102/10682 [1:00:29<29:29,  2.02it/s]
 66%|██████▋   | 7103/10682 [1:00:29<29:30,  2.02it/s]
 67%|██████▋   | 7104/10682 [1:00:30<29:28,  2.02it/s]
 67%|██████▋   | 7105/10682 [1:00:30<29:27,  2.02it/s]
 67%|██████▋   | 7106/10682 [1:00:31<29:27,  2.02it/s]
 67%|██████▋   | 7107/10682 [1:00:31<29:27,  2.02it/s]
 67%|██████▋   | 7108/10682 [1:00:32<29:30,  2.02it/s]
 67%|██████▋   | 7109/10682 [1:00:32<29:27,  2.02it/s]
 67%|██████▋   | 7110/10682 [1:00:33<29:24,  2.02it/s]
 67%|██████▋   | 7111/10682 [1:00:33<29:24,  2.02it/s]
 67%|██████▋   | 7112/10682 [1:00:34<29:23,  2.02it/s]
 67%|██████▋   | 7113/10682 [1:00:34<29:23,  2.02it/s]
 67%|██████▋   | 7114/10682 [1:00:35<29:23,  2.02it/s]
 67%|██████▋   | 7115/10682 [1:00:35<29:23,  2.02it/s]
 67%|██████▋   | 7116/10682 [1:00:36<29:21,  2.02it/s]
 67%|██████▋   | 7117/10682 [1:00:36<29:23,  2.02it/s]
 67%|██████▋   | 7118/10682 [1:00:37<29:22,  2.02it/s]
 67%|██████▋   | 7119/10682 [1:00:37<29:23,  2.02it/s]
 67%|██████▋   | 7120/10682 [1:00:38<29:22,  2.02it/s]
 67%|██████▋   | 7121/10682 [1:00:38<29:21,  2.02it/s]
 67%|██████▋   | 7122/10682 [1:00:38<29:20,  2.02it/s]
 67%|██████▋   | 7123/10682 [1:00:39<29:20,  2.02it/s]
 67%|██████▋   | 7124/10682 [1:00:39<29:17,  2.02it/s]
 67%|██████▋   | 7125/10682 [1:00:40<29:18,  2.02it/s]{'loss': 2.9053, 'grad_norm': 0.27052444219589233, 'learning_rate': 0.0003014545482582672, 'epoch': 9.33}
                                                      
+
 67%|██████▋   | 7125/10682 [1:00:40<29:18,  2.02it/s]
 67%|██████▋   | 7126/10682 [1:00:40<29:21,  2.02it/s]
 67%|██████▋   | 7127/10682 [1:00:41<29:19,  2.02it/s]
 67%|██████▋   | 7128/10682 [1:00:41<29:18,  2.02it/s]
 67%|██████▋   | 7129/10682 [1:00:42<29:19,  2.02it/s]
 67%|██████▋   | 7130/10682 [1:00:42<29:16,  2.02it/s]
 67%|██████▋   | 7131/10682 [1:00:43<29:16,  2.02it/s]
 67%|██████▋   | 7132/10682 [1:00:43<29:14,  2.02it/s]
 67%|██████▋   | 7133/10682 [1:00:44<29:15,  2.02it/s]
 67%|██████▋   | 7134/10682 [1:00:44<29:13,  2.02it/s]
 67%|██████▋   | 7135/10682 [1:00:45<29:14,  2.02it/s]
 67%|██████▋   | 7136/10682 [1:00:45<29:12,  2.02it/s]
 67%|██████▋   | 7137/10682 [1:00:46<29:13,  2.02it/s]
 67%|██████▋   | 7138/10682 [1:00:46<29:11,  2.02it/s]
 67%|██████▋   | 7139/10682 [1:00:47<29:12,  2.02it/s]
 67%|██████▋   | 7140/10682 [1:00:47<29:11,  2.02it/s]
 67%|██████▋   | 7141/10682 [1:00:48<29:10,  2.02it/s]
 67%|██████▋   | 7142/10682 [1:00:48<29:09,  2.02it/s]
 67%|██████▋   | 7143/10682 [1:00:49<29:11,  2.02it/s]
 67%|██████▋   | 7144/10682 [1:00:49<29:08,  2.02it/s]
 67%|██████▋   | 7145/10682 [1:00:50<29:08,  2.02it/s]
 67%|██████▋   | 7146/10682 [1:00:50<29:07,  2.02it/s]
 67%|██████▋   | 7147/10682 [1:00:51<29:07,  2.02it/s]
 67%|██████▋   | 7148/10682 [1:00:51<29:06,  2.02it/s]
 67%|██████▋   | 7149/10682 [1:00:52<29:05,  2.02it/s]
 67%|██████▋   | 7150/10682 [1:00:52<29:06,  2.02it/s]
                                                      
{'loss': 2.9172, 'grad_norm': 0.2685629725456238, 'learning_rate': 0.0002977120114920343, 'epoch': 9.36}
+
 67%|██████▋   | 7150/10682 [1:00:52<29:06,  2.02it/s]
 67%|██████▋   | 7151/10682 [1:00:53<29:06,  2.02it/s]
 67%|██████▋   | 7152/10682 [1:00:53<29:05,  2.02it/s]
 67%|██████▋   | 7153/10682 [1:00:54<29:02,  2.03it/s]
 67%|██████▋   | 7154/10682 [1:00:54<29:04,  2.02it/s]
 67%|██████▋   | 7155/10682 [1:00:55<29:00,  2.03it/s]
 67%|██████▋   | 7156/10682 [1:00:55<29:02,  2.02it/s]
 67%|██████▋   | 7157/10682 [1:00:56<29:00,  2.02it/s]
 67%|██████▋   | 7158/10682 [1:00:56<29:00,  2.02it/s]
 67%|██████▋   | 7159/10682 [1:00:57<28:59,  2.03it/s]
 67%|██████▋   | 7160/10682 [1:00:57<29:01,  2.02it/s]
 67%|██████▋   | 7161/10682 [1:00:58<29:00,  2.02it/s]
 67%|██████▋   | 7162/10682 [1:00:58<29:00,  2.02it/s]
 67%|██████▋   | 7163/10682 [1:00:59<28:59,  2.02it/s]
 67%|██████▋   | 7164/10682 [1:00:59<29:00,  2.02it/s]
 67%|██████▋   | 7165/10682 [1:01:00<28:58,  2.02it/s]
 67%|██████▋   | 7166/10682 [1:01:00<28:57,  2.02it/s]
 67%|██████▋   | 7167/10682 [1:01:01<28:55,  2.03it/s]
 67%|██████▋   | 7168/10682 [1:01:01<28:53,  2.03it/s]
 67%|██████▋   | 7169/10682 [1:01:02<28:54,  2.02it/s]
 67%|██████▋   | 7170/10682 [1:01:02<28:52,  2.03it/s]
 67%|██████▋   | 7171/10682 [1:01:03<28:52,  2.03it/s]
 67%|██████▋   | 7172/10682 [1:01:03<28:51,  2.03it/s]
 67%|██████▋   | 7173/10682 [1:01:04<28:50,  2.03it/s]
 67%|██████▋   | 7174/10682 [1:01:04<28:50,  2.03it/s]
 67%|██████▋   | 7175/10682 [1:01:05<28:51,  2.03it/s]
                                                      {'loss': 2.9093, 'grad_norm': 0.27699965238571167, 'learning_rate': 0.00029398297770374336, 'epoch': 9.4}
+
 67%|██████▋   | 7175/10682 [1:01:05<28:51,  2.03it/s]
 67%|██████▋   | 7176/10682 [1:01:05<28:52,  2.02it/s]
 67%|██████▋   | 7177/10682 [1:01:06<28:51,  2.02it/s]
 67%|██████▋   | 7178/10682 [1:01:06<28:52,  2.02it/s]
 67%|██████▋   | 7179/10682 [1:01:07<28:50,  2.02it/s]
 67%|██████▋   | 7180/10682 [1:01:07<28:50,  2.02it/s]
 67%|██████▋   | 7181/10682 [1:01:08<28:48,  2.03it/s]
 67%|██████▋   | 7182/10682 [1:01:08<28:48,  2.03it/s]
 67%|██████▋   | 7183/10682 [1:01:09<28:48,  2.02it/s]
 67%|██████▋   | 7184/10682 [1:01:09<28:47,  2.02it/s]
 67%|██████▋   | 7185/10682 [1:01:10<28:48,  2.02it/s]
 67%|██████▋   | 7186/10682 [1:01:10<28:47,  2.02it/s]
 67%|██████▋   | 7187/10682 [1:01:11<28:45,  2.03it/s]
 67%|██████▋   | 7188/10682 [1:01:11<28:45,  2.02it/s]
 67%|██████▋   | 7189/10682 [1:01:12<28:44,  2.03it/s]
 67%|██████▋   | 7190/10682 [1:01:12<28:45,  2.02it/s]
 67%|██████▋   | 7191/10682 [1:01:13<28:45,  2.02it/s]
 67%|██████▋   | 7192/10682 [1:01:13<28:44,  2.02it/s]
 67%|██████▋   | 7193/10682 [1:01:14<28:44,  2.02it/s]
 67%|██████▋   | 7194/10682 [1:01:14<28:45,  2.02it/s]
 67%|██████▋   | 7195/10682 [1:01:15<28:43,  2.02it/s]
 67%|██████▋   | 7196/10682 [1:01:15<28:43,  2.02it/s]
 67%|███��██▋   | 7197/10682 [1:01:16<28:42,  2.02it/s]
 67%|██████▋   | 7198/10682 [1:01:16<28:42,  2.02it/s]
 67%|██████▋   | 7199/10682 [1:01:17<28:40,  2.02it/s]
 67%|██████▋   | 7200/10682 [1:01:17<28:39,  2.03it/s]{'loss': 2.9169, 'grad_norm': 0.26536059379577637, 'learning_rate': 0.0002902676958110948, 'epoch': 9.43}
                                                      
+
 67%|██████▋   | 7200/10682 [1:01:17<28:39,  2.03it/s]
 67%|██████▋   | 7201/10682 [1:01:18<28:41,  2.02it/s]
 67%|██████▋   | 7202/10682 [1:01:18<28:39,  2.02it/s]
 67%|██████▋   | 7203/10682 [1:01:19<28:38,  2.02it/s]
 67%|██████▋   | 7204/10682 [1:01:19<28:36,  2.03it/s]
 67%|██████▋   | 7205/10682 [1:01:20<28:38,  2.02it/s]
 67%|██████▋   | 7206/10682 [1:01:20<28:36,  2.02it/s]
 67%|██████▋   | 7207/10682 [1:01:20<28:36,  2.02it/s]
 67%|██████▋   | 7208/10682 [1:01:21<28:34,  2.03it/s]
 67%|██████▋   | 7209/10682 [1:01:21<28:37,  2.02it/s]
 67%|██████▋   | 7210/10682 [1:01:22<28:35,  2.02it/s]
 68%|██████▊   | 7211/10682 [1:01:22<28:37,  2.02it/s]
 68%|██████▊   | 7212/10682 [1:01:23<28:35,  2.02it/s]
 68%|██████▊   | 7213/10682 [1:01:23<28:36,  2.02it/s]
 68%|██████▊   | 7214/10682 [1:01:24<28:35,  2.02it/s]
 68%|██████▊   | 7215/10682 [1:01:24<28:34,  2.02it/s]
 68%|██████▊   | 7216/10682 [1:01:25<28:32,  2.02it/s]
 68%|██████▊   | 7217/10682 [1:01:25<28:32,  2.02it/s]
 68%|██████▊   | 7218/10682 [1:01:26<28:32,  2.02it/s]
 68%|██████▊   | 7219/10682 [1:01:26<28:32,  2.02it/s]
 68%|██████▊   | 7220/10682 [1:01:27<28:30,  2.02it/s]
 68%|██████▊   | 7221/10682 [1:01:27<28:29,  2.02it/s]
 68%|██████▊   | 7222/10682 [1:01:28<28:29,  2.02it/s]
 68%|██████▊   | 7223/10682 [1:01:28<28:29,  2.02it/s]
 68%|██████▊   | 7224/10682 [1:01:29<28:29,  2.02it/s]
 68%|██████▊   | 7225/10682 [1:01:29<28:29,  2.02it/s]
                                                      
{'loss': 2.9294, 'grad_norm': 0.2778185307979584, 'learning_rate': 0.0002865664138138335, 'epoch': 9.46}
+
 68%|██████▊   | 7225/10682 [1:01:29<28:29,  2.02it/s]
 68%|██████▊   | 7226/10682 [1:01:30<28:28,  2.02it/s]
 68%|██████▊   | 7227/10682 [1:01:30<28:28,  2.02it/s]
 68%|██████▊   | 7228/10682 [1:01:31<28:27,  2.02it/s]
 68%|██████▊   | 7229/10682 [1:01:31<28:26,  2.02it/s]
 68%|██████▊   | 7230/10682 [1:01:32<28:25,  2.02it/s]
 68%|██████▊   | 7231/10682 [1:01:32<28:24,  2.02it/s]
 68%|██████▊   | 7232/10682 [1:01:33<28:24,  2.02it/s]
 68%|██████▊   | 7233/10682 [1:01:33<28:23,  2.02it/s]
 68%|██████▊   | 7234/10682 [1:01:34<28:23,  2.02it/s]
 68%|██████▊   | 7235/10682 [1:01:34<28:22,  2.02it/s]
 68%|██████▊   | 7236/10682 [1:01:35<28:23,  2.02it/s]
 68%|██████▊   | 7237/10682 [1:01:35<28:22,  2.02it/s]
 68%|██████▊   | 7238/10682 [1:01:36<28:21,  2.02it/s]
 68%|██████▊   | 7239/10682 [1:01:36<28:21,  2.02it/s]
 68%|██████▊   | 7240/10682 [1:01:37<28:22,  2.02it/s]
 68%|██████▊   | 7241/10682 [1:01:37<28:20,  2.02it/s]
 68%|██████▊   | 7242/10682 [1:01:38<28:20,  2.02it/s]
 68%|██████▊   | 7243/10682 [1:01:38<28:20,  2.02it/s]
 68%|██████▊   | 7244/10682 [1:01:39<28:19,  2.02it/s]
 68%|██████▊   | 7245/10682 [1:01:39<28:20,  2.02it/s]
 68%|██████▊   | 7246/10682 [1:01:40<28:19,  2.02it/s]
 68%|██████▊   | 7247/10682 [1:01:40<28:20,  2.02it/s]
 68%|██████▊   | 7248/10682 [1:01:41<28:20,  2.02it/s]
 68%|██████▊   | 7249/10682 [1:01:41<28:18,  2.02it/s]
 68%|██████▊   | 7250/10682 [1:01:42<28:19,  2.02it/s]
                                                      
{'loss': 2.9175, 'grad_norm': 0.26997700333595276, 'learning_rate': 0.0002828793787771929, 'epoch': 9.5}
+
 68%|██████▊   | 7250/10682 [1:01:42<28:19,  2.02it/s]
 68%|██████▊   | 7251/10682 [1:01:42<28:19,  2.02it/s]
 68%|██████▊   | 7252/10682 [1:01:43<28:18,  2.02it/s]
 68%|██████▊   | 7253/10682 [1:01:43<28:16,  2.02it/s]
 68%|██████▊   | 7254/10682 [1:01:44<28:15,  2.02it/s]
 68%|██████▊   | 7255/10682 [1:01:44<28:13,  2.02it/s]
 68%|██████▊   | 7256/10682 [1:01:45<28:12,  2.02it/s]
 68%|██████▊   | 7257/10682 [1:01:45<28:10,  2.03it/s]
 68%|██████▊   | 7258/10682 [1:01:46<28:09,  2.03it/s]
 68%|██████▊   | 7259/10682 [1:01:46<28:09,  2.03it/s]
 68%|██████▊   | 7260/10682 [1:01:47<28:09,  2.03it/s]
 68%|██████▊   | 7261/10682 [1:01:47<28:09,  2.02it/s]
 68%|██████▊   | 7262/10682 [1:01:48<28:07,  2.03it/s]
 68%|██████▊   | 7263/10682 [1:01:48<28:09,  2.02it/s]
 68%|██████▊   | 7264/10682 [1:01:49<28:07,  2.03it/s]
 68%|██████▊   | 7265/10682 [1:01:49<28:07,  2.02it/s]
 68%|██████▊   | 7266/10682 [1:01:50<28:06,  2.03it/s]
 68%|██████▊   | 7267/10682 [1:01:50<28:08,  2.02it/s]
 68%|██████▊   | 7268/10682 [1:01:51<28:06,  2.02it/s]
 68%|██████▊   | 7269/10682 [1:01:51<28:06,  2.02it/s]
 68%|██████▊   | 7270/10682 [1:01:52<28:04,  2.03it/s]
 68%|██████▊   | 7271/10682 [1:01:52<28:05,  2.02it/s]
 68%|██████▊   | 7272/10682 [1:01:53<28:04,  2.02it/s]
 68%|██████▊   | 7273/10682 [1:01:53<28:04,  2.02it/s]
 68%|██████▊   | 7274/10682 [1:01:54<28:04,  2.02it/s]
 68%|██████▊   | 7275/10682 [1:01:54<28:02,  2.02it/s]
                                                      
{'loss': 2.9256, 'grad_norm': 0.27365386486053467, 'learning_rate': 0.0002792068368154043, 'epoch': 9.53}
+
 68%|██████▊   | 7275/10682 [1:01:54<28:02,  2.02it/s]
 68%|██████▊   | 7276/10682 [1:01:55<28:03,  2.02it/s]
 68%|██████▊   | 7277/10682 [1:01:55<28:02,  2.02it/s]
 68%|██████▊   | 7278/10682 [1:01:56<28:02,  2.02it/s]
 68%|██████▊   | 7279/10682 [1:01:56<28:01,  2.02it/s]
 68%|██████▊   | 7280/10682 [1:01:57<28:01,  2.02it/s]
 68%|██████▊   | 7281/10682 [1:01:57<27:59,  2.03it/s]
 68%|██████▊   | 7282/10682 [1:01:58<27:59,  2.02it/s]
 68%|██████▊   | 7283/10682 [1:01:58<28:00,  2.02it/s]
 68%|██████▊   | 7284/10682 [1:01:59<27:57,  2.03it/s]
 68%|██████▊   | 7285/10682 [1:01:59<27:58,  2.02it/s]
 68%|██████▊   | 7286/10682 [1:02:00<27:58,  2.02it/s]
 68%|██████▊   | 7287/10682 [1:02:00<27:58,  2.02it/s]
 68%|██████▊   | 7288/10682 [1:02:01<27:56,  2.02it/s]
 68%|██████▊   | 7289/10682 [1:02:01<27:57,  2.02it/s]
 68%|██████▊   | 7290/10682 [1:02:02<27:54,  2.03it/s]
 68%|██████▊   | 7291/10682 [1:02:02<27:56,  2.02it/s]
 68%|██████▊   | 7292/10682 [1:02:03<27:54,  2.02it/s]
 68%|██████▊   | 7293/10682 [1:02:03<27:54,  2.02it/s]
 68%|██████▊   | 7294/10682 [1:02:03<27:53,  2.02it/s]
 68%|██████▊   | 7295/10682 [1:02:04<27:54,  2.02it/s]
 68%|██████▊   | 7296/10682 [1:02:04<27:53,  2.02it/s]
 68%|██████▊   | 7297/10682 [1:02:05<27:52,  2.02it/s]
 68%|██████▊   | 7298/10682 [1:02:05<27:51,  2.02it/s]
 68%|██████▊   | 7299/10682 [1:02:06<27:50,  2.03it/s]
 68%|██████▊   | 7300/10682 [1:02:06<27:50,  2.02it/s]{'loss': 2.9171, 'grad_norm': 0.2633053660392761, 'learning_rate': 0.0002755490330752679, 'epoch': 9.56}
                                                      
+
 68%|██████▊   | 7300/10682 [1:02:06<27:50,  2.02it/s]
 68%|██████▊   | 7301/10682 [1:02:07<27:51,  2.02it/s]
 68%|██████▊   | 7302/10682 [1:02:07<27:51,  2.02it/s]
 68%|██████▊   | 7303/10682 [1:02:08<27:51,  2.02it/s]
 68%|██████▊   | 7304/10682 [1:02:08<27:50,  2.02it/s]
 68%|██████▊   | 7305/10682 [1:02:09<27:47,  2.03it/s]
 68%|██████▊   | 7306/10682 [1:02:09<27:48,  2.02it/s]
 68%|██████▊   | 7307/10682 [1:02:10<27:45,  2.03it/s]
 68%|██████▊   | 7308/10682 [1:02:10<27:46,  2.02it/s]
 68%|██████▊   | 7309/10682 [1:02:11<27:45,  2.03it/s]
 68%|██████▊   | 7310/10682 [1:02:11<27:46,  2.02it/s]
 68%|██████▊   | 7311/10682 [1:02:12<27:46,  2.02it/s]
 68%|██████▊   | 7312/10682 [1:02:12<27:44,  2.02it/s]
 68%|██████▊   | 7313/10682 [1:02:13<27:45,  2.02it/s]
 68%|██████▊   | 7314/10682 [1:02:13<27:42,  2.03it/s]
 68%|██████▊   | 7315/10682 [1:02:14<27:43,  2.02it/s]
 68%|██████▊   | 7316/10682 [1:02:14<27:41,  2.03it/s]
 68%|██████▊   | 7317/10682 [1:02:15<27:40,  2.03it/s]
 69%|██████▊   | 7318/10682 [1:02:15<27:42,  2.02it/s]
 69%|██████▊   | 7319/10682 [1:02:16<27:41,  2.02it/s]
 69%|██████▊   | 7320/10682 [1:02:16<27:42,  2.02it/s]
 69%|██████▊   | 7321/10682 [1:02:17<27:41,  2.02it/s]
 69%|██████▊   | 7322/10682 [1:02:17<27:40,  2.02it/s]
 69%|██████▊   | 7323/10682 [1:02:18<27:40,  2.02it/s]
 69%|██████▊   | 7324/10682 [1:02:18<27:38,  2.03it/s]
 69%|██████▊   | 7325/10682 [1:02:19<27:39,  2.02it/s]{'loss': 2.9291, 'grad_norm': 0.26645758748054504, 'learning_rate': 0.0002719062117197888, 'epoch': 9.59}
                                                      
+
 69%|██████▊   | 7325/10682 [1:02:19<27:39,  2.02it/s]
 69%|██████▊   | 7326/10682 [1:02:19<27:38,  2.02it/s]
 69%|██████▊   | 7327/10682 [1:02:20<27:38,  2.02it/s]
 69%|██████▊   | 7328/10682 [1:02:20<27:37,  2.02it/s]
 69%|██████▊   | 7329/10682 [1:02:21<27:37,  2.02it/s]
 69%|██████▊   | 7330/10682 [1:02:21<27:37,  2.02it/s]
 69%|██████▊   | 7331/10682 [1:02:22<27:35,  2.02it/s]
 69%|██████▊   | 7332/10682 [1:02:22<27:36,  2.02it/s]
 69%|██████▊   | 7333/10682 [1:02:23<27:35,  2.02it/s]
 69%|██████▊   | 7334/10682 [1:02:23<27:36,  2.02it/s]
 69%|██████▊   | 7335/10682 [1:02:24<27:33,  2.02it/s]
 69%|██████▊   | 7336/10682 [1:02:24<27:34,  2.02it/s]
 69%|██████▊   | 7337/10682 [1:02:25<27:31,  2.03it/s]
 69%|██████▊   | 7338/10682 [1:02:25<27:32,  2.02it/s]
 69%|██████▊   | 7339/10682 [1:02:26<27:31,  2.02it/s]
 69%|██████▊   | 7340/10682 [1:02:26<27:32,  2.02it/s]
 69%|██████▊   | 7341/10682 [1:02:27<27:30,  2.02it/s]
 69%|██████▊   | 7342/10682 [1:02:27<27:29,  2.02it/s]
 69%|██████▊   | 7343/10682 [1:02:28<27:28,  2.02it/s]
 69%|██████▉   | 7344/10682 [1:02:28<27:28,  2.02it/s]
 69%|██████▉   | 7345/10682 [1:02:29<27:27,  2.03it/s]
 69%|██████▉   | 7346/10682 [1:02:29<27:27,  2.03it/s]
 69%|██████▉   | 7347/10682 [1:02:30<27:27,  2.02it/s]
 69%|██████▉   | 7348/10682 [1:02:30<27:26,  2.03it/s]
 69%|██████▉   | 7349/10682 [1:02:31<27:27,  2.02it/s]
 69%|██████▉   | 7350/10682 [1:02:31<27:26,  2.02it/s]
                                                      {'loss': 2.9307, 'grad_norm': 0.2683810591697693, 'learning_rate': 0.00026827861591187944, 'epoch': 9.63}
+
 69%|██████▉   | 7350/10682 [1:02:31<27:26,  2.02it/s]
 69%|██████▉   | 7351/10682 [1:02:32<27:27,  2.02it/s]
 69%|██████▉   | 7352/10682 [1:02:32<27:24,  2.03it/s]
 69%|██████▉   | 7353/10682 [1:02:33<27:26,  2.02it/s]
 69%|██████▉   | 7354/10682 [1:02:33<27:24,  2.02it/s]
 69%|██████▉   | 7355/10682 [1:02:34<27:24,  2.02it/s]
 69%|██████▉   | 7356/10682 [1:02:34<27:22,  2.03it/s]
 69%|██████▉   | 7357/10682 [1:02:35<27:23,  2.02it/s]
 69%|██████▉   | 7358/10682 [1:02:35<27:21,  2.03it/s]
 69%|██████▉   | 7359/10682 [1:02:36<27:22,  2.02it/s]
 69%|██████▉   | 7360/10682 [1:02:36<27:21,  2.02it/s]
 69%|██████▉   | 7361/10682 [1:02:37<27:20,  2.02it/s]
 69%|██████▉   | 7362/10682 [1:02:37<27:18,  2.03it/s]
 69%|██████▉   | 7363/10682 [1:02:38<27:19,  2.02it/s]
 69%|██████▉   | 7364/10682 [1:02:38<27:19,  2.02it/s]
 69%|██████▉   | 7365/10682 [1:02:39<27:18,  2.02it/s]
 69%|██████▉   | 7366/10682 [1:02:39<27:18,  2.02it/s]
 69%|██████▉   | 7367/10682 [1:02:40<27:16,  2.03it/s]
 69%|██████▉   | 7368/10682 [1:02:40<27:16,  2.02it/s]
 69%|██████▉   | 7369/10682 [1:02:41<27:15,  2.03it/s]
 69%|██████▉   | 7370/10682 [1:02:41<27:15,  2.03it/s]
 69%|██████▉   | 7371/10682 [1:02:42<27:13,  2.03it/s]
 69%|██████▉   | 7372/10682 [1:02:42<27:16,  2.02it/s]
 69%|██████▉   | 7373/10682 [1:02:43<27:14,  2.02it/s]
 69%|██████▉   | 7374/10682 [1:02:43<27:14,  2.02it/s]
 69%|██████▉   | 7375/10682 [1:02:44<27:14,  2.02it/s]
                                                      
{'loss': 2.9338, 'grad_norm': 0.26419082283973694, 'learning_rate': 0.00026466648779812794, 'epoch': 9.66}
+
 69%|██████▉   | 7375/10682 [1:02:44<27:14,  2.02it/s]
 69%|██████▉   | 7376/10682 [1:02:44<27:16,  2.02it/s]
 69%|██████▉   | 7377/10682 [1:02:45<27:13,  2.02it/s]
 69%|██████▉   | 7378/10682 [1:02:45<27:16,  2.02it/s]
 69%|██████▉   | 7379/10682 [1:02:46<27:20,  2.01it/s]
 69%|██████▉   | 7380/10682 [1:02:46<27:15,  2.02it/s]
 69%|██████▉   | 7381/10682 [1:02:46<27:15,  2.02it/s]
 69%|██████▉   | 7382/10682 [1:02:47<27:12,  2.02it/s]
 69%|██████▉   | 7383/10682 [1:02:47<27:09,  2.02it/s]
 69%|██████▉   | 7384/10682 [1:02:48<27:09,  2.02it/s]
 69%|██████▉   | 7385/10682 [1:02:48<27:07,  2.03it/s]
 69%|██████▉   | 7386/10682 [1:02:49<27:09,  2.02it/s]
 69%|██████▉   | 7387/10682 [1:02:49<27:07,  2.02it/s]
 69%|██████▉   | 7388/10682 [1:02:50<27:08,  2.02it/s]
 69%|██████▉   | 7389/10682 [1:02:50<27:05,  2.03it/s]
 69%|██████▉   | 7390/10682 [1:02:51<27:06,  2.02it/s]
 69%|██████▉   | 7391/10682 [1:02:51<27:03,  2.03it/s]
 69%|██████▉   | 7392/10682 [1:02:52<27:04,  2.03it/s]
 69%|██████▉   | 7393/10682 [1:02:52<27:03,  2.03it/s]
 69%|██████▉   | 7394/10682 [1:02:53<27:03,  2.03it/s]
 69%|██████▉   | 7395/10682 [1:02:53<27:03,  2.02it/s]
 69%|██████▉   | 7396/10682 [1:02:54<27:01,  2.03it/s]
 69%|██████▉   | 7397/10682 [1:02:54<27:02,  2.02it/s]
 69%|██████▉   | 7398/10682 [1:02:55<27:00,  2.03it/s]
 69%|██████▉   | 7399/10682 [1:02:55<27:03,  2.02it/s]
 69%|██████▉   | 7400/10682 [1:02:56<27:01,  2.02it/s]
                                                      
{'loss': 2.9327, 'grad_norm': 0.2663451135158539, 'learning_rate': 0.0002610700684926339, 'epoch': 9.69}
+
 69%|██████▉   | 7400/10682 [1:02:56<27:01,  2.02it/s]
 69%|██████▉   | 7401/10682 [1:02:56<27:01,  2.02it/s]
 69%|██████▉   | 7402/10682 [1:02:57<27:00,  2.02it/s]
 69%|██████▉   | 7403/10682 [1:02:57<26:59,  2.02it/s]
 69%|██████▉   | 7404/10682 [1:02:58<26:58,  2.02it/s]
 69%|██████▉   | 7405/10682 [1:02:58<26:59,  2.02it/s]
 69%|██████▉   | 7406/10682 [1:02:59<26:59,  2.02it/s]
 69%|██████▉   | 7407/10682 [1:02:59<26:58,  2.02it/s]
 69%|██████▉   | 7408/10682 [1:03:00<26:57,  2.02it/s]
 69%|██████▉   | 7409/10682 [1:03:00<26:57,  2.02it/s]
 69%|██████▉   | 7410/10682 [1:03:01<26:56,  2.02it/s]
 69%|██████▉   | 7411/10682 [1:03:01<26:56,  2.02it/s]
 69%|██████▉   | 7412/10682 [1:03:02<26:56,  2.02it/s]
 69%|██████▉   | 7413/10682 [1:03:02<26:54,  2.02it/s]
 69%|██████▉   | 7414/10682 [1:03:03<27:04,  2.01it/s]
 69%|██████▉   | 7415/10682 [1:03:03<27:01,  2.01it/s]
 69%|██████▉   | 7416/10682 [1:03:04<26:58,  2.02it/s]
 69%|██████▉   | 7417/10682 [1:03:04<26:56,  2.02it/s]
 69%|██████▉   | 7418/10682 [1:03:05<26:54,  2.02it/s]
 69%|██████▉   | 7419/10682 [1:03:05<26:54,  2.02it/s]
 69%|██████▉   | 7420/10682 [1:03:06<26:52,  2.02it/s]
 69%|██████▉   | 7421/10682 [1:03:06<26:52,  2.02it/s]
 69%|██████▉   | 7422/10682 [1:03:07<26:51,  2.02it/s]
 69%|██████▉   | 7423/10682 [1:03:07<26:51,  2.02it/s]
 70%|██████▉   | 7424/10682 [1:03:08<26:50,  2.02it/s]
 70%|██████▉   | 7425/10682 [1:03:08<26:49,  2.02it/s]{'loss': 2.9369, 'grad_norm': 0.27101436257362366, 'learning_rate': 0.00025748959806091423, 'epoch': 9.72}
                                                      
+
 70%|██████▉   | 7425/10682 [1:03:08<26:49,  2.02it/s]
 70%|██████▉   | 7426/10682 [1:03:09<26:50,  2.02it/s]
 70%|██████▉   | 7427/10682 [1:03:09<26:48,  2.02it/s]
 70%|██████▉   | 7428/10682 [1:03:10<26:48,  2.02it/s]
 70%|██████▉   | 7429/10682 [1:03:10<26:47,  2.02it/s]
 70%|██████▉   | 7430/10682 [1:03:11<26:47,  2.02it/s]
 70%|██████▉   | 7431/10682 [1:03:11<26:44,  2.03it/s]
 70%|██████▉   | 7432/10682 [1:03:12<26:47,  2.02it/s]
 70%|██████▉   | 7433/10682 [1:03:12<26:45,  2.02it/s]
 70%|██████▉   | 7434/10682 [1:03:13<26:46,  2.02it/s]
 70%|██████▉   | 7435/10682 [1:03:13<26:44,  2.02it/s]
 70%|██████▉   | 7436/10682 [1:03:14<26:45,  2.02it/s]
 70%|██████▉   | 7437/10682 [1:03:14<26:42,  2.03it/s]
 70%|██████▉   | 7438/10682 [1:03:15<26:42,  2.02it/s]
 70%|██████▉   | 7439/10682 [1:03:15<26:41,  2.02it/s]
 70%|██████▉   | 7440/10682 [1:03:16<26:41,  2.02it/s]
 70%|██████▉   | 7441/10682 [1:03:16<26:41,  2.02it/s]
 70%|██████▉   | 7442/10682 [1:03:17<26:40,  2.02it/s]
 70%|██████▉   | 7443/10682 [1:03:17<26:40,  2.02it/s]
 70%|██████▉   | 7444/10682 [1:03:18<26:41,  2.02it/s]
 70%|██████▉   | 7445/10682 [1:03:18<26:39,  2.02it/s]
 70%|██████▉   | 7446/10682 [1:03:19<26:38,  2.02it/s]
 70%|██████▉   | 7447/10682 [1:03:19<26:39,  2.02it/s]
 70%|██████▉   | 7448/10682 [1:03:20<26:37,  2.02it/s]
 70%|██████▉   | 7449/10682 [1:03:20<26:38,  2.02it/s]
 70%|██████▉   | 7450/10682 [1:03:21<26:36,  2.02it/s]
                                                      
{'loss': 2.9208, 'grad_norm': 0.2662382423877716, 'learning_rate': 0.00025392531550387867, 'epoch': 9.76}
+
 70%|██████▉   | 7450/10682 [1:03:21<26:36,  2.02it/s]
 70%|██████▉   | 7451/10682 [1:03:21<26:38,  2.02it/s]
 70%|██████▉   | 7452/10682 [1:03:22<26:37,  2.02it/s]
 70%|██████▉   | 7453/10682 [1:03:22<26:37,  2.02it/s]
 70%|██████▉   | 7454/10682 [1:03:23<26:36,  2.02it/s]
 70%|██████▉   | 7455/10682 [1:03:23<26:35,  2.02it/s]
 70%|██████▉   | 7456/10682 [1:03:24<26:35,  2.02it/s]
 70%|���█████▉   | 7457/10682 [1:03:24<26:34,  2.02it/s]
 70%|██████▉   | 7458/10682 [1:03:25<26:34,  2.02it/s]
 70%|██████▉   | 7459/10682 [1:03:25<26:33,  2.02it/s]
 70%|██████▉   | 7460/10682 [1:03:26<26:32,  2.02it/s]
 70%|██████▉   | 7461/10682 [1:03:26<26:32,  2.02it/s]
 70%|██████▉   | 7462/10682 [1:03:27<26:30,  2.02it/s]
 70%|██████▉   | 7463/10682 [1:03:27<26:30,  2.02it/s]
 70%|██████▉   | 7464/10682 [1:03:28<26:28,  2.03it/s]
 70%|██████▉   | 7465/10682 [1:03:28<26:28,  2.02it/s]
 70%|██████▉   | 7466/10682 [1:03:28<26:27,  2.03it/s]
 70%|██████▉   | 7467/10682 [1:03:29<26:27,  2.02it/s]
 70%|██████▉   | 7468/10682 [1:03:29<26:26,  2.03it/s]
 70%|██████▉   | 7469/10682 [1:03:30<26:26,  2.03it/s]
 70%|██████▉   | 7470/10682 [1:03:30<26:27,  2.02it/s]
 70%|██████▉   | 7471/10682 [1:03:31<26:25,  2.03it/s]
 70%|██████▉   | 7472/10682 [1:03:31<26:26,  2.02it/s]
 70%|██████▉   | 7473/10682 [1:03:32<26:25,  2.02it/s]
 70%|██████▉   | 7474/10682 [1:03:32<26:25,  2.02it/s]
 70%|██████▉   | 7475/10682 [1:03:33<26:24,  2.02it/s]
                                                      
{'loss': 2.933, 'grad_norm': 0.2700079083442688, 'learning_rate': 0.0002503774587418758, 'epoch': 9.79}
+
 70%|██████▉   | 7475/10682 [1:03:33<26:24,  2.02it/s]
 70%|██████▉   | 7476/10682 [1:03:33<26:25,  2.02it/s]
 70%|██████▉   | 7477/10682 [1:03:34<26:23,  2.02it/s]
 70%|███████   | 7478/10682 [1:03:34<26:23,  2.02it/s]
 70%|███████   | 7479/10682 [1:03:35<26:21,  2.03it/s]
 70%|███████   | 7480/10682 [1:03:35<26:22,  2.02it/s]
 70%|███████   | 7481/10682 [1:03:36<26:21,  2.02it/s]
 70%|███████   | 7482/10682 [1:03:36<26:22,  2.02it/s]
 70%|███████   | 7483/10682 [1:03:37<26:20,  2.02it/s]
 70%|███████   | 7484/10682 [1:03:37<26:20,  2.02it/s]
 70%|███████   | 7485/10682 [1:03:38<26:20,  2.02it/s]
 70%|███████   | 7486/10682 [1:03:38<26:20,  2.02it/s]
 70%|███████   | 7487/10682 [1:03:39<26:19,  2.02it/s]
 70%|███████   | 7488/10682 [1:03:39<26:19,  2.02it/s]
 70%|███████   | 7489/10682 [1:03:40<26:18,  2.02it/s]
 70%|███████   | 7490/10682 [1:03:40<26:16,  2.03it/s]
 70%|███████   | 7491/10682 [1:03:41<26:15,  2.02it/s]
 70%|███████   | 7492/10682 [1:03:41<26:14,  2.03it/s]
 70%|███████   | 7493/10682 [1:03:42<26:16,  2.02it/s]
 70%|███████   | 7494/10682 [1:03:42<26:14,  2.03it/s]
 70%|███████   | 7495/10682 [1:03:43<26:13,  2.03it/s]
 70%|███████   | 7496/10682 [1:03:43<26:12,  2.03it/s]
 70%|███████   | 7497/10682 [1:03:44<26:13,  2.02it/s]
 70%|███████   | 7498/10682 [1:03:44<26:13,  2.02it/s]
 70%|███████   | 7499/10682 [1:03:45<26:13,  2.02it/s]
 70%|███████   | 7500/10682 [1:03:45<26:12,  2.02it/s]{'loss': 2.9348, 'grad_norm': 0.2688334882259369, 'learning_rate': 0.00024684626459881156, 'epoch': 9.82}
                                                      
+
 70%|███████   | 7500/10682 [1:03:45<26:12,  2.02it/s]
 70%|███████   | 7501/10682 [1:03:46<26:12,  2.02it/s]
 70%|███████   | 7502/10682 [1:03:46<26:11,  2.02it/s]
 70%|███████   | 7503/10682 [1:03:47<26:11,  2.02it/s]
 70%|███████   | 7504/10682 [1:03:47<26:10,  2.02it/s]
 70%|███████   | 7505/10682 [1:03:48<26:11,  2.02it/s]
 70%|███████   | 7506/10682 [1:03:48<26:10,  2.02it/s]
 70%|███████   | 7507/10682 [1:03:49<26:10,  2.02it/s]
 70%|███████   | 7508/10682 [1:03:49<26:08,  2.02it/s]
 70%|███████   | 7509/10682 [1:03:50<26:08,  2.02it/s]
 70%|███████   | 7510/10682 [1:03:50<26:07,  2.02it/s]
 70%|███████   | 7511/10682 [1:03:51<26:08,  2.02it/s]
 70%|███████   | 7512/10682 [1:03:51<26:07,  2.02it/s]
 70%|███████   | 7513/10682 [1:03:52<26:06,  2.02it/s]
 70%|███████   | 7514/10682 [1:03:52<26:05,  2.02it/s]
 70%|███████   | 7515/10682 [1:03:53<26:07,  2.02it/s]
 70%|███████   | 7516/10682 [1:03:53<26:06,  2.02it/s]
 70%|███████   | 7517/10682 [1:03:54<26:04,  2.02it/s]
 70%|███████   | 7518/10682 [1:03:54<26:03,  2.02it/s]
 70%|███████   | 7519/10682 [1:03:55<26:03,  2.02it/s]
 70%|███████   | 7520/10682 [1:03:55<26:02,  2.02it/s]
 70%|███████   | 7521/10682 [1:03:56<26:02,  2.02it/s]
 70%|███████   | 7522/10682 [1:03:56<26:01,  2.02it/s]
 70%|███████   | 7523/10682 [1:03:57<25:59,  2.03it/s]
 70%|███████   | 7524/10682 [1:03:57<26:00,  2.02it/s]
 70%|███████   | 7525/10682 [1:03:58<25:58,  2.03it/s]
                                                      {'loss': 2.9204, 'grad_norm': 0.2628709673881531, 'learning_rate': 0.00024333196878634163, 'epoch': 9.86}
+

 70%|███████   | 7525/10682 [1:03:58<25:58,  2.03it/s]
 70%|███████   | 7526/10682 [1:03:58<26:01,  2.02it/s]
 70%|███████   | 7527/10682 [1:03:59<25:58,  2.02it/s]
 70%|███████   | 7528/10682 [1:03:59<25:59,  2.02it/s]
 70%|███████   | 7529/10682 [1:04:00<25:58,  2.02it/s]
 70%|███████   | 7530/10682 [1:04:00<25:58,  2.02it/s]
 71%|███████   | 7531/10682 [1:04:01<25:55,  2.03it/s]
 71%|███████   | 7532/10682 [1:04:01<25:56,  2.02it/s]
 71%|███████   | 7533/10682 [1:04:02<25:55,  2.02it/s]
 71%|███████   | 7534/10682 [1:04:02<25:53,  2.03it/s]
 71%|███████   | 7535/10682 [1:04:03<25:53,  2.03it/s]
 71%|███████   | 7536/10682 [1:04:03<25:51,  2.03it/s]
 71%|███████   | 7537/10682 [1:04:04<25:52,  2.03it/s]
 71%|███████   | 7538/10682 [1:04:04<25:50,  2.03it/s]
 71%|███████   | 7539/10682 [1:04:05<25:51,  2.03it/s]
 71%|███████   | 7540/10682 [1:04:05<25:50,  2.03it/s]
 71%|███████   | 7541/10682 [1:04:06<25:49,  2.03it/s]
 71%|███████   | 7542/10682 [1:04:06<25:50,  2.03it/s]
 71%|███████   | 7543/10682 [1:04:07<25:49,  2.03it/s]
 71%|███████   | 7544/10682 [1:04:07<25:49,  2.03it/s]
 71%|███████   | 7545/10682 [1:04:08<25:48,  2.03it/s]
 71%|███████   | 7546/10682 [1:04:08<25:48,  2.02it/s]
 71%|███████   | 7547/10682 [1:04:09<25:47,  2.03it/s]
 71%|███████   | 7548/10682 [1:04:09<25:47,  2.02it/s]
 71%|███████   | 7549/10682 [1:04:10<25:46,  2.03it/s]
 71%|███████   | 7550/10682 [1:04:10<25:46,  2.03it/s]
                                                      
{'loss': 2.9279, 'grad_norm': 0.26213338971138, 'learning_rate': 0.00023983480588813622, 'epoch': 9.89}
+
 71%|███████   | 7550/10682 [1:04:10<25:46,  2.03it/s]
 71%|███████   | 7551/10682 [1:04:10<25:45,  2.03it/s]
 71%|███████   | 7552/10682 [1:04:11<25:45,  2.03it/s]
 71%|███████   | 7553/10682 [1:04:11<25:44,  2.03it/s]
 71%|███████   | 7554/10682 [1:04:12<25:45,  2.02it/s]
 71%|███████   | 7555/10682 [1:04:12<25:45,  2.02it/s]
 71%|███████   | 7556/10682 [1:04:13<25:43,  2.03it/s]
 71%|███████   | 7557/10682 [1:04:13<25:43,  2.02it/s]
 71%|███████   | 7558/10682 [1:04:14<25:41,  2.03it/s]
 71%|███████   | 7559/10682 [1:04:14<25:42,  2.02it/s]
 71%|███████   | 7560/10682 [1:04:15<25:40,  2.03it/s]
 71%|███████   | 7561/10682 [1:04:15<25:40,  2.03it/s]
 71%|███████   | 7562/10682 [1:04:16<25:40,  2.03it/s]
 71%|███████   | 7563/10682 [1:04:17<27:56,  1.86it/s]
 71%|███████   | 7564/10682 [1:04:17<27:13,  1.91it/s]
 71%|███████   | 7565/10682 [1:04:18<26:47,  1.94it/s]
 71%|███████   | 7566/10682 [1:04:18<26:24,  1.97it/s]
 71%|███████   | 7567/10682 [1:04:19<26:11,  1.98it/s]
 71%|███████   | 7568/10682 [1:04:19<26:00,  2.00it/s]
 71%|███████   | 7569/10682 [1:04:20<25:52,  2.00it/s]
 71%|███████   | 7570/10682 [1:04:20<25:48,  2.01it/s]
 71%|███████   | 7571/10682 [1:04:21<25:43,  2.02it/s]
 71%|███████   | 7572/10682 [1:04:21<25:42,  2.02it/s]
 71%|███████   | 7573/10682 [1:04:22<25:40,  2.02it/s]
 71%|███████   | 7574/10682 [1:04:22<25:38,  2.02it/s]
 71%|███████   | 7575/10682 [1:04:22<25:35,  2.02it/s]
                                                      {'loss': 2.9543, 'grad_norm': 0.27111098170280457, 'learning_rate': 0.00023635500934422239, 'epoch': 9.92}
+
 71%|███████   | 7575/10682 [1:04:22<25:35,  2.02it/s]
 71%|███████   | 7576/10682 [1:04:23<25:37,  2.02it/s]
 71%|███████   | 7577/10682 [1:04:23<25:35,  2.02it/s]
 71%|███████   | 7578/10682 [1:04:24<25:35,  2.02it/s]
 71%|███████   | 7579/10682 [1:04:24<25:34,  2.02it/s]
 71%|███████   | 7580/10682 [1:04:25<25:34,  2.02it/s]
 71%|███████   | 7581/10682 [1:04:25<25:32,  2.02it/s]
 71%|███████   | 7582/10682 [1:04:26<25:31,  2.02it/s]
 71%|███████   | 7583/10682 [1:04:26<25:31,  2.02it/s]
 71%|███████   | 7584/10682 [1:04:27<25:30,  2.02it/s]
 71%|███████   | 7585/10682 [1:04:27<25:30,  2.02it/s]
 71%|███████   | 7586/10682 [1:04:28<25:29,  2.02it/s]
 71%|███████   | 7587/10682 [1:04:28<25:31,  2.02it/s]
 71%|███████   | 7588/10682 [1:04:29<25:28,  2.02it/s]
 71%|███████   | 7589/10682 [1:04:29<25:29,  2.02it/s]
 71%|███████   | 7590/10682 [1:04:30<25:27,  2.02it/s]
 71%|███████   | 7591/10682 [1:04:30<25:27,  2.02it/s]
 71%|███████   | 7592/10682 [1:04:31<25:27,  2.02it/s]
 71%|███████   | 7593/10682 [1:04:31<25:27,  2.02it/s]
 71%|███████   | 7594/10682 [1:04:32<25:28,  2.02it/s]
 71%|███████   | 7595/10682 [1:04:32<25:27,  2.02it/s]
 71%|███████   | 7596/10682 [1:04:33<25:25,  2.02it/s]
 71%|███████   | 7597/10682 [1:04:33<25:26,  2.02it/s]
 71%|███████   | 7598/10682 [1:04:34<25:23,  2.02it/s]
 71%|███████   | 7599/10682 [1:04:34<25:24,  2.02it/s]
 71%|███████   | 7600/10682 [1:04:35<25:22,  2.02it/s]
                                                      
{'loss': 2.9468, 'grad_norm': 0.27557408809661865, 'learning_rate': 0.00023289281143540065, 'epoch': 9.95}
+
 71%|███████   | 7600/10682 [1:04:35<25:22,  2.02it/s]
 71%|███████   | 7601/10682 [1:04:35<25:25,  2.02it/s]
 71%|███████   | 7602/10682 [1:04:36<25:23,  2.02it/s]
 71%|███████   | 7603/10682 [1:04:36<25:22,  2.02it/s]
 71%|███████   | 7604/10682 [1:04:37<25:20,  2.02it/s]
 71%|███████   | 7605/10682 [1:04:37<25:21,  2.02it/s]
 71%|███████   | 7606/10682 [1:04:38<25:19,  2.02it/s]
 71%|███████   | 7607/10682 [1:04:38<25:20,  2.02it/s]
 71%|███████   | 7608/10682 [1:04:39<25:19,  2.02it/s]
 71%|███████   | 7609/10682 [1:04:39<25:18,  2.02it/s]
 71%|███████   | 7610/10682 [1:04:40<25:17,  2.02it/s]
 71%|███████▏  | 7611/10682 [1:04:40<25:18,  2.02it/s]
 71%|███████▏  | 7612/10682 [1:04:41<25:17,  2.02it/s]
 71%|███████▏  | 7613/10682 [1:04:41<25:17,  2.02it/s]
 71%|███████▏  | 7614/10682 [1:04:42<25:16,  2.02it/s]
 71%|███████▏  | 7615/10682 [1:04:42<25:18,  2.02it/s]
 71%|███████▏  | 7616/10682 [1:04:43<25:15,  2.02it/s]
 71%|███████▏  | 7617/10682 [1:04:43<25:15,  2.02it/s]
 71%|███████▏  | 7618/10682 [1:04:44<25:14,  2.02it/s]
 71%|███████▏  | 7619/10682 [1:04:44<25:13,  2.02it/s]
 71%|███████▏  | 7620/10682 [1:04:45<25:12,  2.03it/s]
 71%|███████▏  | 7621/10682 [1:04:45<25:12,  2.02it/s]
 71%|███████▏  | 7622/10682 [1:04:46<25:10,  2.03it/s]
 71%|███████▏  | 7623/10682 [1:04:46<25:11,  2.02it/s]
 71%|███████▏  | 7624/10682 [1:04:47<25:09,  2.03it/s]
 71%|███████▏  | 7625/10682 [1:04:47<25:08,  2.03it/s]{'loss': 2.9439, 'grad_norm': 0.2665178179740906, 'learning_rate': 0.00022944844326774121, 'epoch': 9.99}
                                                      
+
 71%|███████▏  | 7625/10682 [1:04:47<25:08,  2.03it/s]
 71%|███████▏  | 7626/10682 [1:04:48<25:09,  2.02it/s]
 71%|███████▏  | 7627/10682 [1:04:48<25:09,  2.02it/s]
 71%|███████▏  | 7628/10682 [1:04:49<25:09,  2.02it/s]
 71%|███████▏  | 7629/10682 [1:04:49<25:07,  2.02it/s]
 71%|███████▏  | 7630/10682 [1:04:50<25:08,  2.02it/s]
 71%|███████▏  | 7631/10682 [1:04:50<25:07,  2.02it/s]
 71%|███████▏  | 7632/10682 [1:04:51<25:07,  2.02it/s]
 71%|███████▏  | 7633/10682 [1:04:51<25:05,  2.02it/s]
 71%|███████▏  | 7634/10682 [1:04:52<25:05,  2.02it/s]
 71%|███████▏  | 7635/10682 [1:04:52<24:48,  2.05it/s]
 71%|███████▏  | 7636/10682 [1:05:04<3:21:36,  3.97s/it]
 71%|███████▏  | 7637/10682 [1:05:05<2:28:36,  2.93s/it]
 72%|███████▏  | 7638/10682 [1:05:05<1:51:28,  2.20s/it]
 72%|███████▏  | 7639/10682 [1:05:06<1:25:31,  1.69s/it]
 72%|███████▏  | 7640/10682 [1:05:06<1:07:26,  1.33s/it]
 72%|███████▏  | 7641/10682 [1:05:07<54:50,  1.08s/it]  
 72%|███████▏  | 7642/10682 [1:05:07<45:52,  1.10it/s]
 72%|███████▏  | 7643/10682 [1:05:08<39:41,  1.28it/s]
 72%|███████▏  | 7644/10682 [1:05:08<35:15,  1.44it/s]
 72%|███████▏  | 7645/10682 [1:05:09<32:09,  1.57it/s]
 72%|███████▏  | 7646/10682 [1:05:09<30:00,  1.69it/s]
 72%|███████▏  | 7647/10682 [1:05:10<28:28,  1.78it/s]
 72%|███████▏  | 7648/10682 [1:05:10<27:25,  1.84it/s]
 72%|███████▏  | 7649/10682 [1:05:11<26:42,  1.89it/s]
 72%|███████▏  | 7650/10682 [1:05:11<26:10,  1.93it/s]
                                                      
{'loss': 2.8611, 'grad_norm': 0.281035840511322, 'learning_rate': 0.00022602213475715589, 'epoch': 10.02}
+
 72%|███████▏  | 7650/10682 [1:05:11<26:10,  1.93it/s]
 72%|███████▏  | 7651/10682 [1:05:12<25:49,  1.96it/s]
 72%|███████▏  | 7652/10682 [1:05:12<25:34,  1.97it/s]
 72%|███████▏  | 7653/10682 [1:05:13<27:27,  1.84it/s]
 72%|███████▏  | 7654/10682 [1:05:13<26:42,  1.89it/s]
 72%|███████▏  | 7655/10682 [1:05:14<26:13,  1.92it/s]
 72%|███████▏  | 7656/10682 [1:05:14<25:50,  1.95it/s]
 72%|███████▏  | 7657/10682 [1:05:15<25:33,  1.97it/s]
 72%|███████▏  | 7658/10682 [1:05:15<25:24,  1.98it/s]
 72%|███████▏  | 7659/10682 [1:05:16<25:13,  2.00it/s]
\ No newline at end of file