diff --git "a/run-2024-07-04T00:18:50+00:00.log" "b/run-2024-07-04T00:18:50+00:00.log"
--- "a/run-2024-07-04T00:18:50+00:00.log"
+++ "b/run-2024-07-04T00:18:50+00:00.log"
@@ -2561,4 +2561,508 @@ Non-default generation parameters: {'max_length': 200, 'early_stopping': True, '
 
 76%|███████▋  | 50010/65536 [8:36:58<4:44:02,  1.10s/it]
 76%|███████▋  | 50011/65536 [8:36:59<4:08:19,  1.04it/s]
 76%|███████▋  | 50012/65536 [8:36:59<3:43:28,  1.16it/s]
 76%|███████▋  | 50013/65536 [8:37:00<3:23:41,  1.27it/s]
 76%|███████▋  | 50014/65536 [8:37:01<3:15:37,  1.32it/s]
 76%|███████▋  | 50015/65536 [8:37:01<3:10:12,  1.36it/s]
 76%|███████▋  | 50016/65536 [8:37:02<3:03:37,  1.41it/s]
 76%|███████▋  | 50017/65536 [8:37:03<2:56:22,  1.47it/s]
 76%|███████▋  | 50018/65536 [8:37:03<2:53:09,  1.49it/s]
 76%|███████▋  | 50019/65536 [8:37:04<2:49:23,  1.53it/s]
 76%|███████▋  | 50020/65536 [8:37:04<2:42:20,  1.59it/s]
                                                         
{'loss': 1.5363, 'learning_rate': 3.1373209256765026e-07, 'epoch': 3087.65}
 
 76%|███████▋  | 50020/65536 [8:37:04<2:42:20,  1.59it/s]
 76%|███████▋  | 50021/65536 [8:37:05<2:44:29,  1.57it/s]
 76%|███████▋  | 50022/65536 [8:37:06<2:46:31,  1.55it/s]
 76%|███████▋  | 50023/65536 [8:37:06<2:46:07,  1.56it/s]
 76%|███████▋  | 50024/65536 [8:37:07<2:46:00,  1.56it/s]
 76%|███████▋  | 50025/65536 [8:37:08<2:43:43,  1.58it/s]
 76%|███████▋  | 50026/65536 [8:37:08<2:49:41,  1.52it/s]
 76%|███████▋  | 50027/65536 [8:37:09<2:49:52,  1.52it/s]
 76%|███████▋  | 50028/65536 [8:37:10<2:48:26,  1.53it/s]
 76%|███████▋  | 50029/65536 [8:37:10<2:44:00,  1.58it/s]
 76%|███████▋  | 50030/65536 [8:37:11<2:40:33,  1.61it/s]
 76%|███████▋  | 50031/65536 [8:37:11<2:41:13,  1.60it/s]
 76%|███████▋  | 50032/65536 [8:37:12<2:46:25,  1.55it/s]
 76%|███████▋  | 50033/65536 [8:37:13<2:48:37,  1.53it/s]
 76%|███████▋  | 50034/65536 [8:37:13<2:47:47,  1.54it/s]
 76%|███████▋  | 50035/65536 [8:37:14<2:46:49,  1.55it/s]
 76%|███████▋  | 50036/65536 [8:37:15<2:42:47,  1.59it/s]
 76%|███████▋  | 50037/65536 [8:37:15<2:42:44,  1.59it/s]
 76%|███████▋  | 50038/65536 [8:37:16<2:45:13,  1.56it/s]
 76%|███████▋  | 50039/65536 [8:37:17<2:46:11,  1.55it/s]
 76%|███████▋  | 50040/65536 [8:37:17<2:46:03,  1.56it/s]
                                                         
{'loss': 1.5509, 'learning_rate': 3.134565936084241e-07, 'epoch': 3088.89}
 
 76%|███████▋  | 50040/65536 [8:37:17<2:46:03,  1.56it/s]
 76%|███████▋  | 50041/65536 [8:37:18<2:54:18,  1.48it/s]
 76%|███████▋  | 50042/65536 [8:37:19<3:00:54,  1.43it/s]
 76%|███████▋  | 50043/65536 [8:37:19<3:00:27,  1.43it/s]
 76%|███████▋  | 50044/65536 [8:37:20<3:00:40,  1.43it/s]
 76%|███████▋  | 50045/65536 [8:37:21<2:54:15,  1.48it/s]
 76%|███████▋  | 50046/65536 [8:37:21<2:52:59,  1.49it/s]
 76%|███████▋  | 50047/65536 [8:37:22<2:54:37,  1.48it/s]
 76%|███████▋  | 50048/65536 [8:37:23<2:53:27,  1.49it/s]
 76%|███████▋  | 50049/65536 [8:37:23<2:49:20,  1.52it/s]
 76%|███████▋  | 50050/65536 [8:37:24<2:51:37,  1.50it/s]
 76%|███████▋  | 50051/65536 [8:37:25<2:49:20,  1.52it/s]
 76%|███████▋  | 50052/65536 [8:37:25<2:47:18,  1.54it/s]
 76%|███████▋  | 50053/65536 [8:37:26<2:47:28,  1.54it/s]
 76%|███████▋  | 50054/65536 [8:37:27<2:45:26,  1.56it/s]
 76%|███████▋  | 50055/65536 [8:37:27<2:44:08,  1.57it/s]
 76%|███████▋  | 50056/65536 [8:37:28<2:42:23,  1.59it/s]
 76%|███████▋  | 50057/65536 [8:37:29<2:50:43,  1.51it/s]
 76%|███████▋  | 50058/65536 [8:37:29<2:58:22,  1.45it/s]
 76%|███████▋  | 50059/65536 [8:37:30<3:06:14,  1.38it/s]
 76%|███████▋  | 50060/65536 [8:37:31<3:04:03,  1.40it/s]
                                                         
{'loss': 1.5741, 'learning_rate': 3.13181094649198e-07, 'epoch': 3090.12}
-
 76%|███████▋  | 50060/65536 [8:37:31<3:04:03,  1.40it/s]
\ No newline at end of file
+
 76%|███████▋  | 50060/65536 [8:37:31<3:04:03,  1.40it/s]
 76%|███████▋  | 50061/65536 [8:37:32<2:59:54,  1.43it/s]
 76%|███████▋  | 50062/65536 [8:37:32<2:53:22,  1.49it/s]
 76%|███████▋  | 50063/65536 [8:37:33<2:44:49,  1.56it/s]
 76%|███████▋  | 50064/65536 [8:37:33<2:44:53,  1.56it/s]
 76%|███████▋  | 50065/65536 [8:37:34<2:44:12,  1.57it/s]
 76%|███████▋  | 50066/65536 [8:37:35<2:43:17,  1.58it/s]
 76%|███████▋  | 50067/65536 [8:37:35<2:41:39,  1.59it/s]
 76%|███████▋  | 50068/65536 [8:37:36<2:42:24,  1.59it/s]
 76%|███████▋  | 50069/65536 [8:37:36<2:41:28,  1.60it/s]
 76%|███████▋  | 50070/65536 [8:37:37<2:42:16,  1.59it/s]
 76%|███████▋  | 50071/65536 [8:37:38<2:42:48,  1.58it/s]
 76%|███████▋  | 50072/65536 [8:37:38<2:38:21,  1.63it/s]
 76%|███████▋  | 50073/65536 [8:37:39<2:42:49,  1.58it/s]
 76%|███████▋  | 50074/65536 [8:37:40<2:46:01,  1.55it/s]
 76%|███████▋  | 50075/65536 [8:37:40<2:47:26,  1.54it/s]
 76%|███████▋  | 50076/65536 [8:37:41<2:48:23,  1.53it/s]
 76%|███████▋  | 50077/65536 [8:37:42<2:51:31,  1.50it/s]
 76%|███████▋  | 50078/65536 [8:37:42<2:44:12,  1.57it/s]
 76%|███████▋  | 50079/65536 [8:37:43<2:41:01,  1.60it/s]
 76%|███████▋  | 50080/65536 [8:37:43<2:43:38,  1.57it/s]
                                                         
{'loss': 1.58, 'learning_rate': 3.1290559568997183e-07, 'epoch': 3091.36}
+
 76%|███████▋  | 50080/65536 [8:37:43<2:43:38,  1.57it/s]
 76%|███████▋  | 50081/65536 [8:37:44<2:41:01,  1.60it/s]
 76%|███████▋  | 50082/65536 [8:37:45<2:37:57,  1.63it/s]
 76%|███████▋  | 50083/65536 [8:37:45<2:36:14,  1.65it/s]
 76%|███████▋  | 50084/65536 [8:37:46<2:40:49,  1.60it/s]
 76%|███████▋  | 50085/65536 [8:37:47<2:43:23,  1.58it/s]
 76%|███████▋  | 50086/65536 [8:37:47<2:44:01,  1.57it/s]
 76%|███████▋  | 50087/65536 [8:37:48<2:41:43,  1.59it/s]
 76%|███████▋  | 50088/65536 [8:37:48<2:38:37,  1.62it/s]
 76%|███████▋  | 50089/65536 [8:37:49<2:39:49,  1.61it/s]
 76%|███████▋  | 50090/65536 [8:37:50<2:39:29,  1.61it/s]
 76%|███████▋  | 50091/65536 [8:37:50<2:42:28,  1.58it/s]
 76%|███████▋  | 50092/65536 [8:37:51<2:41:31,  1.59it/s]
 76%|███████▋  | 50093/65536 [8:37:52<2:44:32,  1.56it/s]
 76%|███████▋  | 50094/65536 [8:37:52<2:40:20,  1.61it/s]
 76%|███████▋  | 50095/65536 [8:37:53<2:41:15,  1.60it/s]
 76%|███████▋  | 50096/65536 [8:37:53<2:42:29,  1.58it/s]
 76%|███████▋  | 50097/65536 [8:37:54<2:44:41,  1.56it/s]
 76%|███████▋  | 50098/65536 [8:37:55<2:47:18,  1.54it/s]
 76%|███████▋  | 50099/65536 [8:37:55<2:43:43,  1.57it/s]
 76%|███████▋  | 50100/65536 [8:37:56<2:51:02,  1.50it/s]
                                                         
{'loss': 1.5934, 'learning_rate': 3.126300967307457e-07, 'epoch': 3092.59}
+
 76%|███████▋  | 50100/65536 [8:37:56<2:51:02,  1.50it/s]
 76%|███████▋  | 50101/65536 [8:37:57<2:46:34,  1.54it/s]
 76%|███████▋  | 50102/65536 [8:37:57<2:42:25,  1.58it/s]
 76%|███████▋  | 50103/65536 [8:37:58<2:41:48,  1.59it/s]
 76%|███████▋  | 50104/65536 [8:37:59<2:46:35,  1.54it/s]
 76%|███████▋  | 50105/65536 [8:37:59<2:44:06,  1.57it/s]
 76%|███████▋  | 50106/65536 [8:38:00<2:47:23,  1.54it/s]
 76%|███████▋  | 50107/65536 [8:38:01<2:53:10,  1.48it/s]
 76%|███████▋  | 50108/65536 [8:38:01<2:53:10,  1.48it/s]
 76%|███████▋  | 50109/65536 [8:38:02<2:50:29,  1.51it/s]
 76%|███████▋  | 50110/65536 [8:38:03<2:48:09,  1.53it/s]
 76%|███████▋  | 50111/65536 [8:38:03<2:50:30,  1.51it/s]
 76%|███████▋  | 50112/65536 [8:38:04<2:46:23,  1.55it/s]
 76%|███████▋  | 50113/65536 [8:38:05<2:49:52,  1.51it/s]
 76%|███████▋  | 50114/65536 [8:38:05<2:55:39,  1.46it/s]
 76%|███████▋  | 50115/65536 [8:38:06<2:57:49,  1.45it/s]
 76%|███████▋  | 50116/65536 [8:38:07<2:50:25,  1.51it/s]
 76%|███████▋  | 50117/65536 [8:38:07<2:44:36,  1.56it/s]
 76%|███████▋  | 50118/65536 [8:38:08<2:46:13,  1.55it/s]
 76%|███████▋  | 50119/65536 [8:38:09<2:47:02,  1.54it/s]
 76%|███████▋  | 50120/65536 [8:38:09<2:43:27,  1.57it/s]
                                                         
{'loss': 1.5982, 'learning_rate': 3.1235459777151955e-07, 'epoch': 3093.83}
+
 76%|███████▋  | 50120/65536 [8:38:09<2:43:27,  1.57it/s]
 76%|███████▋  | 50121/65536 [8:38:10<2:47:42,  1.53it/s]
 76%|███████▋  | 50122/65536 [8:38:10<2:43:51,  1.57it/s]
 76%|███████▋  | 50123/65536 [8:38:11<2:45:32,  1.55it/s]
 76%|███████▋  | 50124/65536 [8:38:12<2:45:07,  1.56it/s]
 76%|███████▋  | 50125/65536 [8:38:12<2:41:43,  1.59it/s]
 76%|███████▋  | 50126/65536 [8:38:13<2:39:32,  1.61it/s]
 76%|███████▋  | 50127/65536 [8:38:14<2:40:23,  1.60it/s]
 76%|███████▋  | 50128/65536 [8:38:14<2:44:31,  1.56it/s]
 76%|███████▋  | 50129/65536 [8:38:15<2:41:17,  1.59it/s]
 76%|███████▋  | 50130/65536 [8:38:15<2:41:01,  1.59it/s]
 76%|███████▋  | 50131/65536 [8:38:16<2:42:11,  1.58it/s]
 76%|███████▋  | 50132/65536 [8:38:17<2:49:30,  1.51it/s]
 76%|███████▋  | 50133/65536 [8:38:18<2:48:43,  1.52it/s]
 76%|███████▋  | 50134/65536 [8:38:18<2:50:28,  1.51it/s]
 76%|███████▋  | 50135/65536 [8:38:19<2:47:02,  1.54it/s]
 77%|███████▋  | 50136/65536 [8:38:19<2:46:17,  1.54it/s]
 77%|███████▋  | 50137/65536 [8:38:20<2:42:39,  1.58it/s]
 77%|███████▋  | 50138/65536 [8:38:21<2:41:57,  1.58it/s]
 77%|███████▋  | 50139/65536 [8:38:21<2:41:28,  1.59it/s]
 77%|███████▋  | 50140/65536 [8:38:22<2:44:36,  1.56it/s]
                                                         
{'loss': 1.5727, 'learning_rate': 3.120790988122933e-07, 'epoch': 3095.06}
+
 77%|███████▋  | 50140/65536 [8:38:22<2:44:36,  1.56it/s]
 77%|███████▋  | 50141/65536 [8:38:23<2:44:14,  1.56it/s]
 77%|███████▋  | 50142/65536 [8:38:23<2:43:42,  1.57it/s]
 77%|███████▋  | 50143/65536 [8:38:24<2:41:18,  1.59it/s]
 77%|███████▋  | 50144/65536 [8:38:25<2:43:13,  1.57it/s]
 77%|███████▋  | 50145/65536 [8:38:25<2:42:57,  1.57it/s]
 77%|███████▋  | 50146/65536 [8:38:26<2:42:57,  1.57it/s]
 77%|███████▋  | 50147/65536 [8:38:26<2:42:56,  1.57it/s]
 77%|███████▋  | 50148/65536 [8:38:27<2:41:06,  1.59it/s]
 77%|███████▋  | 50149/65536 [8:38:28<2:38:04,  1.62it/s]
 77%|███████▋  | 50150/65536 [8:38:28<2:43:40,  1.57it/s]
 77%|███████▋  | 50151/65536 [8:38:29<2:42:46,  1.58it/s]
 77%|███████▋  | 50152/65536 [8:38:30<2:42:49,  1.57it/s]
 77%|███████▋  | 50153/65536 [8:38:30<2:48:29,  1.52it/s]
 77%|███████▋  | 50154/65536 [8:38:31<2:47:39,  1.53it/s]
 77%|███████▋  | 50155/65536 [8:38:32<2:55:39,  1.46it/s]
 77%|███████▋  | 50156/65536 [8:38:32<3:00:00,  1.42it/s]
 77%|███████▋  | 50157/65536 [8:38:33<2:55:08,  1.46it/s]
 77%|███████▋  | 50158/65536 [8:38:34<2:52:13,  1.49it/s]
 77%|███████▋  | 50159/65536 [8:38:34<2:54:18,  1.47it/s]
 77%|███████▋  | 50160/65536 [8:38:35<2:54:58,  1.46it/s]
                                                         
{'loss': 1.5596, 'learning_rate': 3.1180359985306716e-07, 'epoch': 3096.3}
+
 77%|███████▋  | 50160/65536 [8:38:35<2:54:58,  1.46it/s]
 77%|███████▋  | 50161/65536 [8:38:36<2:50:41,  1.50it/s]
 77%|███████▋  | 50162/65536 [8:38:36<2:54:16,  1.47it/s]
 77%|███████▋  | 50163/65536 [8:38:37<2:49:45,  1.51it/s]
 77%|███████▋  | 50164/65536 [8:38:38<2:49:16,  1.51it/s]
 77%|███████▋  | 50165/65536 [8:38:38<2:51:06,  1.50it/s]
 77%|███████▋  | 50166/65536 [8:38:39<2:44:40,  1.56it/s]
 77%|███████▋  | 50167/65536 [8:38:40<2:40:44,  1.59it/s]
 77%|███████▋  | 50168/65536 [8:38:40<2:39:45,  1.60it/s]
 77%|███████▋  | 50169/65536 [8:38:41<2:39:52,  1.60it/s]
 77%|███████▋  | 50170/65536 [8:38:41<2:42:47,  1.57it/s]
 77%|███████▋  | 50171/65536 [8:38:42<2:44:36,  1.56it/s]
 77%|███████▋  | 50172/65536 [8:38:43<2:47:19,  1.53it/s]
 77%|███████▋  | 50173/65536 [8:38:43<2:44:11,  1.56it/s]
 77%|███████▋  | 50174/65536 [8:38:44<2:46:01,  1.54it/s]
 77%|███████▋  | 50175/65536 [8:38:45<2:50:29,  1.50it/s]
 77%|███████▋  | 50176/65536 [8:38:46<2:55:28,  1.46it/s]
 77%|███████▋  | 50177/65536 [8:38:46<2:51:51,  1.49it/s]
 77%|███████▋  | 50178/65536 [8:38:47<2:50:06,  1.50it/s]
 77%|███████▋  | 50179/65536 [8:38:47<2:48:58,  1.51it/s]
 77%|███████▋  | 50180/65536 [8:38:48<2:47:03,  1.53it/s]
                                                         
{'loss': 1.5842, 'learning_rate': 3.11528100893841e-07, 'epoch': 3097.53}
+
 77%|███████▋  | 50180/65536 [8:38:48<2:47:03,  1.53it/s]
 77%|███████▋  | 50181/65536 [8:38:49<2:45:20,  1.55it/s]
 77%|███████▋  | 50182/65536 [8:38:49<2:42:52,  1.57it/s]
 77%|███████▋  | 50183/65536 [8:38:50<2:45:47,  1.54it/s]
 77%|███████▋  | 50184/65536 [8:38:51<2:43:54,  1.56it/s]
 77%|███████▋  | 50185/65536 [8:38:51<2:42:39,  1.57it/s]
 77%|███████▋  | 50186/65536 [8:38:52<2:45:52,  1.54it/s]
 77%|███████▋  | 50187/65536 [8:38:53<2:45:23,  1.55it/s]
 77%|███████▋  | 50188/65536 [8:38:53<2:52:31,  1.48it/s]
 77%|███████▋  | 50189/65536 [8:38:54<2:52:00,  1.49it/s]
 77%|███████▋  | 50190/65536 [8:38:55<2:48:03,  1.52it/s]
 77%|███████▋  | 50191/65536 [8:38:55<2:46:06,  1.54it/s]
 77%|███████▋  | 50192/65536 [8:38:56<2:46:25,  1.54it/s]
 77%|███████▋  | 50193/65536 [8:38:57<2:52:29,  1.48it/s]
 77%|███████▋  | 50194/65536 [8:38:57<2:50:09,  1.50it/s]
 77%|███████▋  | 50195/65536 [8:38:58<2:48:51,  1.51it/s]
 77%|███████▋  | 50196/65536 [8:38:59<2:44:47,  1.55it/s]
 77%|███████▋  | 50197/65536 [8:38:59<2:40:40,  1.59it/s]
 77%|███████▋  | 50198/65536 [8:39:00<2:39:59,  1.60it/s]
 77%|███████▋  | 50199/65536 [8:39:00<2:37:05,  1.63it/s]
 77%|███████▋  | 50200/65536 [8:39:01<2:38:45,  1.61it/s]
                                                         
{'loss': 1.5643, 'learning_rate': 3.112526019346149e-07, 'epoch': 3098.77}
+
 77%|███████▋  | 50200/65536 [8:39:01<2:38:45,  1.61it/s]
 77%|███████▋  | 50201/65536 [8:39:02<2:36:29,  1.63it/s]
 77%|███████▋  | 50202/65536 [8:39:02<2:34:50,  1.65it/s]
 77%|███████▋  | 50203/65536 [8:39:03<2:36:23,  1.63it/s]
 77%|███████▋  | 50204/65536 [8:39:03<2:38:36,  1.61it/s]
 77%|███████▋  | 50205/65536 [8:39:04<2:36:00,  1.64it/s]
 77%|███████▋  | 50206/65536 [8:39:05<2:34:31,  1.65it/s]
 77%|███████▋  | 50207/65536 [8:39:05<2:34:50,  1.65it/s]
 77%|███████▋  | 50208/65536 [8:39:06<2:32:50,  1.67it/s]
 77%|███████▋  | 50209/65536 [8:39:06<2:38:24,  1.61it/s]
 77%|███████▋  | 50210/65536 [8:39:07<2:38:40,  1.61it/s]
 77%|███████▋  | 50211/65536 [8:39:08<2:37:52,  1.62it/s]
 77%|███████▋  | 50212/65536 [8:39:08<2:34:37,  1.65it/s]
 77%|███████▋  | 50213/65536 [8:39:09<2:36:14,  1.63it/s]
 77%|███████▋  | 50214/65536 [8:39:10<2:38:23,  1.61it/s]
 77%|███████▋  | 50215/65536 [8:39:10<2:37:49,  1.62it/s]
 77%|███████▋  | 50216/65536 [8:39:11<2:34:41,  1.65it/s]
 77%|███████▋  | 50217/65536 [8:39:11<2:35:01,  1.65it/s]
 77%|███████▋  | 50218/65536 [8:39:12<2:42:00,  1.58it/s]
 77%|███████▋  | 50219/65536 [8:39:13<2:41:22,  1.58it/s]
 77%|███████▋  | 50220/65536 [8:39:13<2:40:47,  1.59it/s]
                                                         
{'loss': 1.5784, 'learning_rate': 3.1097710297538874e-07, 'epoch': 3100.0}
+
 77%|███████▋  | 50220/65536 [8:39:13<2:40:47,  1.59it/s]
 77%|███████▋  | 50221/65536 [8:39:14<2:47:08,  1.53it/s]
 77%|███████▋  | 50222/65536 [8:39:15<2:46:03,  1.54it/s]
 77%|███████▋  | 50223/65536 [8:39:15<2:41:53,  1.58it/s]
 77%|███████▋  | 50224/65536 [8:39:16<2:43:53,  1.56it/s]
 77%|███████▋  | 50225/65536 [8:39:17<2:42:35,  1.57it/s]
 77%|███████▋  | 50226/65536 [8:39:17<2:39:22,  1.60it/s]
 77%|███████▋  | 50227/65536 [8:39:18<2:38:33,  1.61it/s]
 77%|███████▋  | 50228/65536 [8:39:18<2:36:44,  1.63it/s]
 77%|███████▋  | 50229/65536 [8:39:19<2:37:04,  1.62it/s]
 77%|███████▋  | 50230/65536 [8:39:20<2:34:53,  1.65it/s]
 77%|███████▋  | 50231/65536 [8:39:20<2:40:40,  1.59it/s]
 77%|███████▋  | 50232/65536 [8:39:21<2:41:35,  1.58it/s]
 77%|███████▋  | 50233/65536 [8:39:21<2:37:50,  1.62it/s]
 77%|███████▋  | 50234/65536 [8:39:22<2:36:11,  1.63it/s]
 77%|███████▋  | 50235/65536 [8:39:23<2:39:31,  1.60it/s]
 77%|███████▋  | 50236/65536 [8:39:23<2:37:19,  1.62it/s]
 77%|███████▋  | 50237/65536 [8:39:24<2:39:38,  1.60it/s]
 77%|███████▋  | 50238/65536 [8:39:25<2:38:00,  1.61it/s]
 77%|███████▋  | 50239/65536 [8:39:25<2:39:53,  1.59it/s]
 77%|███████▋  | 50240/65536 [8:39:26<2:40:40,  1.59it/s]
                                                         
{'loss': 1.5627, 'learning_rate': 3.107016040161626e-07, 'epoch': 3101.23}
+
 77%|███████▋  | 50240/65536 [8:39:26<2:40:40,  1.59it/s]
 77%|███████▋  | 50241/65536 [8:39:27<2:44:28,  1.55it/s]
 77%|███████▋  | 50242/65536 [8:39:27<2:37:24,  1.62it/s]
 77%|███████▋  | 50243/65536 [8:39:28<2:41:50,  1.57it/s]
 77%|███████▋  | 50244/65536 [8:39:28<2:37:10,  1.62it/s]
 77%|███████▋  | 50245/65536 [8:39:29<2:34:23,  1.65it/s]
 77%|███████▋  | 50246/65536 [8:39:29<2:32:39,  1.67it/s]
 77%|███████▋  | 50247/65536 [8:39:30<2:30:55,  1.69it/s]
 77%|███████▋  | 50248/65536 [8:39:31<2:33:50,  1.66it/s]
 77%|███████▋  | 50249/65536 [8:39:31<2:34:54,  1.64it/s]
 77%|███████▋  | 50250/65536 [8:39:32<2:35:56,  1.63it/s]
 77%|███████▋  | 50251/65536 [8:39:33<2:36:48,  1.62it/s]
 77%|███████▋  | 50252/65536 [8:39:33<2:40:07,  1.59it/s]
 77%|███████▋  | 50253/65536 [8:39:34<2:46:10,  1.53it/s]
 77%|███████▋  | 50254/65536 [8:39:35<2:42:47,  1.56it/s]
 77%|███████▋  | 50255/65536 [8:39:35<2:39:25,  1.60it/s]
 77%|███████▋  | 50256/65536 [8:39:36<2:41:54,  1.57it/s]
 77%|███████▋  | 50257/65536 [8:39:36<2:42:14,  1.57it/s]
 77%|███████▋  | 50258/65536 [8:39:37<2:39:58,  1.59it/s]
 77%|███████▋  | 50259/65536 [8:39:38<2:40:23,  1.59it/s]
 77%|███████▋  | 50260/65536 [8:39:38<2:41:29,  1.58it/s]
                                                         
{'loss': 1.54, 'learning_rate': 3.1042610505693646e-07, 'epoch': 3102.47}
+
 77%|███████▋  | 50260/65536 [8:39:38<2:41:29,  1.58it/s]
 77%|███████▋  | 50261/65536 [8:39:39<2:43:42,  1.56it/s]
 77%|███████▋  | 50262/65536 [8:39:40<2:43:07,  1.56it/s]
 77%|███████▋  | 50263/65536 [8:39:40<2:39:16,  1.60it/s]
 77%|███████▋  | 50264/65536 [8:39:41<2:35:49,  1.63it/s]
 77%|███████▋  | 50265/65536 [8:39:41<2:34:31,  1.65it/s]
 77%|███████▋  | 50266/65536 [8:39:42<2:34:20,  1.65it/s]
 77%|███████▋  | 50267/65536 [8:39:43<2:34:13,  1.65it/s]
 77%|███████▋  | 50268/65536 [8:39:43<2:31:11,  1.68it/s]
 77%|███████▋  | 50269/65536 [8:39:44<2:39:05,  1.60it/s]
 77%|███████▋  | 50270/65536 [8:39:44<2:41:03,  1.58it/s]
 77%|███████▋  | 50271/65536 [8:39:45<2:41:22,  1.58it/s]
 77%|███████▋  | 50272/65536 [8:39:46<2:39:59,  1.59it/s]
 77%|███████▋  | 50273/65536 [8:39:46<2:35:00,  1.64it/s]
 77%|███████▋  | 50274/65536 [8:39:47<2:34:08,  1.65it/s]
 77%|███████▋  | 50275/65536 [8:39:48<2:37:24,  1.62it/s]
 77%|███████▋  | 50276/65536 [8:39:48<2:37:36,  1.61it/s]
 77%|███████▋  | 50277/65536 [8:39:49<2:35:42,  1.63it/s]
 77%|███████▋  | 50278/65536 [8:39:49<2:38:49,  1.60it/s]
 77%|███████▋  | 50279/65536 [8:39:50<2:42:07,  1.57it/s]
 77%|███████▋  | 50280/65536 [8:39:51<2:39:30,  1.59it/s]
                                                         
{'loss': 1.5772, 'learning_rate': 3.101506060977102e-07, 'epoch': 3103.7}
+
 77%|███████▋  | 50280/65536 [8:39:51<2:39:30,  1.59it/s]
 77%|███████▋  | 50281/65536 [8:39:51<2:37:39,  1.61it/s]
 77%|███████▋  | 50282/65536 [8:39:52<2:40:30,  1.58it/s]
 77%|███████▋  | 50283/65536 [8:39:53<2:38:54,  1.60it/s]
 77%|███████▋  | 50284/65536 [8:39:53<2:37:17,  1.62it/s]
 77%|███████▋  | 50285/65536 [8:39:54<2:41:27,  1.57it/s]
 77%|███████▋  | 50286/65536 [8:39:54<2:38:06,  1.61it/s]
 77%|███████▋  | 50287/65536 [8:39:55<2:34:18,  1.65it/s]
 77%|███████▋  | 50288/65536 [8:39:56<2:32:47,  1.66it/s]
 77%|███████▋  | 50289/65536 [8:39:56<2:34:11,  1.65it/s]
 77%|███████▋  | 50290/65536 [8:39:57<2:41:14,  1.58it/s]
 77%|███████▋  | 50291/65536 [8:39:58<2:38:26,  1.60it/s]
 77%|███████▋  | 50292/65536 [8:39:58<2:43:25,  1.55it/s]
 77%|███████▋  | 50293/65536 [8:39:59<2:38:29,  1.60it/s]
 77%|███████▋  | 50294/65536 [8:39:59<2:34:28,  1.64it/s]
 77%|███████▋  | 50295/65536 [8:40:00<2:36:46,  1.62it/s]
 77%|███████▋  | 50296/65536 [8:40:01<2:37:41,  1.61it/s]
 77%|███████▋  | 50297/65536 [8:40:01<2:34:45,  1.64it/s]
 77%|███████▋  | 50298/65536 [8:40:02<2:37:34,  1.61it/s]
 77%|███████▋  | 50299/65536 [8:40:02<2:36:28,  1.62it/s]
 77%|███████▋  | 50300/65536 [8:40:03<2:34:51,  1.64it/s]
                                                         
{'loss': 1.5741, 'learning_rate': 3.0987510713848407e-07, 'epoch': 3104.94}
+
 77%|███████▋  | 50300/65536 [8:40:03<2:34:51,  1.64it/s]
 77%|███████▋  | 50301/65536 [8:40:04<2:35:52,  1.63it/s]
 77%|███████▋  | 50302/65536 [8:40:04<2:39:27,  1.59it/s]
 77%|███████▋  | 50303/65536 [8:40:05<2:38:02,  1.61it/s]
 77%|███████▋  | 50304/65536 [8:40:06<2:35:20,  1.63it/s]
 77%|███████▋  | 50305/65536 [8:40:06<2:33:07,  1.66it/s]
 77%|███████▋  | 50306/65536 [8:40:07<2:34:48,  1.64it/s]
 77%|███████▋  | 50307/65536 [8:40:07<2:39:14,  1.59it/s]
 77%|███████▋  | 50308/65536 [8:40:08<2:35:35,  1.63it/s]
 77%|███████▋  | 50309/65536 [8:40:09<2:40:04,  1.59it/s]
 77%|███████▋  | 50310/65536 [8:40:09<2:38:39,  1.60it/s]
 77%|███████▋  | 50311/65536 [8:40:10<2:35:22,  1.63it/s]
 77%|███████▋  | 50312/65536 [8:40:10<2:31:47,  1.67it/s]
 77%|███████▋  | 50313/65536 [8:40:11<2:33:10,  1.66it/s]
 77%|███████▋  | 50314/65536 [8:40:12<2:35:07,  1.64it/s]
 77%|███████▋  | 50315/65536 [8:40:12<2:32:56,  1.66it/s]
 77%|███████▋  | 50316/65536 [8:40:13<2:38:54,  1.60it/s]
 77%|███████▋  | 50317/65536 [8:40:14<2:38:39,  1.60it/s]
 77%|███████▋  | 50318/65536 [8:40:14<2:43:46,  1.55it/s]
 77%|███████▋  | 50319/65536 [8:40:15<2:43:01,  1.56it/s]
 77%|███████▋  | 50320/65536 [8:40:16<2:43:13,  1.55it/s]
                                                         
{'loss': 1.5732, 'learning_rate': 3.09599608179258e-07, 'epoch': 3106.17}
+
 77%|███████▋  | 50320/65536 [8:40:16<2:43:13,  1.55it/s]
 77%|███████▋  | 50321/65536 [8:40:16<2:38:15,  1.60it/s]
 77%|███████▋  | 50322/65536 [8:40:17<2:36:40,  1.62it/s]
 77%|███████▋  | 50323/65536 [8:40:17<2:33:22,  1.65it/s]
 77%|███████▋  | 50324/65536 [8:40:18<2:36:18,  1.62it/s]
 77%|███████▋  | 50325/65536 [8:40:19<2:37:21,  1.61it/s]
 77%|███████▋  | 50326/65536 [8:40:19<2:38:18,  1.60it/s]
 77%|███████▋  | 50327/65536 [8:40:20<2:36:16,  1.62it/s]
 77%|███████▋  | 50328/65536 [8:40:20<2:35:47,  1.63it/s]
 77%|███████▋  | 50329/65536 [8:40:21<2:38:01,  1.60it/s]
 77%|███████▋  | 50330/65536 [8:40:22<2:38:20,  1.60it/s]
 77%|███████▋  | 50331/65536 [8:40:22<2:36:53,  1.62it/s]
 77%|███████▋  | 50332/65536 [8:40:23<2:37:14,  1.61it/s]
 77%|███████▋  | 50333/65536 [8:40:23<2:34:59,  1.63it/s]
 77%|███████▋  | 50334/65536 [8:40:24<2:42:00,  1.56it/s]
 77%|███████▋  | 50335/65536 [8:40:25<2:41:58,  1.56it/s]
 77%|███████▋  | 50336/65536 [8:40:26<2:47:03,  1.52it/s]
 77%|███████▋  | 50337/65536 [8:40:26<2:42:12,  1.56it/s]
 77%|███████▋  | 50338/65536 [8:40:27<2:42:32,  1.56it/s]
 77%|███████▋  | 50339/65536 [8:40:27<2:42:24,  1.56it/s]
 77%|███████▋  | 50340/65536 [8:40:28<2:36:07,  1.62it/s]
                                                         
{'loss': 1.581, 'learning_rate': 3.0932410922003184e-07, 'epoch': 3107.41}
+
 77%|███████▋  | 50340/65536 [8:40:28<2:36:07,  1.62it/s]
 77%|███████▋  | 50341/65536 [8:40:29<2:39:01,  1.59it/s]
 77%|███████▋  | 50342/65536 [8:40:29<2:37:04,  1.61it/s]
 77%|███████▋  | 50343/65536 [8:40:30<2:33:18,  1.65it/s]
 77%|███████▋  | 50344/65536 [8:40:30<2:31:49,  1.67it/s]
 77%|███████▋  | 50345/65536 [8:40:31<2:30:27,  1.68it/s]
 77%|███████▋  | 50346/65536 [8:40:32<2:29:05,  1.70it/s]
 77%|███████▋  | 50347/65536 [8:40:32<2:31:39,  1.67it/s]
 77%|███████▋  | 50348/65536 [8:40:33<2:30:57,  1.68it/s]
 77%|███████▋  | 50349/65536 [8:40:33<2:32:08,  1.66it/s]
 77%|███████▋  | 50350/65536 [8:40:34<2:39:04,  1.59it/s]
 77%|███████▋  | 50351/65536 [8:40:35<2:41:37,  1.57it/s]
 77%|███████▋  | 50352/65536 [8:40:35<2:37:44,  1.60it/s]
 77%|███████▋  | 50353/65536 [8:40:36<2:41:22,  1.57it/s]
 77%|███████▋  | 50354/65536 [8:40:37<2:43:06,  1.55it/s]
 77%|███████▋  | 50355/65536 [8:40:37<2:41:11,  1.57it/s]
 77%|███████▋  | 50356/65536 [8:40:38<2:36:26,  1.62it/s]
 77%|███████▋  | 50357/65536 [8:40:38<2:35:27,  1.63it/s]
 77%|███████▋  | 50358/65536 [8:40:39<2:35:39,  1.63it/s]
 77%|███████▋  | 50359/65536 [8:40:40<2:33:29,  1.65it/s]
 77%|███████▋  | 50360/65536 [8:40:40<2:30:40,  1.68it/s]
                                                         
{'loss': 1.5868, 'learning_rate': 3.090486102608057e-07, 'epoch': 3108.64}
+
 77%|███████▋  | 50360/65536 [8:40:40<2:30:40,  1.68it/s]
 77%|███████▋  | 50361/65536 [8:40:41<2:36:16,  1.62it/s]
 77%|███████▋  | 50362/65536 [8:40:42<2:36:12,  1.62it/s]
 77%|███████▋  | 50363/65536 [8:40:42<2:34:58,  1.63it/s]
 77%|███████▋  | 50364/65536 [8:40:43<2:31:49,  1.67it/s]
 77%|███████▋  | 50365/65536 [8:40:43<2:35:46,  1.62it/s]
 77%|███████▋  | 50366/65536 [8:40:44<2:38:27,  1.60it/s]
 77%|███████▋  | 50367/65536 [8:40:45<2:44:58,  1.53it/s]
 77%|███████▋  | 50368/65536 [8:40:45<2:37:09,  1.61it/s]
 77%|███████▋  | 50369/65536 [8:40:46<2:38:02,  1.60it/s]
 77%|███████▋  | 50370/65536 [8:40:47<2:38:20,  1.60it/s]
 77%|███████▋  | 50371/65536 [8:40:47<2:35:39,  1.62it/s]
 77%|███████▋  | 50372/65536 [8:40:48<2:36:48,  1.61it/s]
 77%|███████▋  | 50373/65536 [8:40:48<2:36:52,  1.61it/s]
 77%|███████▋  | 50374/65536 [8:40:49<2:33:01,  1.65it/s]
 77%|███████▋  | 50375/65536 [8:40:50<2:35:07,  1.63it/s]
 77%|███████▋  | 50376/65536 [8:40:50<2:36:33,  1.61it/s]
 77%|███████▋  | 50377/65536 [8:40:51<2:35:23,  1.63it/s]
 77%|███████▋  | 50378/65536 [8:40:51<2:33:19,  1.65it/s]
 77%|███████▋  | 50379/65536 [8:40:52<2:33:13,  1.65it/s]
 77%|███████▋  | 50380/65536 [8:40:53<2:33:19,  1.65it/s]
                                                         
{'loss': 1.5879, 'learning_rate': 3.0877311130157956e-07, 'epoch': 3109.88}
+
 77%|███████▋  | 50380/65536 [8:40:53<2:33:19,  1.65it/s]
 77%|███████▋  | 50381/65536 [8:40:53<2:33:03,  1.65it/s]
 77%|███████▋  | 50382/65536 [8:40:54<2:37:12,  1.61it/s]
 77%|███████▋  | 50383/65536 [8:40:55<2:40:45,  1.57it/s]
 77%|███████▋  | 50384/65536 [8:40:55<2:38:02,  1.60it/s]
 77%|███████▋  | 50385/65536 [8:40:56<2:36:06,  1.62it/s]
 77%|███████▋  | 50386/65536 [8:40:56<2:36:05,  1.62it/s]
 77%|███████▋  | 50387/65536 [8:40:57<2:35:38,  1.62it/s]
 77%|███████▋  | 50388/65536 [8:40:58<2:36:44,  1.61it/s]
 77%|███████▋  | 50389/65536 [8:40:58<2:39:55,  1.58it/s]
 77%|███████▋  | 50390/65536 [8:40:59<2:37:34,  1.60it/s]
 77%|███████▋  | 50391/65536 [8:40:59<2:37:04,  1.61it/s]
 77%|███████▋  | 50392/65536 [8:41:00<2:36:36,  1.61it/s]
 77%|███████▋  | 50393/65536 [8:41:01<2:38:26,  1.59it/s]
 77%|███████▋  | 50394/65536 [8:41:01<2:36:04,  1.62it/s]
 77%|███████▋  | 50395/65536 [8:41:02<2:36:42,  1.61it/s]
 77%|███████▋  | 50396/65536 [8:41:03<2:37:01,  1.61it/s]
 77%|███████▋  | 50397/65536 [8:41:03<2:40:51,  1.57it/s]
 77%|███████▋  | 50398/65536 [8:41:04<2:40:29,  1.57it/s]
 77%|███████▋  | 50399/65536 [8:41:05<2:41:12,  1.56it/s]
 77%|███████▋  | 50400/65536 [8:41:05<2:36:34,  1.61it/s]
                                                         
{'loss': 1.5726, 'learning_rate': 3.084976123423534e-07, 'epoch': 3111.11}
+
 77%|███████▋  | 50400/65536 [8:41:05<2:36:34,  1.61it/s]
 77%|███████▋  | 50401/65536 [8:41:06<2:39:07,  1.59it/s]
 77%|███████▋  | 50402/65536 [8:41:06<2:41:56,  1.56it/s]
 77%|███████▋  | 50403/65536 [8:41:07<2:41:01,  1.57it/s]
 77%|███████▋  | 50404/65536 [8:41:08<2:40:44,  1.57it/s]
 77%|███████▋  | 50405/65536 [8:41:08<2:41:26,  1.56it/s]
 77%|███████▋  | 50406/65536 [8:41:09<2:36:01,  1.62it/s]
 77%|███████▋  | 50407/65536 [8:41:10<2:35:23,  1.62it/s]
 77%|███████▋  | 50408/65536 [8:41:10<2:37:19,  1.60it/s]
 77%|███████▋  | 50409/65536 [8:41:11<2:35:55,  1.62it/s]
 77%|███████▋  | 50410/65536 [8:41:11<2:35:41,  1.62it/s]
 77%|███████▋  | 50411/65536 [8:41:12<2:39:08,  1.58it/s]
 77%|███████▋  | 50412/65536 [8:41:13<2:37:48,  1.60it/s]
 77%|███████▋  | 50413/65536 [8:41:13<2:34:54,  1.63it/s]
 77%|███████▋  | 50414/65536 [8:41:14<2:32:56,  1.65it/s]
 77%|███████▋  | 50415/65536 [8:41:15<2:37:03,  1.60it/s]
 77%|███████▋  | 50416/65536 [8:41:15<2:37:41,  1.60it/s]
 77%|███████▋  | 50417/65536 [8:41:16<2:37:38,  1.60it/s]
 77%|███████▋  | 50418/65536 [8:41:16<2:32:50,  1.65it/s]
 77%|███████▋  | 50419/65536 [8:41:17<2:35:05,  1.62it/s]
 77%|███████▋  | 50420/65536 [8:41:18<2:33:03,  1.65it/s]
                                                         
{'loss': 1.5422, 'learning_rate': 3.0822211338312717e-07, 'epoch': 3112.35}
+
 77%|███████▋  | 50420/65536 [8:41:18<2:33:03,  1.65it/s]
 77%|███████▋  | 50421/65536 [8:41:18<2:33:45,  1.64it/s]
 77%|███████▋  | 50422/65536 [8:41:19<2:33:43,  1.64it/s]
 77%|███████▋  | 50423/65536 [8:41:19<2:34:15,  1.63it/s]
 77%|███████▋  | 50424/65536 [8:41:20<2:39:48,  1.58it/s]
 77%|███████▋  | 50425/65536 [8:41:21<2:37:57,  1.59it/s]
 77%|███████▋  | 50426/65536 [8:41:21<2:37:46,  1.60it/s]
 77%|███████▋  | 50427/65536 [8:41:22<2:35:53,  1.62it/s]
 77%|███████▋  | 50428/65536 [8:41:23<2:36:27,  1.61it/s]
 77%|███████▋  | 50429/65536 [8:41:23<2:34:40,  1.63it/s]
 77%|███████▋  | 50430/65536 [8:41:24<2:36:40,  1.61it/s]
 77%|███████▋  | 50431/65536 [8:41:24<2:39:01,  1.58it/s]
 77%|██��████▋  | 50432/65536 [8:41:25<2:38:14,  1.59it/s]
 77%|███████▋  | 50433/65536 [8:41:26<2:38:33,  1.59it/s]
 77%|███████▋  | 50434/65536 [8:41:26<2:34:19,  1.63it/s]
 77%|███████▋  | 50435/65536 [8:41:27<2:36:21,  1.61it/s]
 77%|███████▋  | 50436/65536 [8:41:28<2:36:38,  1.61it/s]
 77%|███████▋  | 50437/65536 [8:41:28<2:37:18,  1.60it/s]
 77%|███████▋  | 50438/65536 [8:41:29<2:38:07,  1.59it/s]
 77%|███████▋  | 50439/65536 [8:41:29<2:35:11,  1.62it/s]
 77%|███████▋  | 50440/65536 [8:41:30<2:34:46,  1.63it/s]
                                                         
{'loss': 1.5575, 'learning_rate': 3.0794661442390103e-07, 'epoch': 3113.58}
+
 77%|███████▋  | 50440/65536 [8:41:30<2:34:46,  1.63it/s]
 77%|███████▋  | 50441/65536 [8:41:31<2:37:20,  1.60it/s]
 77%|███████▋  | 50442/65536 [8:41:31<2:34:55,  1.62it/s]
 77%|███████▋  | 50443/65536 [8:41:32<2:33:49,  1.64it/s]
 77%|███████▋  | 50444/65536 [8:41:32<2:36:10,  1.61it/s]
 77%|███████▋  | 50445/65536 [8:41:33<2:34:10,  1.63it/s]
 77%|███████▋  | 50446/65536 [8:41:34<2:36:06,  1.61it/s]
 77%|███████▋  | 50447/65536 [8:41:34<2:44:38,  1.53it/s]
 77%|███████▋  | 50448/65536 [8:41:35<2:40:12,  1.57it/s]
 77%|███████▋  | 50449/65536 [8:41:36<2:37:19,  1.60it/s]
 77%|███████▋  | 50450/65536 [8:41:36<2:36:23,  1.61it/s]
 77%|███████▋  | 50451/65536 [8:41:37<2:40:12,  1.57it/s]
 77%|███████▋  | 50452/65536 [8:41:38<2:36:34,  1.61it/s]
 77%|███████▋  | 50453/65536 [8:41:38<2:38:19,  1.59it/s]
 77%|███████▋  | 50454/65536 [8:41:39<2:37:13,  1.60it/s]
 77%|███████▋  | 50455/65536 [8:41:39<2:37:46,  1.59it/s]
 77%|███████▋  | 50456/65536 [8:41:40<2:33:43,  1.63it/s]
 77%|███████▋  | 50457/65536 [8:41:41<2:39:13,  1.58it/s]
 77%|███████▋  | 50458/65536 [8:41:41<2:39:12,  1.58it/s]
 77%|███████▋  | 50459/65536 [8:41:42<2:33:51,  1.63it/s]
 77%|███████▋  | 50460/65536 [8:41:42<2:33:36,  1.64it/s]
                                                         
{'loss': 1.5783, 'learning_rate': 3.076711154646749e-07, 'epoch': 3114.81}
+
 77%|███████▋  | 50460/65536 [8:41:42<2:33:36,  1.64it/s]
 77%|███████▋  | 50461/65536 [8:41:43<2:36:14,  1.61it/s]
 77%|███████▋  | 50462/65536 [8:41:44<2:36:21,  1.61it/s]
 77%|███████▋  | 50463/65536 [8:41:44<2:34:50,  1.62it/s]
 77%|███████▋  | 50464/65536 [8:41:45<2:38:29,  1.58it/s]
 77%|███████▋  | 50465/65536 [8:41:46<2:35:57,  1.61it/s]
 77%|███████▋  | 50466/65536 [8:41:46<2:40:59,  1.56it/s]
 77%|███████▋  | 50467/65536 [8:41:47<2:35:31,  1.61it/s]
 77%|███████▋  | 50468/65536 [8:41:48<2:39:10,  1.58it/s]
 77%|███████▋  | 50469/65536 [8:41:48<2:39:10,  1.58it/s]
 77%|███████▋  | 50470/65536 [8:41:49<2:39:23,  1.58it/s]
 77%|███████▋  | 50471/65536 [8:41:49<2:37:04,  1.60it/s]
 77%|███████▋  | 50472/65536 [8:41:50<2:34:06,  1.63it/s]
 77%|███████▋  | 50473/65536 [8:41:51<2:34:54,  1.62it/s]
 77%|███████▋  | 50474/65536 [8:41:51<2:35:23,  1.62it/s]
 77%|███████▋  | 50475/65536 [8:41:52<2:33:20,  1.64it/s]
 77%|███████▋  | 50476/65536 [8:41:52<2:33:09,  1.64it/s]
 77%|███████▋  | 50477/65536 [8:41:53<2:31:28,  1.66it/s]
 77%|███████▋  | 50478/65536 [8:41:54<2:29:30,  1.68it/s]
 77%|███████▋  | 50479/65536 [8:41:54<2:31:54,  1.65it/s]
 77%|███████▋  | 50480/65536 [8:41:55<2:35:05,  1.62it/s]
                                                         
{'loss': 1.5554, 'learning_rate': 3.0739561650544874e-07, 'epoch': 3116.05}
+
 77%|███████▋  | 50480/65536 [8:41:55<2:35:05,  1.62it/s]
 77%|███████▋  | 50481/65536 [8:41:56<2:35:26,  1.61it/s]
 77%|███████▋  | 50482/65536 [8:41:56<2:35:41,  1.61it/s]
 77%|███████▋  | 50483/65536 [8:41:57<2:35:56,  1.61it/s]
 77%|███████▋  | 50484/65536 [8:41:57<2:38:31,  1.58it/s]
 77%|███████▋  | 50485/65536 [8:41:58<2:34:42,  1.62it/s]
 77%|███████▋  | 50486/65536 [8:41:59<2:35:28,  1.61it/s]
 77%|███████▋  | 50487/65536 [8:41:59<2:31:40,  1.65it/s]
 77%|███████▋  | 50488/65536 [8:42:00<2:33:11,  1.64it/s]
 77%|███████▋  | 50489/65536 [8:42:00<2:32:56,  1.64it/s]
 77%|███████▋  | 50490/65536 [8:42:01<2:33:31,  1.63it/s]
 77%|███████▋  | 50491/65536 [8:42:02<2:29:44,  1.67it/s]
 77%|███████▋  | 50492/65536 [8:42:02<2:30:04,  1.67it/s]
 77%|███████▋  | 50493/65536 [8:42:03<2:34:40,  1.62it/s]
 77%|███████▋  | 50494/65536 [8:42:03<2:33:38,  1.63it/s]
 77%|███████▋  | 50495/65536 [8:42:04<2:32:35,  1.64it/s]
 77%|███████▋  | 50496/65536 [8:42:05<2:36:00,  1.61it/s]
 77%|███████▋  | 50497/65536 [8:42:05<2:34:59,  1.62it/s]
 77%|███████▋  | 50498/65536 [8:42:06<2:33:41,  1.63it/s]
 77%|███████▋  | 50499/65536 [8:42:07<2:31:09,  1.66it/s]
 77%|███████▋  | 50500/65536 [8:42:07<2:30:45,  1.66it/s]
                                                         
{'loss': 1.5733, 'learning_rate': 3.071201175462226e-07, 'epoch': 3117.28}
+
 77%|███████▋  | 50500/65536 [8:42:07<2:30:45,  1.66it/s]
 77%|███████▋  | 50501/65536 [8:42:08<2:33:48,  1.63it/s]
 77%|███████▋  | 50502/65536 [8:42:08<2:37:00,  1.60it/s]
 77%|███████▋  | 50503/65536 [8:42:09<2:35:55,  1.61it/s]
 77%|███████▋  | 50504/65536 [8:42:10<2:43:25,  1.53it/s]
 77%|███████▋  | 50505/65536 [8:42:10<2:47:45,  1.49it/s]
 77%|███████▋  | 50506/65536 [8:42:11<2:41:47,  1.55it/s]
 77%|███████▋  | 50507/65536 [8:42:12<2:43:40,  1.53it/s]
 77%|███████▋  | 50508/65536 [8:42:12<2:43:31,  1.53it/s]
 77%|███████▋  | 50509/65536 [8:42:13<2:39:23,  1.57it/s]
 77%|███████▋  | 50510/65536 [8:42:14<2:38:07,  1.58it/s]
 77%|███████▋  | 50511/65536 [8:42:14<2:34:51,  1.62it/s]
 77%|███████▋  | 50512/65536 [8:42:15<2:35:58,  1.61it/s]
 77%|███████▋  | 50513/65536 [8:42:15<2:33:13,  1.63it/s]
 77%|███████▋  | 50514/65536 [8:42:16<2:36:00,  1.60it/s]
 77%|███████▋  | 50515/65536 [8:42:17<2:38:08,  1.58it/s]
 77%|███████▋  | 50516/65536 [8:42:17<2:37:04,  1.59it/s]
 77%|███████▋  | 50517/65536 [8:42:18<2:36:35,  1.60it/s]
 77%|███████▋  | 50518/65536 [8:42:19<2:37:29,  1.59it/s]
 77%|███████▋  | 50519/65536 [8:42:19<2:36:59,  1.59it/s]
 77%|███████▋  | 50520/65536 [8:42:20<2:35:34,  1.61it/s]
                                                         
{'loss': 1.5429, 'learning_rate': 3.0684461858699646e-07, 'epoch': 3118.52}
+
 77%|███████▋  | 50520/65536 [8:42:20<2:35:34,  1.61it/s]
 77%|███████▋  | 50521/65536 [8:42:20<2:33:55,  1.63it/s]
 77%|███████▋  | 50522/65536 [8:42:21<2:33:27,  1.63it/s]
 77%|███████▋  | 50523/65536 [8:42:22<2:29:17,  1.68it/s]
 77%|███████▋  | 50524/65536 [8:42:22<2:32:35,  1.64it/s]
 77%|███████▋  | 50525/65536 [8:42:23<2:34:38,  1.62it/s]
 77%|███████▋  | 50526/65536 [8:42:23<2:36:08,  1.60it/s]
 77%|███████▋  | 50527/65536 [8:42:24<2:34:48,  1.62it/s]
 77%|███████▋  | 50528/65536 [8:42:25<2:38:01,  1.58it/s]
 77%|███████▋  | 50529/65536 [8:42:25<2:36:46,  1.60it/s]
 77%|███████▋  | 50530/65536 [8:42:26<2:36:41,  1.60it/s]
 77%|███████▋  | 50531/65536 [8:42:27<2:36:12,  1.60it/s]
 77%|███████▋  | 50532/65536 [8:42:27<2:32:56,  1.64it/s]
 77%|███████▋  | 50533/65536 [8:42:28<2:29:02,  1.68it/s]
 77%|███████▋  | 50534/65536 [8:42:28<2:32:14,  1.64it/s]
 77%|███████▋  | 50535/65536 [8:42:29<2:34:03,  1.62it/s]
 77%|███████▋  | 50536/65536 [8:42:30<2:37:59,  1.58it/s]
 77%|███████▋  | 50537/65536 [8:42:30<2:35:46,  1.60it/s]
 77%|███████▋  | 50538/65536 [8:42:31<2:34:46,  1.62it/s]
 77%|███████▋  | 50539/65536 [8:42:31<2:30:51,  1.66it/s]
 77%|███████▋  | 50540/65536 [8:42:32<2:25:58,  1.71it/s]
                                                         
{'loss': 1.5658, 'learning_rate': 3.0656911962777027e-07, 'epoch': 3119.75}
+
 77%|███████▋  | 50540/65536 [8:42:32<2:25:58,  1.71it/s]
 77%|███████▋  | 50541/65536 [8:42:33<2:25:45,  1.71it/s]
 77%|███████▋  | 50542/65536 [8:42:33<2:32:12,  1.64it/s]
 77%|███████▋  | 50543/65536 [8:42:34<2:36:48,  1.59it/s]
 77%|███████▋  | 50544/65536 [8:42:35<2:32:47,  1.64it/s]
 77%|███████▋  | 50545/65536 [8:42:35<2:38:25,  1.58it/s]
 77%|███████▋  | 50546/65536 [8:42:36<2:37:29,  1.59it/s]
 77%|███████▋  | 50547/65536 [8:42:36<2:38:35,  1.58it/s]
 77%|███████▋  | 50548/65536 [8:42:37<2:35:52,  1.60it/s]
 77%|███████▋  | 50549/65536 [8:42:38<2:36:32,  1.60it/s]
 77%|███████▋  | 50550/65536 [8:42:38<2:33:38,  1.63it/s]
 77%|███████▋  | 50551/65536 [8:42:39<2:32:16,  1.64it/s]
 77%|███████▋  | 50552/65536 [8:42:40<2:33:53,  1.62it/s]
 77%|███████▋  | 50553/65536 [8:42:40<2:34:43,  1.61it/s]
 77%|███████▋  | 50554/65536 [8:42:41<2:34:43,  1.61it/s]
 77%|███████▋  | 50555/65536 [8:42:42<2:42:51,  1.53it/s]
 77%|███████▋  | 50556/65536 [8:42:42<2:41:42,  1.54it/s]
 77%|███████▋  | 50557/65536 [8:42:43<2:39:28,  1.57it/s]
 77%|███████▋  | 50558/65536 [8:42:43<2:38:34,  1.57it/s]
 77%|███████▋  | 50559/65536 [8:42:44<2:34:31,  1.62it/s]
 77%|███████▋  | 50560/65536 [8:42:45<2:32:05,  1.64it/s]
                                                         
{'loss': 1.5662, 'learning_rate': 3.0629362066854413e-07, 'epoch': 3120.99}
+
 77%|███████▋  | 50560/65536 [8:42:45<2:32:05,  1.64it/s]
 77%|███████▋  | 50561/65536 [8:42:45<2:41:36,  1.54it/s]
 77%|███████▋  | 50562/65536 [8:42:46<2:41:29,  1.55it/s]
 77%|███████▋  | 50563/65536 [8:42:47<2:37:38,  1.58it/s]
 77%|███████▋  | 50564/65536 [8:42:47<2:35:24,  1.61it/s]
 77%|███████▋  | 50565/65536 [8:42:48<2:38:59,  1.57it/s]
 77%|███████▋  | 50566/65536 [8:42:48<2:36:10,  1.60it/s]
 77%|███████▋  | 50567/65536 [8:42:49<2:35:13,  1.61it/s]
 77%|███████▋  | 50568/65536 [8:42:50<2:32:01,  1.64it/s]
 77%|███████▋  | 50569/65536 [8:42:50<2:32:44,  1.63it/s]
 77%|███████▋  | 50570/65536 [8:42:51<2:30:22,  1.66it/s]
 77%|███████▋  | 50571/65536 [8:42:51<2:34:35,  1.61it/s]
 77%|███████▋  | 50572/65536 [8:42:52<2:36:14,  1.60it/s]
 77%|███████▋  | 50573/65536 [8:42:53<2:38:38,  1.57it/s]
 77%|███████▋  | 50574/65536 [8:42:53<2:35:54,  1.60it/s]
 77%|███████▋  | 50575/65536 [8:42:54<2:33:44,  1.62it/s]
 77%|███████▋  | 50576/65536 [8:42:54<2:28:41,  1.68it/s]
 77%|███████▋  | 50577/65536 [8:42:55<2:32:43,  1.63it/s]
 77%|███████▋  | 50578/65536 [8:42:56<2:34:18,  1.62it/s]
 77%|███████▋  | 50579/65536 [8:42:56<2:32:57,  1.63it/s]
 77%|███████▋  | 50580/65536 [8:42:57<2:40:15,  1.56it/s]
                                                         
{'loss': 1.5615, 'learning_rate': 3.06018121709318e-07, 'epoch': 3122.22}
+
 77%|███████▋  | 50580/65536 [8:42:57<2:40:15,  1.56it/s]
 77%|███████▋  | 50581/65536 [8:42:58<2:36:41,  1.59it/s]
 77%|███████▋  | 50582/65536 [8:42:58<2:34:53,  1.61it/s]
 77%|███████▋  | 50583/65536 [8:42:59<2:35:39,  1.60it/s]
 77%|███████▋  | 50584/65536 [8:43:00<2:31:49,  1.64it/s]
 77%|███████▋  | 50585/65536 [8:43:00<2:30:19,  1.66it/s]
 77%|███████▋  | 50586/65536 [8:43:01<2:31:24,  1.65it/s]
 77%|███████▋  | 50587/65536 [8:43:01<2:33:55,  1.62it/s]
 77%|███████▋  | 50588/65536 [8:43:02<2:32:04,  1.64it/s]
 77%|███████▋  | 50589/65536 [8:43:03<2:32:46,  1.63it/s]
 77%|███████▋  | 50590/65536 [8:43:03<2:33:46,  1.62it/s]
 77%|███████▋  | 50591/65536 [8:43:04<2:32:04,  1.64it/s]
 77%|███████▋  | 50592/65536 [8:43:04<2:32:37,  1.63it/s]
 77%|███████▋  | 50593/65536 [8:43:05<2:36:44,  1.59it/s]
 77%|███████▋  | 50594/65536 [8:43:06<2:38:18,  1.57it/s]
 77%|███████▋  | 50595/65536 [8:43:06<2:38:51,  1.57it/s]
 77%|███████▋  | 50596/65536 [8:43:07<2:39:40,  1.56it/s]
 77%|███████▋  | 50597/65536 [8:43:08<2:35:16,  1.60it/s]
 77%|███████▋  | 50598/65536 [8:43:08<2:34:08,  1.62it/s]
 77%|███████▋  | 50599/65536 [8:43:09<2:30:32,  1.65it/s]
 77%|███████▋  | 50600/65536 [8:43:09<2:30:20,  1.66it/s]
                                                         
{'loss': 1.5678, 'learning_rate': 3.0574262275009185e-07, 'epoch': 3123.46}
+
 77%|███████▋  | 50600/65536 [8:43:09<2:30:20,  1.66it/s]
 77%|███████▋  | 50601/65536 [8:43:10<2:28:23,  1.68it/s]
 77%|███████▋  | 50602/65536 [8:43:11<2:27:18,  1.69it/s]
 77%|███████▋  | 50603/65536 [8:43:11<2:28:27,  1.68it/s]
 77%|███████▋  | 50604/65536 [8:43:12<2:29:57,  1.66it/s]
 77%|███████▋  | 50605/65536 [8:43:12<2:32:01,  1.64it/s]
 77%|███████▋  | 50606/65536 [8:43:13<2:32:17,  1.63it/s]
 77%|███████▋  | 50607/65536 [8:43:14<2:29:31,  1.66it/s]
 77%|███████▋  | 50608/65536 [8:43:14<2:33:09,  1.62it/s]
 77%|███████▋  | 50609/65536 [8:43:15<2:40:12,  1.55it/s]
 77%|███████▋  | 50610/65536 [8:43:16<2:39:25,  1.56it/s]
 77%|███████▋  | 50611/65536 [8:43:16<2:36:37,  1.59it/s]
 77%|███████▋  | 50612/65536 [8:43:17<2:35:22,  1.60it/s]
 77%|███████▋  | 50613/65536 [8:43:17<2:38:59,  1.56it/s]
 77%|████��██▋  | 50614/65536 [8:43:18<2:35:34,  1.60it/s]
 77%|███████▋  | 50615/65536 [8:43:19<2:33:29,  1.62it/s]
 77%|███████▋  | 50616/65536 [8:43:19<2:33:18,  1.62it/s]
 77%|███████▋  | 50617/65536 [8:43:20<2:34:11,  1.61it/s]
 77%|███████▋  | 50618/65536 [8:43:20<2:32:04,  1.63it/s]
 77%|███████▋  | 50619/65536 [8:43:21<2:35:26,  1.60it/s]
 77%|███████▋  | 50620/65536 [8:43:22<2:34:23,  1.61it/s]
                                                         
{'loss': 1.5542, 'learning_rate': 3.054671237908657e-07, 'epoch': 3124.69}
+
 77%|███████▋  | 50620/65536 [8:43:22<2:34:23,  1.61it/s]
 77%|███████▋  | 50621/65536 [8:43:22<2:32:40,  1.63it/s]
 77%|███████▋  | 50622/65536 [8:43:23<2:32:04,  1.63it/s]
 77%|███████▋  | 50623/65536 [8:43:24<2:30:17,  1.65it/s]
 77%|███████▋  | 50624/65536 [8:43:24<2:36:38,  1.59it/s]
 77%|███████▋  | 50625/65536 [8:43:25<2:33:31,  1.62it/s]
 77%|███████▋  | 50626/65536 [8:43:26<2:39:07,  1.56it/s]
 77%|███████▋  | 50627/65536 [8:43:26<2:35:26,  1.60it/s]
 77%|███████▋  | 50628/65536 [8:43:27<2:35:04,  1.60it/s]
 77%|███████▋  | 50629/65536 [8:43:27<2:33:08,  1.62it/s]
 77%|███████▋  | 50630/65536 [8:43:28<2:33:44,  1.62it/s]
 77%|███████▋  | 50631/65536 [8:43:29<2:33:59,  1.61it/s]
 77%|███████▋  | 50632/65536 [8:43:29<2:32:21,  1.63it/s]
 77%|███████▋  | 50633/65536 [8:43:30<2:36:44,  1.58it/s]
 77%|███████▋  | 50634/65536 [8:43:30<2:35:46,  1.59it/s]
 77%|███████▋  | 50635/65536 [8:43:31<2:32:00,  1.63it/s]
 77%|███████▋  | 50636/65536 [8:43:32<2:34:43,  1.60it/s]
 77%|███████▋  | 50637/65536 [8:43:32<2:34:57,  1.60it/s]
 77%|███████▋  | 50638/65536 [8:43:33<2:34:35,  1.61it/s]
 77%|███████▋  | 50639/65536 [8:43:34<2:32:47,  1.62it/s]
 77%|███████▋  | 50640/65536 [8:43:34<2:28:01,  1.68it/s]
                                                         
{'loss': 1.5804, 'learning_rate': 3.0519162483163956e-07, 'epoch': 3125.93}
+
 77%|███████▋  | 50640/65536 [8:43:34<2:28:01,  1.68it/s]
 77%|███████▋  | 50641/65536 [8:43:35<2:31:32,  1.64it/s]
 77%|███████▋  | 50642/65536 [8:43:35<2:34:51,  1.60it/s]
 77%|███████▋  | 50643/65536 [8:43:36<2:29:15,  1.66it/s]
 77%|███████▋  | 50644/65536 [8:43:37<2:32:53,  1.62it/s]
 77%|███████▋  | 50645/65536 [8:43:37<2:32:57,  1.62it/s]
 77%|███████▋  | 50646/65536 [8:43:38<2:31:53,  1.63it/s]
 77%|███████▋  | 50647/65536 [8:43:38<2:31:38,  1.64it/s]
 77%|███████▋  | 50648/65536 [8:43:39<2:31:33,  1.64it/s]
 77%|███████▋  | 50649/65536 [8:43:40<2:31:05,  1.64it/s]
 77%|███████▋  | 50650/65536 [8:43:40<2:34:28,  1.61it/s]
 77%|███████▋  | 50651/65536 [8:43:41<2:32:51,  1.62it/s]
 77%|███████▋  | 50652/65536 [8:43:42<2:36:13,  1.59it/s]
 77%|███████▋  | 50653/65536 [8:43:42<2:35:29,  1.60it/s]
 77%|███████▋  | 50654/65536 [8:43:43<2:42:44,  1.52it/s]
 77%|███████▋  | 50655/65536 [8:43:44<2:39:14,  1.56it/s]
 77%|███████▋  | 50656/65536 [8:43:44<2:37:04,  1.58it/s]
 77%|███████▋  | 50657/65536 [8:43:45<2:33:52,  1.61it/s]
 77%|███████▋  | 50658/65536 [8:43:45<2:34:39,  1.60it/s]
 77%|███████▋  | 50659/65536 [8:43:46<2:36:12,  1.59it/s]
 77%|███████▋  | 50660/65536 [8:43:47<2:38:42,  1.56it/s]
                                                         
{'loss': 1.5672, 'learning_rate': 3.049161258724134e-07, 'epoch': 3127.16}
+
 77%|███████▋  | 50660/65536 [8:43:47<2:38:42,  1.56it/s]
 77%|███████▋  | 50661/65536 [8:43:47<2:32:16,  1.63it/s]
 77%|███████▋  | 50662/65536 [8:43:48<2:31:44,  1.63it/s]
 77%|███████▋  | 50663/65536 [8:43:48<2:30:11,  1.65it/s]
 77%|███████▋  | 50664/65536 [8:43:49<2:35:06,  1.60it/s]
 77%|███████▋  | 50665/65536 [8:43:50<2:32:45,  1.62it/s]
 77%|███████▋  | 50666/65536 [8:43:50<2:31:57,  1.63it/s]
 77%|███████▋  | 50667/65536 [8:43:51<2:30:05,  1.65it/s]
 77%|███████▋  | 50668/65536 [8:43:51<2:32:22,  1.63it/s]
 77%|███████▋  | 50669/65536 [8:43:52<2:32:44,  1.62it/s]
 77%|███████▋  | 50670/65536 [8:43:53<2:36:49,  1.58it/s]
 77%|███████▋  | 50671/65536 [8:43:53<2:36:17,  1.59it/s]
 77%|███████▋  | 50672/65536 [8:43:54<2:36:35,  1.58it/s]
 77%|███████▋  | 50673/65536 [8:43:55<2:35:14,  1.60it/s]
 77%|███████▋  | 50674/65536 [8:43:55<2:38:51,  1.56it/s]
 77%|███████▋  | 50675/65536 [8:43:56<2:40:16,  1.55it/s]
 77%|███████▋  | 50676/65536 [8:43:57<2:41:16,  1.54it/s]
 77%|███████▋  | 50677/65536 [8:43:57<2:36:26,  1.58it/s]
 77%|███████▋  | 50678/65536 [8:43:58<2:35:32,  1.59it/s]
 77%|███████▋  | 50679/65536 [8:43:58<2:31:06,  1.64it/s]
 77%|███████▋  | 50680/65536 [8:43:59<2:29:52,  1.65it/s]
                                                         
{'loss': 1.5756, 'learning_rate': 3.046406269131872e-07, 'epoch': 3128.4}
+
 77%|███████▋  | 50680/65536 [8:43:59<2:29:52,  1.65it/s]
 77%|███████▋  | 50681/65536 [8:44:00<2:35:44,  1.59it/s]
 77%|███████▋  | 50682/65536 [8:44:00<2:40:51,  1.54it/s]
 77%|███████▋  | 50683/65536 [8:44:01<2:38:43,  1.56it/s]
 77%|███████▋  | 50684/65536 [8:44:02<2:34:54,  1.60it/s]
 77%|███████▋  | 50685/65536 [8:44:02<2:34:06,  1.61it/s]
 77%|███████▋  | 50686/65536 [8:44:03<2:29:01,  1.66it/s]
 77%|███████▋  | 50687/65536 [8:44:03<2:25:18,  1.70it/s]
 77%|███████▋  | 50688/65536 [8:44:04<2:29:35,  1.65it/s]
 77%|███████▋  | 50689/65536 [8:44:05<2:25:58,  1.70it/s]
 77%|███████▋  | 50690/65536 [8:44:05<2:32:26,  1.62it/s]
 77%|███████▋  | 50691/65536 [8:44:06<2:30:02,  1.65it/s]
 77%|███████▋  | 50692/65536 [8:44:06<2:28:50,  1.66it/s]
 77%|███████▋  | 50693/65536 [8:44:07<2:29:11,  1.66it/s]
 77%|███████▋  | 50694/65536 [8:44:08<2:32:14,  1.62it/s]
 77%|███████▋  | 50695/65536 [8:44:08<2:34:56,  1.60it/s]
 77%|███████▋  | 50696/65536 [8:44:09<2:33:08,  1.62it/s]
 77%|███████▋  | 50697/65536 [8:44:09<2:30:54,  1.64it/s]
 77%|███████▋  | 50698/65536 [8:44:10<2:29:25,  1.66it/s]
 77%|███████▋  | 50699/65536 [8:44:11<2:28:29,  1.67it/s]
 77%|███████▋  | 50700/65536 [8:44:11<2:32:58,  1.62it/s]
                                                         
{'loss': 1.5722, 'learning_rate': 3.0436512795396103e-07, 'epoch': 3129.63}
+
 77%|███████▋  | 50700/65536 [8:44:11<2:32:58,  1.62it/s]
 77%|███████▋  | 50701/65536 [8:44:12<2:33:31,  1.61it/s]
 77%|███████▋  | 50702/65536 [8:44:13<2:36:15,  1.58it/s]
 77%|███████▋  | 50703/65536 [8:44:13<2:32:54,  1.62it/s]
 77%|███████▋  | 50704/65536 [8:44:14<2:32:56,  1.62it/s]
 77%|███████▋  | 50705/65536 [8:44:14<2:33:18,  1.61it/s]
 77%|███████▋  | 50706/65536 [8:44:15<2:35:40,  1.59it/s]
 77%|███████▋  | 50707/65536 [8:44:16<2:37:11,  1.57it/s]
 77%|███████▋  | 50708/65536 [8:44:16<2:36:46,  1.58it/s]
 77%|███████▋  | 50709/65536 [8:44:17<2:38:49,  1.56it/s]
 77%|███████▋  | 50710/65536 [8:44:18<2:35:15,  1.59it/s]
 77%|███████▋  | 50711/65536 [8:44:18<2:33:16,  1.61it/s]
 77%|███████▋  | 50712/65536 [8:44:19<2:34:20,  1.60it/s]
 77%|███████▋  | 50713/65536 [8:44:19<2:33:32,  1.61it/s]
 77%|███████▋  | 50714/65536 [8:44:20<2:35:03,  1.59it/s]
 77%|███████▋  | 50715/65536 [8:44:21<2:34:22,  1.60it/s]
 77%|███████▋  | 50716/65536 [8:44:21<2:32:30,  1.62it/s]
 77%|███████▋  | 50717/65536 [8:44:22<2:29:55,  1.65it/s]
 77%|███████▋  | 50718/65536 [8:44:23<2:33:33,  1.61it/s]
 77%|███████▋  | 50719/65536 [8:44:23<2:30:59,  1.64it/s]
 77%|███████▋  | 50720/65536 [8:44:24<2:29:58,  1.65it/s]
                                                         
{'loss': 1.557, 'learning_rate': 3.040896289947349e-07, 'epoch': 3130.86}
+
 77%|███████▋  | 50720/65536 [8:44:24<2:29:58,  1.65it/s]
 77%|███████▋  | 50721/65536 [8:44:24<2:31:40,  1.63it/s]
 77%|███████▋  | 50722/65536 [8:44:25<2:31:21,  1.63it/s]
 77%|███████▋  | 50723/65536 [8:44:26<2:38:14,  1.56it/s]
 77%|███████▋  | 50724/65536 [8:44:26<2:39:05,  1.55it/s]
 77%|███████▋  | 50725/65536 [8:44:27<2:34:39,  1.60it/s]
 77%|███████▋  | 50726/65536 [8:44:28<2:38:54,  1.55it/s]
 77%|███████▋  | 50727/65536 [8:44:28<2:38:37,  1.56it/s]
 77%|███████▋  | 50728/65536 [8:44:29<2:37:58,  1.56it/s]
 77%|███████▋  | 50729/65536 [8:44:30<2:34:25,  1.60it/s]
 77%|███████▋  | 50730/65536 [8:44:30<2:34:41,  1.60it/s]
 77%|███████▋  | 50731/65536 [8:44:31<2:30:31,  1.64it/s]
 77%|███████▋  | 50732/65536 [8:44:31<2:28:07,  1.67it/s]
 77%|███████▋  | 50733/65536 [8:44:32<2:30:50,  1.64it/s]
 77%|███████▋  | 50734/65536 [8:44:33<2:31:06,  1.63it/s]
 77%|███████▋  | 50735/65536 [8:44:33<2:30:41,  1.64it/s]
 77%|███████▋  | 50736/65536 [8:44:34<2:30:24,  1.64it/s]
 77%|███████▋  | 50737/65536 [8:44:34<2:30:45,  1.64it/s]
 77%|███████▋  | 50738/65536 [8:44:35<2:29:47,  1.65it/s]
 77%|███████▋  | 50739/65536 [8:44:36<2:39:02,  1.55it/s]
 77%|███████▋  | 50740/65536 [8:44:36<2:34:21,  1.60it/s]
                                                         
{'loss': 1.5584, 'learning_rate': 3.0381413003550875e-07, 'epoch': 3132.1}
+
 77%|███████▋  | 50740/65536 [8:44:36<2:34:21,  1.60it/s]
 77%|███████▋  | 50741/65536 [8:44:37<2:33:11,  1.61it/s]
 77%|███████▋  | 50742/65536 [8:44:37<2:32:18,  1.62it/s]
 77%|███████▋  | 50743/65536 [8:44:38<2:29:25,  1.65it/s]
 77%|███████▋  | 50744/65536 [8:44:39<2:27:05,  1.68it/s]
 77%|███████▋  | 50745/65536 [8:44:39<2:30:35,  1.64it/s]
 77%|███████▋  | 50746/65536 [8:44:40<2:28:12,  1.66it/s]
 77%|███████▋  | 50747/65536 [8:44:40<2:28:11,  1.66it/s]
 77%|███████▋  | 50748/65536 [8:44:41<2:36:09,  1.58it/s]
 77%|███████▋  | 50749/65536 [8:44:42<2:32:48,  1.61it/s]
 77%|███████▋  | 50750/65536 [8:44:42<2:34:04,  1.60it/s]
 77%|███████▋  | 50751/65536 [8:44:43<2:34:10,  1.60it/s]
 77%|███████▋  | 50752/65536 [8:44:44<2:32:32,  1.62it/s]
 77%|███████▋  | 50753/65536 [8:44:44<2:31:02,  1.63it/s]
 77%|███████▋  | 50754/65536 [8:44:45<2:28:42,  1.66it/s]
 77%|███████▋  | 50755/65536 [8:44:46<2:34:39,  1.59it/s]
 77%|███████▋  | 50756/65536 [8:44:46<2:34:17,  1.60it/s]
 77%|███████▋  | 50757/65536 [8:44:47<2:30:47,  1.63it/s]
 77%|███████▋  | 50758/65536 [8:44:47<2:36:41,  1.57it/s]
 77%|███████▋  | 50759/65536 [8:44:48<2:29:48,  1.64it/s]
 77%|███████▋  | 50760/65536 [8:44:49<2:32:42,  1.61it/s]
                                                         
{'loss': 1.5748, 'learning_rate': 3.035386310762826e-07, 'epoch': 3133.33}
+
 77%|███████▋  | 50760/65536 [8:44:49<2:32:42,  1.61it/s]
 77%|███████▋  | 50761/65536 [8:44:49<2:29:17,  1.65it/s]
 77%|███████▋  | 50762/65536 [8:44:50<2:29:19,  1.65it/s]
 77%|███████▋  | 50763/65536 [8:44:50<2:29:48,  1.64it/s]
 77%|███████▋  | 50764/65536 [8:44:51<2:31:10,  1.63it/s]
 77%|███████▋  | 50765/65536 [8:44:52<2:30:34,  1.64it/s]
 77%|███████▋  | 50766/65536 [8:44:52<2:26:22,  1.68it/s]
 77%|███████▋  | 50767/65536 [8:44:53<2:26:43,  1.68it/s]
 77%|███████▋  | 50768/65536 [8:44:53<2:26:04,  1.69it/s]
 77%|███████▋  | 50769/65536 [8:44:54<2:29:29,  1.65it/s]
 77%|███████▋  | 50770/65536 [8:44:55<2:33:51,  1.60it/s]
 77%|███████▋  | 50771/65536 [8:44:55<2:38:56,  1.55it/s]
 77%|███████▋  | 50772/65536 [8:44:56<2:35:10,  1.59it/s]
 77%|███████▋  | 50773/65536 [8:44:57<2:32:44,  1.61it/s]
 77%|███████▋  | 50774/65536 [8:44:57<2:32:32,  1.61it/s]
 77%|███████▋  | 50775/65536 [8:44:58<2:34:43,  1.59it/s]
 77%|███████▋  | 50776/65536 [8:44:58<2:31:10,  1.63it/s]
 77%|███████▋  | 50777/65536 [8:44:59<2:30:53,  1.63it/s]
 77%|███████▋  | 50778/65536 [8:45:00<2:30:02,  1.64it/s]
 77%|███████▋  | 50779/65536 [8:45:00<2:29:42,  1.64it/s]
 77%|███████▋  | 50780/65536 [8:45:01<2:34:38,  1.59it/s]
                                                         
{'loss': 1.5931, 'learning_rate': 3.0326313211705647e-07, 'epoch': 3134.57}
+
 77%|███████▋  | 50780/65536 [8:45:01<2:34:38,  1.59it/s]
 77%|███████▋  | 50781/65536 [8:45:01<2:30:02,  1.64it/s]
 77%|███████▋  | 50782/65536 [8:45:02<2:28:31,  1.66it/s]
 77%|███████▋  | 50783/65536 [8:45:03<2:28:14,  1.66it/s]
 77%|███████▋  | 50784/65536 [8:45:03<2:31:25,  1.62it/s]
 77%|███████▋  | 50785/65536 [8:45:04<2:40:18,  1.53it/s]
 77%|███████▋  | 50786/65536 [8:45:05<2:35:11,  1.58it/s]
 77%|███████▋  | 50787/65536 [8:45:05<2:29:47,  1.64it/s]
 77%|███████▋  | 50788/65536 [8:45:06<2:36:21,  1.57it/s]
 77%|███████▋  | 50789/65536 [8:45:07<2:37:02,  1.57it/s]
 77%|███████▋  | 50790/65536 [8:45:07<2:35:51,  1.58it/s]
 78%|███████▊  | 50791/65536 [8:45:08<2:36:16,  1.57it/s]
 78%|███████▊  | 50792/65536 [8:45:08<2:32:47,  1.61it/s]
 78%|███████▊  | 50793/65536 [8:45:09<2:31:45,  1.62it/s]
 78%|███████▊  | 50794/65536 [8:45:10<2:37:55,  1.56it/s]
 78%|███████▊  | 50795/65536 [8:45:10<2:30:42,  1.63it/s]
 78%|███████▊  | 50796/65536 [8:45:11<2:30:18,  1.63it/s]
 78%|███████▊  | 50797/65536 [8:45:11<2:31:17,  1.62it/s]
 78%|███████▊  | 50798/65536 [8:45:12<2:30:43,  1.63it/s]
 78%|███████▊  | 50799/65536 [8:45:13<2:31:13,  1.62it/s]
 78%|███████▊  | 50800/65536 [8:45:13<2:30:26,  1.63it/s]
                                                         
{'loss': 1.5797, 'learning_rate': 3.0298763315783033e-07, 'epoch': 3135.8}
+
 78%|███████▊  | 50800/65536 [8:45:13<2:30:26,  1.63it/s]
 78%|███████▊  | 50801/65536 [8:45:14<2:28:36,  1.65it/s]
 78%|███████▊  | 50802/65536 [8:45:14<2:26:38,  1.67it/s]
 78%|███████▊  | 50803/65536 [8:45:15<2:28:08,  1.66it/s]
 78%|███████▊  | 50804/65536 [8:45:16<2:33:11,  1.60it/s]
 78%|███████▊  | 50805/65536 [8:45:16<2:33:47,  1.60it/s]
 78%|███████▊  | 50806/65536 [8:45:17<2:33:07,  1.60it/s]
 78%|███████▊  | 50807/65536 [8:45:18<2:34:52,  1.59it/s]
 78%|███████▊  | 50808/65536 [8:45:18<2:36:31,  1.57it/s]
 78%|███████▊  | 50809/65536 [8:45:19<2:34:08,  1.59it/s]
 78%|███████▊  | 50810/65536 [8:45:19<2:30:46,  1.63it/s]
 78%|███████▊  | 50811/65536 [8:45:20<2:33:41,  1.60it/s]
 78%|███████▊  | 50812/65536 [8:45:21<2:33:26,  1.60it/s]
 78%|███████▊  | 50813/65536 [8:45:21<2:30:58,  1.63it/s]
 78%|███████▊  | 50814/65536 [8:45:22<2:26:52,  1.67it/s]
 78%|███████▊  | 50815/65536 [8:45:23<2:27:49,  1.66it/s]
 78%|███████▊  | 50816/65536 [8:45:23<2:28:55,  1.65it/s]
 78%|███████▊  | 50817/65536 [8:45:24<2:29:16,  1.64it/s]
 78%|███████▊  | 50818/65536 [8:45:24<2:31:57,  1.61it/s]
 78%|███████▊  | 50819/65536 [8:45:25<2:32:40,  1.61it/s]
 78%|███████▊  | 50820/65536 [8:45:26<2:36:08,  1.57it/s]
                                                         
{'loss': 1.5638, 'learning_rate': 3.027121341986041e-07, 'epoch': 3137.04}
+
 78%|███████▊  | 50820/65536 [8:45:26<2:36:08,  1.57it/s]
 78%|███████▊  | 50821/65536 [8:45:26<2:33:40,  1.60it/s]
 78%|███████▊  | 50822/65536 [8:45:27<2:33:51,  1.59it/s]
 78%|███████▊  | 50823/65536 [8:45:28<2:30:43,  1.63it/s]
 78%|███████▊  | 50824/65536 [8:45:28<2:38:51,  1.54it/s]
 78%|███████▊  | 50825/65536 [8:45:29<2:37:20,  1.56it/s]
 78%|███████▊  | 50826/65536 [8:45:29<2:35:18,  1.58it/s]
 78%|███████▊  | 50827/65536 [8:45:30<2:30:55,  1.62it/s]
 78%|███████▊  | 50828/65536 [8:45:31<2:34:24,  1.59it/s]
 78%|███████▊  | 50829/65536 [8:45:31<2:33:37,  1.60it/s]
 78%|███████▊  | 50830/65536 [8:45:32<2:29:26,  1.64it/s]
 78%|███████▊  | 50831/65536 [8:45:33<2:31:18,  1.62it/s]
 78%|███████▊  | 50832/65536 [8:45:33<2:33:46,  1.59it/s]
 78%|███████▊  | 50833/65536 [8:45:34<2:31:23,  1.62it/s]
 78%|███████▊  | 50834/65536 [8:45:34<2:29:03,  1.64it/s]
 78%|███████▊  | 50835/65536 [8:45:35<2:25:18,  1.69it/s]
 78%|███████▊  | 50836/65536 [8:45:36<2:35:06,  1.58it/s]
 78%|███████▊  | 50837/65536 [8:45:36<2:38:25,  1.55it/s]
 78%|███████▊  | 50838/65536 [8:45:37<2:36:19,  1.57it/s]
 78%|███████▊  | 50839/65536 [8:45:38<2:35:07,  1.58it/s]
 78%|███████▊  | 50840/65536 [8:45:38<2:31:37,  1.62it/s]
                                                         
{'loss': 1.5431, 'learning_rate': 3.0243663523937794e-07, 'epoch': 3138.27}
+
 78%|███████▊  | 50840/65536 [8:45:38<2:31:37,  1.62it/s]
 78%|███████▊  | 50841/65536 [8:45:39<2:31:52,  1.61it/s]
 78%|███████▊  | 50842/65536 [8:45:39<2:31:26,  1.62it/s]
 78%|███████▊  | 50843/65536 [8:45:40<2:36:35,  1.56it/s]
 78%|███████▊  | 50844/65536 [8:45:41<2:34:44,  1.58it/s]
 78%|███████▊  | 50845/65536 [8:45:41<2:31:38,  1.61it/s]
 78%|███████▊  | 50846/65536 [8:45:42<2:31:24,  1.62it/s]
 78%|███████▊  | 50847/65536 [8:45:43<2:32:40,  1.60it/s]
 78%|███████▊  | 50848/65536 [8:45:43<2:29:37,  1.64it/s]
 78%|███████▊  | 50849/65536 [8:45:44<2:28:04,  1.65it/s]
 78%|███████▊  | 50850/65536 [8:45:44<2:33:16,  1.60it/s]
 78%|███████▊  | 50851/65536 [8:45:45<2:29:30,  1.64it/s]
 78%|███████▊  | 50852/65536 [8:45:46<2:37:09,  1.56it/s]
 78%|███████▊  | 50853/65536 [8:45:46<2:32:05,  1.61it/s]
 78%|███████▊  | 50854/65536 [8:45:47<2:31:16,  1.62it/s]
 78%|███████▊  | 50855/65536 [8:45:48<2:32:42,  1.60it/s]
 78%|███████▊  | 50856/65536 [8:45:48<2:31:56,  1.61it/s]
 78%|��██████▊  | 50857/65536 [8:45:49<2:30:08,  1.63it/s]
 78%|███████▊  | 50858/65536 [8:45:49<2:31:28,  1.62it/s]
 78%|███████▊  | 50859/65536 [8:45:50<2:29:15,  1.64it/s]
 78%|███████▊  | 50860/65536 [8:45:51<2:27:38,  1.66it/s]
                                                         
{'loss': 1.5679, 'learning_rate': 3.021611362801518e-07, 'epoch': 3139.51}
+
 78%|███████▊  | 50860/65536 [8:45:51<2:27:38,  1.66it/s]
 78%|███████▊  | 50861/65536 [8:45:51<2:30:41,  1.62it/s]
 78%|███████▊  | 50862/65536 [8:45:52<2:33:20,  1.59it/s]
 78%|███████▊  | 50863/65536 [8:45:52<2:35:42,  1.57it/s]
 78%|███████▊  | 50864/65536 [8:45:53<2:32:33,  1.60it/s]
 78%|███████▊  | 50865/65536 [8:45:54<2:34:13,  1.59it/s]
 78%|███████▊  | 50866/65536 [8:45:54<2:30:54,  1.62it/s]
 78%|███████▊  | 50867/65536 [8:45:55<2:29:48,  1.63it/s]
 78%|███████▊  | 50868/65536 [8:45:55<2:27:31,  1.66it/s]
 78%|███████▊  | 50869/65536 [8:45:56<2:31:40,  1.61it/s]
 78%|███████▊  | 50870/65536 [8:45:57<2:30:50,  1.62it/s]
 78%|███████▊  | 50871/65536 [8:45:57<2:36:18,  1.56it/s]
 78%|███████▊  | 50872/65536 [8:45:58<2:32:12,  1.61it/s]
 78%|███████▊  | 50873/65536 [8:45:59<2:35:32,  1.57it/s]
 78%|███████▊  | 50874/65536 [8:45:59<2:31:31,  1.61it/s]
 78%|███████▊  | 50875/65536 [8:46:00<2:28:59,  1.64it/s]
 78%|███████▊  | 50876/65536 [8:46:00<2:29:08,  1.64it/s]
 78%|███████▊  | 50877/65536 [8:46:01<2:29:12,  1.64it/s]
 78%|███████▊  | 50878/65536 [8:46:02<2:29:12,  1.64it/s]
 78%|███████▊  | 50879/65536 [8:46:02<2:27:37,  1.65it/s]
 78%|███████▊  | 50880/65536 [8:46:03<2:30:00,  1.63it/s]
                                                         
{'loss': 1.5609, 'learning_rate': 3.0188563732092566e-07, 'epoch': 3140.74}
+
 78%|███████▊  | 50880/65536 [8:46:03<2:30:00,  1.63it/s]
 78%|███████▊  | 50881/65536 [8:46:04<2:30:10,  1.63it/s]
 78%|███████▊  | 50882/65536 [8:46:04<2:28:33,  1.64it/s]
 78%|███████▊  | 50883/65536 [8:46:05<2:27:15,  1.66it/s]
 78%|███████▊  | 50884/65536 [8:46:05<2:26:27,  1.67it/s]
 78%|███████▊  | 50885/65536 [8:46:06<2:37:32,  1.55it/s]
 78%|███████▊  | 50886/65536 [8:46:07<2:34:22,  1.58it/s]
 78%|███████▊  | 50887/65536 [8:46:07<2:32:37,  1.60it/s]
 78%|███████▊  | 50888/65536 [8:46:08<2:32:36,  1.60it/s]
 78%|███████▊  | 50889/65536 [8:46:09<2:30:10,  1.63it/s]
 78%|███████▊  | 50890/65536 [8:46:09<2:28:37,  1.64it/s]
 78%|███████▊  | 50891/65536 [8:46:10<2:33:13,  1.59it/s]
 78%|███████▊  | 50892/65536 [8:46:10<2:32:28,  1.60it/s]
 78%|███████▊  | 50893/65536 [8:46:11<2:33:17,  1.59it/s]
 78%|███████▊  | 50894/65536 [8:46:12<2:37:41,  1.55it/s]
 78%|███████▊  | 50895/65536 [8:46:12<2:33:34,  1.59it/s]
 78%|███████▊  | 50896/65536 [8:46:13<2:31:01,  1.62it/s]
 78%|███████▊  | 50897/65536 [8:46:13<2:28:16,  1.65it/s]
 78%|███████▊  | 50898/65536 [8:46:14<2:29:48,  1.63it/s]
 78%|███████▊  | 50899/65536 [8:46:15<2:29:57,  1.63it/s]
 78%|███████▊  | 50900/65536 [8:46:15<2:27:53,  1.65it/s]
                                                         
{'loss': 1.5893, 'learning_rate': 3.016101383616995e-07, 'epoch': 3141.98}
+
 78%|███████▊  | 50900/65536 [8:46:15<2:27:53,  1.65it/s]
 78%|███████▊  | 50901/65536 [8:46:16<2:37:19,  1.55it/s]
 78%|███████▊  | 50902/65536 [8:46:17<2:36:01,  1.56it/s]
 78%|███████▊  | 50903/65536 [8:46:17<2:32:57,  1.59it/s]
 78%|███████▊  | 50904/65536 [8:46:18<2:30:24,  1.62it/s]
 78%|███████▊  | 50905/65536 [8:46:18<2:30:23,  1.62it/s]
 78%|███████▊  | 50906/65536 [8:46:19<2:29:40,  1.63it/s]
 78%|███████▊  | 50907/65536 [8:46:20<2:32:35,  1.60it/s]
 78%|███████▊  | 50908/65536 [8:46:20<2:32:56,  1.59it/s]
 78%|███████▊  | 50909/65536 [8:46:21<2:34:01,  1.58it/s]
 78%|███████▊  | 50910/65536 [8:46:22<2:31:30,  1.61it/s]
 78%|███████▊  | 50911/65536 [8:46:22<2:28:02,  1.65it/s]
 78%|███████▊  | 50912/65536 [8:46:23<2:26:49,  1.66it/s]
 78%|███████▊  | 50913/65536 [8:46:23<2:26:56,  1.66it/s]
 78%|███████▊  | 50914/65536 [8:46:24<2:27:25,  1.65it/s]
 78%|███████▊  | 50915/65536 [8:46:25<2:33:21,  1.59it/s]
 78%|███████▊  | 50916/65536 [8:46:25<2:30:40,  1.62it/s]
 78%|███████▊  | 50917/65536 [8:46:26<2:35:29,  1.57it/s]
 78%|███████▊  | 50918/65536 [8:46:27<2:37:11,  1.55it/s]
 78%|███████▊  | 50919/65536 [8:46:27<2:32:18,  1.60it/s]
 78%|███████▊  | 50920/65536 [8:46:28<2:29:15,  1.63it/s]
                                                         
{'loss': 1.5805, 'learning_rate': 3.013346394024734e-07, 'epoch': 3143.21}
+
 78%|███████▊  | 50920/65536 [8:46:28<2:29:15,  1.63it/s]
 78%|███████▊  | 50921/65536 [8:46:28<2:32:43,  1.59it/s]
 78%|███████▊  | 50922/65536 [8:46:29<2:31:11,  1.61it/s]
 78%|███████▊  | 50923/65536 [8:46:30<2:29:13,  1.63it/s]
 78%|███████▊  | 50924/65536 [8:46:30<2:30:09,  1.62it/s]
 78%|███████▊  | 50925/65536 [8:46:31<2:28:00,  1.65it/s]
 78%|███████▊  | 50926/65536 [8:46:32<2:31:36,  1.61it/s]
 78%|███████▊  | 50927/65536 [8:46:32<2:29:18,  1.63it/s]
 78%|███████▊  | 50928/65536 [8:46:33<2:29:23,  1.63it/s]
 78%|███████▊  | 50929/65536 [8:46:33<2:28:53,  1.63it/s]
 78%|███████▊  | 50930/65536 [8:46:34<2:29:32,  1.63it/s]
 78%|███████▊  | 50931/65536 [8:46:35<2:27:38,  1.65it/s]
 78%|███████▊  | 50932/65536 [8:46:35<2:30:00,  1.62it/s]
 78%|███████▊  | 50933/65536 [8:46:36<2:32:17,  1.60it/s]
 78%|███████▊  | 50934/65536 [8:46:36<2:33:27,  1.59it/s]
 78%|███████▊  | 50935/65536 [8:46:37<2:29:50,  1.62it/s]
 78%|███████▊  | 50936/65536 [8:46:38<2:32:34,  1.59it/s]
 78%|███████▊  | 50937/65536 [8:46:38<2:29:23,  1.63it/s]
 78%|███████▊  | 50938/65536 [8:46:39<2:27:35,  1.65it/s]
 78%|███████▊  | 50939/65536 [8:46:40<2:30:29,  1.62it/s]
 78%|███████▊  | 50940/65536 [8:46:40<2:26:45,  1.66it/s]
                                                         
{'loss': 1.5756, 'learning_rate': 3.010591404432472e-07, 'epoch': 3144.44}
+
 78%|███████▊  | 50940/65536 [8:46:40<2:26:45,  1.66it/s]
 78%|███████▊  | 50941/65536 [8:46:41<2:27:13,  1.65it/s]
 78%|███████▊  | 50942/65536 [8:46:41<2:28:34,  1.64it/s]
 78%|███████▊  | 50943/65536 [8:46:42<2:28:03,  1.64it/s]
 78%|███████▊  | 50944/65536 [8:46:43<2:29:33,  1.63it/s]
 78%|███████▊  | 50945/65536 [8:46:43<2:30:23,  1.62it/s]
 78%|███████▊  | 50946/65536 [8:46:44<2:32:20,  1.60it/s]
 78%|███████▊  | 50947/65536 [8:46:44<2:33:31,  1.58it/s]
 78%|███████▊  | 50948/65536 [8:46:45<2:31:32,  1.60it/s]
 78%|███████▊  | 50949/65536 [8:46:46<2:31:35,  1.60it/s]
 78%|███████▊  | 50950/65536 [8:46:46<2:35:20,  1.56it/s]
 78%|███████▊  | 50951/65536 [8:46:47<2:34:11,  1.58it/s]
 78%|███████▊  | 50952/65536 [8:46:48<2:34:12,  1.58it/s]
 78%|███████▊  | 50953/65536 [8:46:48<2:32:19,  1.60it/s]
 78%|███████▊  | 50954/65536 [8:46:49<2:34:18,  1.58it/s]
 78%|███████▊  | 50955/65536 [8:46:49<2:31:37,  1.60it/s]
 78%|███████▊  | 50956/65536 [8:46:50<2:31:01,  1.61it/s]
 78%|███████▊  | 50957/65536 [8:46:51<2:35:11,  1.57it/s]
 78%|███████▊  | 50958/65536 [8:46:51<2:32:30,  1.59it/s]
 78%|███████▊  | 50959/65536 [8:46:52<2:30:53,  1.61it/s]
 78%|███████▊  | 50960/65536 [8:46:53<2:30:07,  1.62it/s]
                                                         
{'loss': 1.5364, 'learning_rate': 3.00783641484021e-07, 'epoch': 3145.68}
+
 78%|███████▊  | 50960/65536 [8:46:53<2:30:07,  1.62it/s]
 78%|███████▊  | 50961/65536 [8:46:53<2:30:01,  1.62it/s]
 78%|███████▊  | 50962/65536 [8:46:54<2:32:27,  1.59it/s]
 78%|███████▊  | 50963/65536 [8:46:54<2:27:50,  1.64it/s]
 78%|███████▊  | 50964/65536 [8:46:55<2:28:10,  1.64it/s]
 78%|███████▊  | 50965/65536 [8:46:56<2:26:34,  1.66it/s]
 78%|███████▊  | 50966/65536 [8:46:56<2:32:51,  1.59it/s]
 78%|███████▊  | 50967/65536 [8:46:57<2:29:50,  1.62it/s]
 78%|███████▊  | 50968/65536 [8:46:58<2:29:19,  1.63it/s]
 78%|███████▊  | 50969/65536 [8:46:58<2:26:43,  1.65it/s]
 78%|███████▊  | 50970/65536 [8:46:59<2:27:22,  1.65it/s]
 78%|███████▊  | 50971/65536 [8:46:59<2:28:31,  1.63it/s]
 78%|███████▊  | 50972/65536 [8:47:00<2:26:10,  1.66it/s]
 78%|███████▊  | 50973/65536 [8:47:01<2:28:17,  1.64it/s]
 78%|███████▊  | 50974/65536 [8:47:01<2:29:15,  1.63it/s]
 78%|███████▊  | 50975/65536 [8:47:02<2:27:49,  1.64it/s]
 78%|███████▊  | 50976/65536 [8:47:02<2:28:48,  1.63it/s]
 78%|███████▊  | 50977/65536 [8:47:03<2:28:15,  1.64it/s]
 78%|███████▊  | 50978/65536 [8:47:04<2:31:01,  1.61it/s]
 78%|███████▊  | 50979/65536 [8:47:04<2:33:56,  1.58it/s]
 78%|███████▊  | 50980/65536 [8:47:05<2:35:52,  1.56it/s]
                                                         
{'loss': 1.5451, 'learning_rate': 3.0050814252479485e-07, 'epoch': 3146.91}
+
 78%|███████▊  | 50980/65536 [8:47:05<2:35:52,  1.56it/s]
 78%|███████▊  | 50981/65536 [8:47:06<2:35:30,  1.56it/s]
 78%|███████▊  | 50982/65536 [8:47:06<2:40:33,  1.51it/s]
 78%|███████▊  | 50983/65536 [8:47:07<2:37:54,  1.54it/s]
 78%|███████▊  | 50984/65536 [8:47:08<2:36:01,  1.55it/s]
 78%|███████▊  | 50985/65536 [8:47:08<2:32:07,  1.59it/s]
 78%|███████▊  | 50986/65536 [8:47:09<2:27:22,  1.65it/s]
 78%|███████▊  | 50987/65536 [8:47:09<2:28:25,  1.63it/s]
 78%|███████▊  | 50988/65536 [8:47:10<2:27:45,  1.64it/s]
 78%|███████▊  | 50989/65536 [8:47:11<2:25:32,  1.67it/s]
 78%|███████▊  | 50990/65536 [8:47:11<2:25:50,  1.66it/s]
 78%|███████▊  | 50991/65536 [8:47:12<2:27:11,  1.65it/s]
 78%|███████▊  | 50992/65536 [8:47:12<2:26:47,  1.65it/s]
 78%|███████▊  | 50993/65536 [8:47:13<2:26:05,  1.66it/s]
 78%|███████▊  | 50994/65536 [8:47:14<2:26:42,  1.65it/s]
 78%|███████▊  | 50995/65536 [8:47:14<2:30:39,  1.61it/s]
 78%|███████▊  | 50996/65536 [8:47:15<2:29:22,  1.62it/s]
 78%|███████▊  | 50997/65536 [8:47:16<2:33:24,  1.58it/s]
 78%|███████▊  | 50998/65536 [8:47:16<2:34:46,  1.57it/s]
 78%|███████▊  | 50999/65536 [8:47:17<2:32:22,  1.59it/s]
 78%|███████▊  | 51000/65536 [8:47:17<2:34:58,  1.56it/s]
                                                         
{'loss': 1.5657, 'learning_rate': 3.002326435655687e-07, 'epoch': 3148.15}
+
 78%|███████▊  | 51000/65536 [8:47:17<2:34:58,  1.56it/s]
 78%|███████▊  | 51001/65536 [8:47:18<2:33:57,  1.57it/s]
 78%|███████▊  | 51002/65536 [8:47:19<2:29:17,  1.62it/s]
 78%|███████▊  | 51003/65536 [8:47:19<2:28:23,  1.63it/s]
 78%|███████▊  | 51004/65536 [8:47:20<2:29:46,  1.62it/s]
 78%|███████▊  | 51005/65536 [8:47:21<2:33:43,  1.58it/s]
 78%|███████▊  | 51006/65536 [8:47:21<2:35:12,  1.56it/s]
 78%|███████▊  | 51007/65536 [8:47:22<2:38:48,  1.52it/s]
 78%|███████▊  | 51008/65536 [8:47:23<2:41:51,  1.50it/s]
 78%|███████▊  | 51009/65536 [8:47:23<2:40:56,  1.50it/s]
 78%|███████▊  | 51010/65536 [8:47:24<2:36:31,  1.55it/s]
 78%|███████▊  | 51011/65536 [8:47:24<2:33:55,  1.57it/s]
 78%|███████▊  | 51012/65536 [8:47:25<2:31:05,  1.60it/s]
 78%|███████▊  | 51013/65536 [8:47:26<2:33:56,  1.57it/s]
 78%|███████▊  | 51014/65536 [8:47:26<2:34:42,  1.56it/s]
 78%|███████▊  | 51015/65536 [8:47:27<2:32:43,  1.58it/s]
 78%|███████▊  | 51016/65536 [8:47:28<2:31:00,  1.60it/s]
 78%|███████▊  | 51017/65536 [8:47:28<2:33:33,  1.58it/s]
 78%|███████▊  | 51018/65536 [8:47:29<2:29:51,  1.61it/s]
 78%|███████▊  | 51019/65536 [8:47:29<2:25:17,  1.67it/s]
 78%|███████▊  | 51020/65536 [8:47:30<2:31:27,  1.60it/s]
                                                         
{'loss': 1.5771, 'learning_rate': 2.9995714460634256e-07, 'epoch': 3149.38}
+
 78%|███████▊  | 51020/65536 [8:47:30<2:31:27,  1.60it/s]
 78%|███████▊  | 51021/65536 [8:47:31<2:32:55,  1.58it/s]
 78%|███████▊  | 51022/65536 [8:47:31<2:30:07,  1.61it/s]
 78%|███████▊  | 51023/65536 [8:47:32<2:35:21,  1.56it/s]
 78%|███████▊  | 51024/65536 [8:47:33<2:30:36,  1.61it/s]
 78%|███████▊  | 51025/65536 [8:47:33<2:28:52,  1.62it/s]
 78%|███████▊  | 51026/65536 [8:47:34<2:28:37,  1.63it/s]
 78%|███████▊  | 51027/65536 [8:47:34<2:31:48,  1.59it/s]
 78%|███████▊  | 51028/65536 [8:47:35<2:30:42,  1.60it/s]
 78%|███████▊  | 51029/65536 [8:47:36<2:28:28,  1.63it/s]
 78%|███████▊  | 51030/65536 [8:47:36<2:28:47,  1.62it/s]
 78%|███████▊  | 51031/65536 [8:47:37<2:33:21,  1.58it/s]
 78%|███████▊  | 51032/65536 [8:47:38<2:30:34,  1.61it/s]
 78%|███████▊  | 51033/65536 [8:47:38<2:30:00,  1.61it/s]
 78%|███████▊  | 51034/65536 [8:47:39<2:38:19,  1.53it/s]
 78%|███████▊  | 51035/65536 [8:47:40<2:36:06,  1.55it/s]
 78%|███████▊  | 51036/65536 [8:47:40<2:35:42,  1.55it/s]
 78%|███████▊  | 51037/65536 [8:47:41<2:31:53,  1.59it/s]
 78%|███████▊  | 51038/65536 [8:47:41<2:26:37,  1.65it/s]
 78%|███��███▊  | 51039/65536 [8:47:42<2:26:05,  1.65it/s]
 78%|███████▊  | 51040/65536 [8:47:43<2:26:29,  1.65it/s]
                                                         
{'loss': 1.5572, 'learning_rate': 2.996816456471164e-07, 'epoch': 3150.62}
+
 78%|███████▊  | 51040/65536 [8:47:43<2:26:29,  1.65it/s]
 78%|███████▊  | 51041/65536 [8:47:43<2:28:47,  1.62it/s]
 78%|███████▊  | 51042/65536 [8:47:44<2:27:15,  1.64it/s]
 78%|███████▊  | 51043/65536 [8:47:44<2:25:19,  1.66it/s]
 78%|███████▊  | 51044/65536 [8:47:45<2:28:12,  1.63it/s]
 78%|███████▊  | 51045/65536 [8:47:46<2:30:17,  1.61it/s]
 78%|███████▊  | 51046/65536 [8:47:46<2:27:03,  1.64it/s]
 78%|███████▊  | 51047/65536 [8:47:47<2:33:00,  1.58it/s]
 78%|███████▊  | 51048/65536 [8:47:47<2:30:40,  1.60it/s]
 78%|███████▊  | 51049/65536 [8:47:48<2:30:17,  1.61it/s]
 78%|███████▊  | 51050/65536 [8:47:49<2:29:10,  1.62it/s]
 78%|███████▊  | 51051/65536 [8:47:49<2:26:57,  1.64it/s]
 78%|███████▊  | 51052/65536 [8:47:50<2:27:45,  1.63it/s]
 78%|███████▊  | 51053/65536 [8:47:51<2:27:21,  1.64it/s]
 78%|███████▊  | 51054/65536 [8:47:51<2:26:28,  1.65it/s]
 78%|███████▊  | 51055/65536 [8:47:52<2:30:50,  1.60it/s]
 78%|███████▊  | 51056/65536 [8:47:52<2:30:42,  1.60it/s]
 78%|███████▊  | 51057/65536 [8:47:53<2:33:42,  1.57it/s]
 78%|███████▊  | 51058/65536 [8:47:54<2:35:43,  1.55it/s]
 78%|███████▊  | 51059/65536 [8:47:54<2:33:41,  1.57it/s]
 78%|███████▊  | 51060/65536 [8:47:55<2:35:43,  1.55it/s]
                                                         
{'loss': 1.5297, 'learning_rate': 2.994061466878903e-07, 'epoch': 3151.85}
+
 78%|███████▊  | 51060/65536 [8:47:55<2:35:43,  1.55it/s]
 78%|███████▊  | 51061/65536 [8:47:56<2:33:41,  1.57it/s]
 78%|███████▊  | 51062/65536 [8:47:56<2:29:22,  1.62it/s]
 78%|███████▊  | 51063/65536 [8:47:57<2:37:11,  1.53it/s]
 78%|███████▊  | 51064/65536 [8:47:58<2:38:29,  1.52it/s]
 78%|███████▊  | 51065/65536 [8:47:58<2:32:48,  1.58it/s]
 78%|███████▊  | 51066/65536 [8:47:59<2:32:16,  1.58it/s]
 78%|███████▊  | 51067/65536 [8:47:59<2:26:27,  1.65it/s]
 78%|███████▊  | 51068/65536 [8:48:00<2:30:16,  1.60it/s]
 78%|███████▊  | 51069/65536 [8:48:01<2:31:23,  1.59it/s]
 78%|███████▊  | 51070/65536 [8:48:01<2:29:33,  1.61it/s]
 78%|███████▊  | 51071/65536 [8:48:02<2:30:14,  1.60it/s]
 78%|███████▊  | 51072/65536 [8:48:03<2:32:09,  1.58it/s]
 78%|███████▊  | 51073/65536 [8:48:03<2:28:38,  1.62it/s]
 78%|███████▊  | 51074/65536 [8:48:04<2:28:09,  1.63it/s]
 78%|███████▊  | 51075/65536 [8:48:04<2:24:41,  1.67it/s]
 78%|███████▊  | 51076/65536 [8:48:05<2:25:51,  1.65it/s]
 78%|███████▊  | 51077/65536 [8:48:06<2:25:59,  1.65it/s]
 78%|███████▊  | 51078/65536 [8:48:06<2:25:50,  1.65it/s]
 78%|███████▊  | 51079/65536 [8:48:07<2:29:21,  1.61it/s]
 78%|███████▊  | 51080/65536 [8:48:07<2:29:03,  1.62it/s]
                                                         
{'loss': 1.6089, 'learning_rate': 2.991306477286641e-07, 'epoch': 3153.09}
+
 78%|███████▊  | 51080/65536 [8:48:07<2:29:03,  1.62it/s]
 78%|███████▊  | 51081/65536 [8:48:08<2:31:32,  1.59it/s]
 78%|███████▊  | 51082/65536 [8:48:09<2:33:26,  1.57it/s]
 78%|███████▊  | 51083/65536 [8:48:09<2:30:07,  1.60it/s]
 78%|███████▊  | 51084/65536 [8:48:10<2:28:22,  1.62it/s]
 78%|███████▊  | 51085/65536 [8:48:11<2:28:00,  1.63it/s]
 78%|███████▊  | 51086/65536 [8:48:11<2:27:45,  1.63it/s]
 78%|███████▊  | 51087/65536 [8:48:12<2:29:47,  1.61it/s]
 78%|███████▊  | 51088/65536 [8:48:12<2:31:38,  1.59it/s]
 78%|███████▊  | 51089/65536 [8:48:13<2:34:13,  1.56it/s]
 78%|███████▊  | 51090/65536 [8:48:14<2:29:22,  1.61it/s]
 78%|███████▊  | 51091/65536 [8:48:14<2:26:27,  1.64it/s]
 78%|███████▊  | 51092/65536 [8:48:15<2:25:52,  1.65it/s]
 78%|███████▊  | 51093/65536 [8:48:15<2:24:51,  1.66it/s]
 78%|███████▊  | 51094/65536 [8:48:16<2:27:05,  1.64it/s]
 78%|███████▊  | 51095/65536 [8:48:17<2:29:30,  1.61it/s]
 78%|███████▊  | 51096/65536 [8:48:17<2:29:07,  1.61it/s]
 78%|███████▊  | 51097/65536 [8:48:18<2:32:15,  1.58it/s]
 78%|███████▊  | 51098/65536 [8:48:19<2:31:05,  1.59it/s]
 78%|███████▊  | 51099/65536 [8:48:19<2:33:22,  1.57it/s]
 78%|███████▊  | 51100/65536 [8:48:20<2:29:21,  1.61it/s]
                                                         
{'loss': 1.5366, 'learning_rate': 2.9885514876943795e-07, 'epoch': 3154.32}
+
 78%|███████▊  | 51100/65536 [8:48:20<2:29:21,  1.61it/s]
 78%|███████▊  | 51101/65536 [8:48:21<2:32:20,  1.58it/s]
 78%|███████▊  | 51102/65536 [8:48:21<2:28:24,  1.62it/s]
 78%|███████▊  | 51103/65536 [8:48:22<2:31:15,  1.59it/s]
 78%|███████▊  | 51104/65536 [8:48:22<2:27:01,  1.64it/s]
 78%|███████▊  | 51105/65536 [8:48:23<2:26:39,  1.64it/s]
 78%|███████▊  | 51106/65536 [8:48:24<2:26:28,  1.64it/s]
 78%|███████▊  | 51107/65536 [8:48:24<2:25:59,  1.65it/s]
 78%|███████▊  | 51108/65536 [8:48:25<2:27:06,  1.63it/s]
 78%|███████▊  | 51109/65536 [8:48:25<2:28:03,  1.62it/s]
 78%|███████▊  | 51110/65536 [8:48:26<2:27:05,  1.63it/s]
 78%|███████▊  | 51111/65536 [8:48:27<2:24:21,  1.67it/s]
 78%|███████▊  | 51112/65536 [8:48:27<2:28:46,  1.62it/s]
 78%|███████▊  | 51113/65536 [8:48:28<2:29:17,  1.61it/s]
 78%|███████▊  | 51114/65536 [8:48:28<2:28:59,  1.61it/s]
 78%|███████▊  | 51115/65536 [8:48:29<2:26:12,  1.64it/s]
 78%|███████▊  | 51116/65536 [8:48:30<2:29:12,  1.61it/s]
 78%|███████▊  | 51117/65536 [8:48:30<2:30:23,  1.60it/s]
 78%|███████▊  | 51118/65536 [8:48:31<2:33:04,  1.57it/s]
 78%|███████▊  | 51119/65536 [8:48:32<2:32:38,  1.57it/s]
 78%|███████▊  | 51120/65536 [8:48:32<2:33:23,  1.57it/s]
                                                         
{'loss': 1.5776, 'learning_rate': 2.9857964981021186e-07, 'epoch': 3155.56}
+
 78%|███████▊  | 51120/65536 [8:48:32<2:33:23,  1.57it/s]
 78%|███████▊  | 51121/65536 [8:48:33<2:32:12,  1.58it/s]
 78%|███████▊  | 51122/65536 [8:48:34<2:31:41,  1.58it/s]
 78%|███████▊  | 51123/65536 [8:48:34<2:29:51,  1.60it/s]
 78%|███████▊  | 51124/65536 [8:48:35<2:29:59,  1.60it/s]
 78%|███████▊  | 51125/65536 [8:48:35<2:26:47,  1.64it/s]
 78%|███████▊  | 51126/65536 [8:48:36<2:24:48,  1.66it/s]
 78%|███████▊  | 51127/65536 [8:48:36<2:22:40,  1.68it/s]
 78%|███████▊  | 51128/65536 [8:48:37<2:25:46,  1.65it/s]
 78%|███████▊  | 51129/65536 [8:48:38<2:26:08,  1.64it/s]
 78%|███████▊  | 51130/65536 [8:48:38<2:27:07,  1.63it/s]
 78%|███████▊  | 51131/65536 [8:48:39<2:26:46,  1.64it/s]
 78%|███████▊  | 51132/65536 [8:48:40<2:26:17,  1.64it/s]
 78%|███████▊  | 51133/65536 [8:48:40<2:25:10,  1.65it/s]
 78%|███████▊  | 51134/65536 [8:48:41<2:25:11,  1.65it/s]
 78%|███████▊  | 51135/65536 [8:48:41<2:28:03,  1.62it/s]
 78%|███████▊  | 51136/65536 [8:48:42<2:26:33,  1.64it/s]
 78%|███████▊  | 51137/65536 [8:48:43<2:28:17,  1.62it/s]
 78%|███████▊  | 51138/65536 [8:48:43<2:28:48,  1.61it/s]
 78%|███████▊  | 51139/65536 [8:48:44<2:26:01,  1.64it/s]
 78%|███████▊  | 51140/65536 [8:48:44<2:24:17,  1.66it/s]
                                                         
{'loss': 1.6102, 'learning_rate': 2.9830415085098566e-07, 'epoch': 3156.79}
+
 78%|███████▊  | 51140/65536 [8:48:44<2:24:17,  1.66it/s]
 78%|███████▊  | 51141/65536 [8:48:45<2:25:09,  1.65it/s]
 78%|███████▊  | 51142/65536 [8:48:46<2:32:41,  1.57it/s]
 78%|███████▊  | 51143/65536 [8:48:46<2:30:37,  1.59it/s]
 78%|███████▊  | 51144/65536 [8:48:47<2:35:14,  1.55it/s]
 78%|███████▊  | 51145/65536 [8:48:48<2:33:45,  1.56it/s]
 78%|███████▊  | 51146/65536 [8:48:48<2:35:49,  1.54it/s]
 78%|███████▊  | 51147/65536 [8:48:49<2:31:36,  1.58it/s]
 78%|███████▊  | 51148/65536 [8:48:50<2:34:13,  1.55it/s]
 78%|███████▊  | 51149/65536 [8:48:50<2:32:21,  1.57it/s]
 78%|███████▊  | 51150/65536 [8:48:51<2:29:29,  1.60it/s]
 78%|███████▊  | 51151/65536 [8:48:51<2:28:50,  1.61it/s]
 78%|███████▊  | 51152/65536 [8:48:52<2:26:01,  1.64it/s]
 78%|███████▊  | 51153/65536 [8:48:53<2:29:23,  1.60it/s]
 78%|███████▊  | 51154/65536 [8:48:53<2:30:08,  1.60it/s]
 78%|███████▊  | 51155/65536 [8:48:54<2:30:09,  1.60it/s]
 78%|███████▊  | 51156/65536 [8:48:55<2:29:15,  1.61it/s]
 78%|███████▊  | 51157/65536 [8:48:55<2:28:29,  1.61it/s]
 78%|███████▊  | 51158/65536 [8:48:56<2:30:38,  1.59it/s]
 78%|███████▊  | 51159/65536 [8:48:56<2:31:17,  1.58it/s]
 78%|███████▊  | 51160/65536 [8:48:57<2:32:26,  1.57it/s]
                                                         
{'loss': 1.5385, 'learning_rate': 2.980286518917595e-07, 'epoch': 3158.02}
+
 78%|███████▊  | 51160/65536 [8:48:57<2:32:26,  1.57it/s]
 78%|███████▊  | 51161/65536 [8:48:58<2:38:57,  1.51it/s]
 78%|███████▊  | 51162/65536 [8:48:58<2:34:18,  1.55it/s]
 78%|███████▊  | 51163/65536 [8:48:59<2:34:30,  1.55it/s]
 78%|███████▊  | 51164/65536 [8:49:00<2:33:53,  1.56it/s]
 78%|███████▊  | 51165/65536 [8:49:00<2:30:16,  1.59it/s]
 78%|███████▊  | 51166/65536 [8:49:01<2:39:26,  1.50it/s]
 78%|███████▊  | 51167/65536 [8:49:02<2:35:54,  1.54it/s]
 78%|███████▊  | 51168/65536 [8:49:02<2:32:33,  1.57it/s]
 78%|███████▊  | 51169/65536 [8:49:03<2:32:37,  1.57it/s]
 78%|███████▊  | 51170/65536 [8:49:04<2:30:45,  1.59it/s]
 78%|███████▊  | 51171/65536 [8:49:04<2:28:50,  1.61it/s]
 78%|███████▊  | 51172/65536 [8:49:05<2:26:58,  1.63it/s]
 78%|███████▊  | 51173/65536 [8:49:05<2:30:30,  1.59it/s]
 78%|███████▊  | 51174/65536 [8:49:06<2:28:09,  1.62it/s]
 78%|███████▊  | 51175/65536 [8:49:07<2:28:59,  1.61it/s]
 78%|███████▊  | 51176/65536 [8:49:07<2:36:58,  1.52it/s]
 78%|███████▊  | 51177/65536 [8:49:08<2:33:32,  1.56it/s]
 78%|███████▊  | 51178/65536 [8:49:09<2:30:53,  1.59it/s]
 78%|███████▊  | 51179/65536 [8:49:09<2:35:01,  1.54it/s]
 78%|███████▊  | 51180/65536 [8:49:10<2:31:38,  1.58it/s]
                                                         
{'loss': 1.5312, 'learning_rate': 2.977531529325334e-07, 'epoch': 3159.26}
+
 78%|███████▊  | 51180/65536 [8:49:10<2:31:38,  1.58it/s]
 78%|███████▊  | 51181/65536 [8:49:10<2:30:13,  1.59it/s]
 78%|███████▊  | 51182/65536 [8:49:11<2:30:43,  1.59it/s]
 78%|███████▊  | 51183/65536 [8:49:12<2:29:08,  1.60it/s]
 78%|███████▊  | 51184/65536 [8:49:12<2:30:50,  1.59it/s]
 78%|███████▊  | 51185/65536 [8:49:13<2:30:22,  1.59it/s]
 78%|███████▊  | 51186/65536 [8:49:14<2:34:18,  1.55it/s]
 78%|███████▊  | 51187/65536 [8:49:14<2:31:49,  1.58it/s]
 78%|███████▊  | 51188/65536 [8:49:15<2:27:16,  1.62it/s]
 78%|███████▊  | 51189/65536 [8:49:16<2:30:40,  1.59it/s]
 78%|███████▊  | 51190/65536 [8:49:16<2:28:31,  1.61it/s]
 78%|███████▊  | 51191/65536 [8:49:17<2:26:01,  1.64it/s]
 78%|███████▊  | 51192/65536 [8:49:17<2:25:09,  1.65it/s]
 78%|███████▊  | 51193/65536 [8:49:18<2:30:11,  1.59it/s]
 78%|███████▊  | 51194/65536 [8:49:19<2:24:55,  1.65it/s]
 78%|███████▊  | 51195/65536 [8:49:19<2:24:46,  1.65it/s]
 78%|███████▊  | 51196/65536 [8:49:20<2:25:47,  1.64it/s]
 78%|███████▊  | 51197/65536 [8:49:20<2:26:31,  1.63it/s]
 78%|███████▊  | 51198/65536 [8:49:21<2:29:12,  1.60it/s]
 78%|███████▊  | 51199/65536 [8:49:22<2:27:01,  1.63it/s]
 78%|███████▊  | 51200/65536 [8:49:22<2:29:01,  1.60it/s]
                                                         
{'loss': 1.5608, 'learning_rate': 2.9747765397330724e-07, 'epoch': 3160.49}
+
 78%|███████▊  | 51200/65536 [8:49:22<2:29:01,  1.60it/s]
 78%|███████▊  | 51201/65536 [8:49:23<2:30:24,  1.59it/s]
 78%|███████▊  | 51202/65536 [8:49:24<2:31:19,  1.58it/s]
 78%|███████▊  | 51203/65536 [8:49:24<2:28:28,  1.61it/s]
 78%|███████▊  | 51204/65536 [8:49:25<2:28:50,  1.60it/s]
 78%|███████▊  | 51205/65536 [8:49:25<2:28:37,  1.61it/s]
 78%|███████▊  | 51206/65536 [8:49:26<2:33:56,  1.55it/s]
 78%|███████▊  | 51207/65536 [8:49:27<2:36:41,  1.52it/s]
 78%|███████▊  | 51208/65536 [8:49:27<2:34:18,  1.55it/s]
 78%|███████▊  | 51209/65536 [8:49:28<2:39:24,  1.50it/s]
 78%|███████▊  | 51210/65536 [8:49:29<2:34:38,  1.54it/s]
 78%|███████▊  | 51211/65536 [8:49:29<2:35:02,  1.54it/s]
 78%|███████▊  | 51212/65536 [8:49:30<2:33:59,  1.55it/s]
 78%|███████▊  | 51213/65536 [8:49:31<2:34:08,  1.55it/s]
 78%|███████▊  | 51214/65536 [8:49:31<2:32:39,  1.56it/s]
 78%|███████▊  | 51215/65536 [8:49:32<2:30:14,  1.59it/s]
 78%|███████▊  | 51216/65536 [8:49:32<2:26:44,  1.63it/s]
 78%|███████▊  | 51217/65536 [8:49:33<2:23:14,  1.67it/s]
 78%|███████▊  | 51218/65536 [8:49:34<2:22:43,  1.67it/s]
 78%|███████▊  | 51219/65536 [8:49:34<2:22:38,  1.67it/s]
 78%|███████▊  | 51220/65536 [8:49:35<2:23:28,  1.66it/s]
                                                         
{'loss': 1.5325, 'learning_rate': 2.97202155014081e-07, 'epoch': 3161.73}
+
 78%|███████▊  | 51220/65536 [8:49:35<2:23:28,  1.66it/s]
 78%|███████▊  | 51221/65536 [8:49:35<2:26:46,  1.63it/s]
 78%|███████▊  | 51222/65536 [8:49:36<2:28:12,  1.61it/s]
 78%|███████▊  | 51223/65536 [8:49:37<2:27:32,  1.62it/s]
 78%|███████▊  | 51224/65536 [8:49:37<2:26:34,  1.63it/s]
 78%|███████▊  | 51225/65536 [8:49:38<2:33:02,  1.56it/s]
 78%|███████▊  | 51226/65536 [8:49:39<2:30:44,  1.58it/s]
 78%|███████▊  | 51227/65536 [8:49:39<2:26:12,  1.63it/s]
 78%|███████▊  | 51228/65536 [8:49:40<2:24:44,  1.65it/s]
 78%|███████▊  | 51229/65536 [8:49:40<2:24:47,  1.65it/s]
 78%|███████▊  | 51230/65536 [8:49:41<2:25:40,  1.64it/s]
 78%|███████▊  | 51231/65536 [8:49:42<2:25:02,  1.64it/s]
 78%|███████▊  | 51232/65536 [8:49:42<2:30:05,  1.59it/s]
 78%|███████▊  | 51233/65536 [8:49:43<2:30:08,  1.59it/s]
 78%|███████▊  | 51234/65536 [8:49:44<2:29:07,  1.60it/s]
 78%|███████▊  | 51235/65536 [8:49:44<2:28:25,  1.61it/s]
 78%|███████▊  | 51236/65536 [8:49:45<2:30:09,  1.59it/s]
 78%|███████▊  | 51237/65536 [8:49:45<2:29:57,  1.59it/s]
 78%|███████▊  | 51238/65536 [8:49:46<2:31:45,  1.57it/s]
 78%|███████▊  | 51239/65536 [8:49:47<2:30:17,  1.59it/s]
 78%|███████▊  | 51240/65536 [8:49:47<2:29:49,  1.59it/s]
                                                         
{'loss': 1.5552, 'learning_rate': 2.9692665605485485e-07, 'epoch': 3162.96}
+
 78%|███████▊  | 51240/65536 [8:49:47<2:29:49,  1.59it/s]
 78%|███████▊  | 51241/65536 [8:49:48<2:35:27,  1.53it/s]
 78%|███████▊  | 51242/65536 [8:49:49<2:31:48,  1.57it/s]
 78%|███████▊  | 51243/65536 [8:49:49<2:29:06,  1.60it/s]
 78%|███████▊  | 51244/65536 [8:49:50<2:29:13,  1.60it/s]
 78%|███████▊  | 51245/65536 [8:49:51<2:29:06,  1.60it/s]
 78%|███████▊  | 51246/65536 [8:49:51<2:29:31,  1.59it/s]
 78%|███████▊  | 51247/65536 [8:49:52<2:29:50,  1.59it/s]
 78%|███████▊  | 51248/65536 [8:49:52<2:27:32,  1.61it/s]
 78%|███████▊  | 51249/65536 [8:49:53<2:26:53,  1.62it/s]
 78%|███████▊  | 51250/65536 [8:49:54<2:26:20,  1.63it/s]
 78%|███████▊  | 51251/65536 [8:49:54<2:26:48,  1.62it/s]
 78%|███████▊  | 51252/65536 [8:49:55<2:28:39,  1.60it/s]
 78%|███████▊  | 51253/65536 [8:49:55<2:25:40,  1.63it/s]
 78%|███████▊  | 51254/65536 [8:49:56<2:25:01,  1.64it/s]
 78%|███████▊  | 51255/65536 [8:49:57<2:23:17,  1.66it/s]
 78%|███████▊  | 51256/65536 [8:49:57<2:22:01,  1.68it/s]
 78%|███████▊  | 51257/65536 [8:49:58<2:34:12,  1.54it/s]
 78%|███████▊  | 51258/65536 [8:49:59<2:31:57,  1.57it/s]
 78%|███████▊  | 51259/65536 [8:49:59<2:27:51,  1.61it/s]
 78%|███████▊  | 51260/65536 [8:50:00<2:26:19,  1.63it/s]
                                                         
{'loss': 1.5954, 'learning_rate': 2.966511570956287e-07, 'epoch': 3164.2}
+
 78%|███████▊  | 51260/65536 [8:50:00<2:26:19,  1.63it/s]
 78%|███████▊  | 51261/65536 [8:50:00<2:26:53,  1.62it/s]
 78%|███████▊  | 51262/65536 [8:50:01<2:27:09,  1.62it/s]
 78%|███████▊  | 51263/65536 [8:50:02<2:26:57,  1.62it/s]
 78%|███████▊  | 51264/65536 [8:50:02<2:27:13,  1.62it/s]
 78%|███████▊  | 51265/65536 [8:50:03<2:27:32,  1.61it/s]
 78%|███████▊  | 51266/65536 [8:50:04<2:31:12,  1.57it/s]
 78%|███████▊  | 51267/65536 [8:50:04<2:32:14,  1.56it/s]
 78%|███████▊  | 51268/65536 [8:50:05<2:33:21,  1.55it/s]
 78%|███████▊  | 51269/65536 [8:50:05<2:31:31,  1.57it/s]
 78%|███████▊  | 51270/65536 [8:50:06<2:29:45,  1.59it/s]
 78%|███████▊  | 51271/65536 [8:50:07<2:29:17,  1.59it/s]
 78%|███████▊  | 51272/65536 [8:50:07<2:26:13,  1.63it/s]
 78%|███████▊  | 51273/65536 [8:50:08<2:26:18,  1.62it/s]
 78%|███████▊  | 51274/65536 [8:50:09<2:32:57,  1.55it/s]
 78%|███████▊  | 51275/65536 [8:50:09<2:35:09,  1.53it/s]
 78%|███████▊  | 51276/65536 [8:50:10<2:30:11,  1.58it/s]
 78%|███████▊  | 51277/65536 [8:50:11<2:34:59,  1.53it/s]
 78%|███████▊  | 51278/65536 [8:50:11<2:28:53,  1.60it/s]
 78%|███████▊  | 51279/65536 [8:50:12<2:30:01,  1.58it/s]
 78%|███████▊  | 51280/65536 [8:50:12<2:27:02,  1.62it/s]
                                                         
{'loss': 1.5833, 'learning_rate': 2.963756581364026e-07, 'epoch': 3165.43}
+
 78%|███████▊  | 51280/65536 [8:50:12<2:27:02,  1.62it/s]
 78%|███████▊  | 51281/65536 [8:50:13<2:28:30,  1.60it/s]
 78%|███████▊  | 51282/65536 [8:50:14<2:28:33,  1.60it/s]
 78%|███████▊  | 51283/65536 [8:50:14<2:27:51,  1.61it/s]
 78%|███████▊  | 51284/65536 [8:50:15<2:30:15,  1.58it/s]
 78%|███████▊  | 51285/65536 [8:50:16<2:28:54,  1.60it/s]
 78%|███████▊  | 51286/65536 [8:50:16<2:29:46,  1.59it/s]
 78%|███████▊  | 51287/65536 [8:50:17<2:32:50,  1.55it/s]
 78%|███████▊  | 51288/65536 [8:50:17<2:30:44,  1.58it/s]
 78%|███████▊  | 51289/65536 [8:50:18<2:27:54,  1.61it/s]
 78%|███████▊  | 51290/65536 [8:50:19<2:35:44,  1.52it/s]
 78%|███████▊  | 51291/65536 [8:50:19<2:31:35,  1.57it/s]
 78%|███████▊  | 51292/65536 [8:50:20<2:33:24,  1.55it/s]
 78%|███████▊  | 51293/65536 [8:50:21<2:30:55,  1.57it/s]
 78%|███████▊  | 51294/65536 [8:50:21<2:27:55,  1.60it/s]
 78%|███████▊  | 51295/65536 [8:50:22<2:27:29,  1.61it/s]
 78%|███████▊  | 51296/65536 [8:50:22<2:26:47,  1.62it/s]
 78%|███████▊  | 51297/65536 [8:50:23<2:24:43,  1.64it/s]
 78%|███████▊  | 51298/65536 [8:50:24<2:24:04,  1.65it/s]
 78%|███████▊  | 51299/65536 [8:50:24<2:27:10,  1.61it/s]
 78%|███████▊  | 51300/65536 [8:50:25<2:28:01,  1.60it/s]
                                                         
{'loss': 1.5642, 'learning_rate': 2.961001591771765e-07, 'epoch': 3166.67}
+
 78%|███████▊  | 51300/65536 [8:50:25<2:28:01,  1.60it/s]
 78%|███████▊  | 51301/65536 [8:50:26<2:31:08,  1.57it/s]
 78%|███████▊  | 51302/65536 [8:50:26<2:29:29,  1.59it/s]
 78%|███████▊  | 51303/65536 [8:50:27<2:31:12,  1.57it/s]
 78%|███████▊  | 51304/65536 [8:50:28<2:30:07,  1.58it/s]
 78%|███████▊  | 51305/65536 [8:50:28<2:26:43,  1.62it/s]
 78%|███████▊  | 51306/65536 [8:50:29<2:30:00,  1.58it/s]
 78%|███████▊  | 51307/65536 [8:50:29<2:34:25,  1.54it/s]
 78%|███████▊  | 51308/65536 [8:50:30<2:31:14,  1.57it/s]
 78%|███████▊  | 51309/65536 [8:50:31<2:27:19,  1.61it/s]
 78%|███████▊  | 51310/65536 [8:50:31<2:28:58,  1.59it/s]
 78%|███████▊  | 51311/65536 [8:50:32<2:25:36,  1.63it/s]
 78%|███████▊  | 51312/65536 [8:50:32<2:25:42,  1.63it/s]
 78%|███████▊  | 51313/65536 [8:50:33<2:34:25,  1.54it/s]
 78%|███████▊  | 51314/65536 [8:50:34<2:31:57,  1.56it/s]
 78%|███████▊  | 51315/65536 [8:50:34<2:30:17,  1.58it/s]
 78%|███████▊  | 51316/65536 [8:50:35<2:28:42,  1.59it/s]
 78%|███████▊  | 51317/65536 [8:50:36<2:26:04,  1.62it/s]
 78%|███████▊  | 51318/65536 [8:50:36<2:26:00,  1.62it/s]
 78%|███████▊  | 51319/65536 [8:50:37<2:26:02,  1.62it/s]
 78%|███████▊  | 51320/65536 [8:50:38<2:28:10,  1.60it/s]
                                                         
{'loss': 1.5585, 'learning_rate': 2.9582466021795034e-07, 'epoch': 3167.9}
+
 78%|███████▊  | 51320/65536 [8:50:38<2:28:10,  1.60it/s]
 78%|███████▊  | 51321/65536 [8:50:38<2:26:20,  1.62it/s]
 78%|███████▊  | 51322/65536 [8:50:39<2:32:58,  1.55it/s]
 78%|███████▊  | 51323/65536 [8:50:39<2:31:14,  1.57it/s]
 78%|███████▊  | 51324/65536 [8:50:40<2:28:33,  1.59it/s]
 78%|███████▊  | 51325/65536 [8:50:41<2:26:18,  1.62it/s]
 78%|███████▊  | 51326/65536 [8:50:41<2:30:05,  1.58it/s]
 78%|███████▊  | 51327/65536 [8:50:42<2:27:45,  1.60it/s]
 78%|███████▊  | 51328/65536 [8:50:43<2:28:11,  1.60it/s]
 78%|███████▊  | 51329/65536 [8:50:43<2:31:16,  1.57it/s]
 78%|███████▊  | 51330/65536 [8:50:44<2:28:06,  1.60it/s]
 78%|███████▊  | 51331/65536 [8:50:44<2:23:22,  1.65it/s]
 78%|███████▊  | 51332/65536 [8:50:45<2:24:43,  1.64it/s]
 78%|███████▊  | 51333/65536 [8:50:46<2:25:30,  1.63it/s]
 78%|███████▊  | 51334/65536 [8:50:46<2:27:19,  1.61it/s]
 78%|███████▊  | 51335/65536 [8:50:47<2:25:55,  1.62it/s]
 78%|███████▊  | 51336/65536 [8:50:48<2:30:12,  1.58it/s]
 78%|███████▊  | 51337/65536 [8:50:48<2:25:14,  1.63it/s]
 78%|███████▊  | 51338/65536 [8:50:49<2:25:23,  1.63it/s]
 78%|███████▊  | 51339/65536 [8:50:49<2:32:38,  1.55it/s]
 78%|███████▊  | 51340/65536 [8:50:50<2:30:03,  1.58it/s]
                                                         
{'loss': 1.5686, 'learning_rate': 2.955491612587242e-07, 'epoch': 3169.14}
+
 78%|███████▊  | 51340/65536 [8:50:50<2:30:03,  1.58it/s]
 78%|███████▊  | 51341/65536 [8:50:51<2:32:46,  1.55it/s]
 78%|███████▊  | 51342/65536 [8:50:51<2:29:59,  1.58it/s]
 78%|███████▊  | 51343/65536 [8:50:52<2:27:11,  1.61it/s]
 78%|███████▊  | 51344/65536 [8:50:53<2:22:47,  1.66it/s]
 78%|███████▊  | 51345/65536 [8:50:53<2:24:36,  1.64it/s]
 78%|███████▊  | 51346/65536 [8:50:54<2:25:06,  1.63it/s]
 78%|███████▊  | 51347/65536 [8:50:54<2:26:16,  1.62it/s]
 78%|███████▊  | 51348/65536 [8:50:55<2:25:29,  1.63it/s]
 78%|███████▊  | 51349/65536 [8:50:56<2:24:36,  1.64it/s]
 78%|███████▊  | 51350/65536 [8:50:56<2:23:06,  1.65it/s]
 78%|███████▊  | 51351/65536 [8:50:57<2:29:53,  1.58it/s]
 78%|███████▊  | 51352/65536 [8:50:57<2:28:07,  1.60it/s]
 78%|███████▊  | 51353/65536 [8:50:58<2:24:47,  1.63it/s]
 78%|███████▊  | 51354/65536 [8:50:59<2:22:43,  1.66it/s]
 78%|███████▊  | 51355/65536 [8:50:59<2:28:46,  1.59it/s]
 78%|███████▊  | 51356/65536 [8:51:00<2:34:02,  1.53it/s]
 78%|███████▊  | 51357/65536 [8:51:01<2:33:33,  1.54it/s]
 78%|███████▊  | 51358/65536 [8:51:01<2:30:08,  1.57it/s]
 78%|███████▊  | 51359/65536 [8:51:02<2:25:55,  1.62it/s]
 78%|███████▊  | 51360/65536 [8:51:02<2:24:05,  1.64it/s]
                                                         
{'loss': 1.5734, 'learning_rate': 2.9527366229949795e-07, 'epoch': 3170.37}
+
 78%|███████▊  | 51360/65536 [8:51:02<2:24:05,  1.64it/s]
 78%|███████▊  | 51361/65536 [8:51:03<2:25:43,  1.62it/s]
 78%|███████▊  | 51362/65536 [8:51:04<2:26:47,  1.61it/s]
 78%|███████▊  | 51363/65536 [8:51:04<2:23:41,  1.64it/s]
 78%|███████▊  | 51364/65536 [8:51:05<2:27:14,  1.60it/s]
 78%|███████▊  | 51365/65536 [8:51:06<2:27:56,  1.60it/s]
 78%|███████▊  | 51366/65536 [8:51:06<2:28:23,  1.59it/s]
 78%|███████▊  | 51367/65536 [8:51:07<2:26:05,  1.62it/s]
 78%|███████▊  | 51368/65536 [8:51:07<2:25:08,  1.63it/s]
 78%|███████▊  | 51369/65536 [8:51:08<2:24:44,  1.63it/s]
 78%|███████▊  | 51370/65536 [8:51:09<2:26:10,  1.62it/s]
 78%|███████▊  | 51371/65536 [8:51:09<2:29:28,  1.58it/s]
 78%|███████▊  | 51372/65536 [8:51:10<2:28:33,  1.59it/s]
 78%|███████▊  | 51373/65536 [8:51:11<2:26:51,  1.61it/s]
 78%|███████▊  | 51374/65536 [8:51:11<2:25:36,  1.62it/s]
 78%|███████▊  | 51375/65536 [8:51:12<2:21:08,  1.67it/s]
 78%|███████▊  | 51376/65536 [8:51:12<2:24:07,  1.64it/s]
 78%|███████▊  | 51377/65536 [8:51:13<2:23:21,  1.65it/s]
 78%|███████▊  | 51378/65536 [8:51:14<2:22:40,  1.65it/s]
 78%|███████▊  | 51379/65536 [8:51:14<2:22:50,  1.65it/s]
 78%|███████▊  | 51380/65536 [8:51:15<2:25:07,  1.63it/s]
                                                         
{'loss': 1.5736, 'learning_rate': 2.949981633402718e-07, 'epoch': 3171.6}
+
 78%|███████▊  | 51380/65536 [8:51:15<2:25:07,  1.63it/s]
 78%|███████▊  | 51381/65536 [8:51:15<2:25:51,  1.62it/s]
 78%|███████▊  | 51382/65536 [8:51:16<2:24:28,  1.63it/s]
 78%|███████▊  | 51383/65536 [8:51:17<2:23:23,  1.65it/s]
 78%|███████▊  | 51384/65536 [8:51:17<2:26:58,  1.60it/s]
 78%|███████▊  | 51385/65536 [8:51:18<2:32:01,  1.55it/s]
 78%|███████▊  | 51386/65536 [8:51:19<2:32:06,  1.55it/s]
 78%|███████▊  | 51387/65536 [8:51:19<2:32:05,  1.55it/s]
 78%|███████▊  | 51388/65536 [8:51:20<2:27:41,  1.60it/s]
 78%|███████▊  | 51389/65536 [8:51:20<2:25:46,  1.62it/s]
 78%|███████▊  | 51390/65536 [8:51:21<2:25:08,  1.62it/s]
 78%|███████▊  | 51391/65536 [8:51:22<2:25:18,  1.62it/s]
 78%|███████▊  | 51392/65536 [8:51:22<2:23:17,  1.65it/s]
 78%|███████▊  | 51393/65536 [8:51:23<2:22:02,  1.66it/s]
 78%|███████▊  | 51394/65536 [8:51:23<2:23:16,  1.65it/s]
 78%|███████▊  | 51395/65536 [8:51:24<2:25:24,  1.62it/s]
 78%|███████▊  | 51396/65536 [8:51:25<2:26:30,  1.61it/s]
 78%|███████▊  | 51397/65536 [8:51:25<2:27:53,  1.59it/s]
 78%|███████▊  | 51398/65536 [8:51:26<2:31:44,  1.55it/s]
 78%|███████▊  | 51399/65536 [8:51:27<2:31:31,  1.55it/s]
 78%|███████▊  | 51400/65536 [8:51:27<2:29:37,  1.57it/s]
                                                         
{'loss': 1.542, 'learning_rate': 2.9472266438104567e-07, 'epoch': 3172.84}
+
 78%|███████▊  | 51400/65536 [8:51:27<2:29:37,  1.57it/s]
 78%|███████▊  | 51401/65536 [8:51:28<2:30:32,  1.56it/s]
 78%|███████▊  | 51402/65536 [8:51:29<2:29:09,  1.58it/s]
 78%|███████▊  | 51403/65536 [8:51:29<2:31:49,  1.55it/s]
 78%|███████▊  | 51404/65536 [8:51:30<2:31:32,  1.55it/s]
 78%|███████▊  | 51405/65536 [8:51:31<2:27:00,  1.60it/s]
 78%|███████▊  | 51406/65536 [8:51:31<2:26:26,  1.61it/s]
 78%|███████▊  | 51407/65536 [8:51:32<2:26:02,  1.61it/s]
 78%|███████▊  | 51408/65536 [8:51:32<2:24:04,  1.63it/s]
 78%|███████▊  | 51409/65536 [8:51:33<2:23:49,  1.64it/s]
 78%|███████▊  | 51410/65536 [8:51:34<2:26:32,  1.61it/s]
 78%|███████▊  | 51411/65536 [8:51:34<2:25:48,  1.61it/s]
 78%|███████▊  | 51412/65536 [8:51:35<2:26:44,  1.60it/s]
 78%|███████▊  | 51413/65536 [8:51:35<2:29:05,  1.58it/s]
 78%|███████▊  | 51414/65536 [8:51:36<2:26:45,  1.60it/s]
 78%|███████▊  | 51415/65536 [8:51:37<2:28:19,  1.59it/s]
 78%|███████▊  | 51416/65536 [8:51:37<2:25:04,  1.62it/s]
 78%|███████▊  | 51417/65536 [8:51:38<2:21:48,  1.66it/s]
 78%|███████▊  | 51418/65536 [8:51:38<2:21:03,  1.67it/s]
 78%|███████▊  | 51419/65536 [8:51:39<2:29:37,  1.57it/s]
 78%|███████▊  | 51420/65536 [8:51:40<2:27:49,  1.59it/s]
                                                         
{'loss': 1.5903, 'learning_rate': 2.9444716542181953e-07, 'epoch': 3174.07}
+
 78%|███████▊  | 51420/65536 [8:51:40<2:27:49,  1.59it/s]
 78%|███████▊  | 51421/65536 [8:51:40<2:24:04,  1.63it/s]
 78%|███████▊  | 51422/65536 [8:51:41<2:23:34,  1.64it/s]
 78%|███████▊  | 51423/65536 [8:51:42<2:23:07,  1.64it/s]
 78%|███████▊  | 51424/65536 [8:51:42<2:27:35,  1.59it/s]
 78%|███████▊  | 51425/65536 [8:51:43<2:23:52,  1.63it/s]
 78%|███████▊  | 51426/65536 [8:51:43<2:21:19,  1.66it/s]
 78%|███████▊  | 51427/65536 [8:51:44<2:27:56,  1.59it/s]
 78%|███████▊  | 51428/65536 [8:51:45<2:24:31,  1.63it/s]
 78%|███████▊  | 51429/65536 [8:51:45<2:25:16,  1.62it/s]
 78%|███████▊  | 51430/65536 [8:51:46<2:25:54,  1.61it/s]
 78%|███████▊  | 51431/65536 [8:51:47<2:26:41,  1.60it/s]
 78%|███████▊  | 51432/65536 [8:51:47<2:25:23,  1.62it/s]
 78%|███████▊  | 51433/65536 [8:51:48<2:26:20,  1.61it/s]
 78%|███████▊  | 51434/65536 [8:51:48<2:24:43,  1.62it/s]
 78%|███████▊  | 51435/65536 [8:51:49<2:21:26,  1.66it/s]
 78%|███████▊  | 51436/65536 [8:51:50<2:28:27,  1.58it/s]
 78%|███████▊  | 51437/65536 [8:51:50<2:24:46,  1.62it/s]
 78%|███████▊  | 51438/65536 [8:51:51<2:24:51,  1.62it/s]
 78%|███████▊  | 51439/65536 [8:51:51<2:24:18,  1.63it/s]
 78%|███████▊  | 51440/65536 [8:51:52<2:24:43,  1.62it/s]
                                                         
{'loss': 1.5849, 'learning_rate': 2.941716664625934e-07, 'epoch': 3175.31}
+
 78%|███████▊  | 51440/65536 [8:51:52<2:24:43,  1.62it/s]
 78%|███████▊  | 51441/65536 [8:51:53<2:25:28,  1.61it/s]
 78%|███████▊  | 51442/65536 [8:51:53<2:29:45,  1.57it/s]
 78%|███████▊  | 51443/65536 [8:51:54<2:28:41,  1.58it/s]
 78%|███████▊  | 51444/65536 [8:51:55<2:25:51,  1.61it/s]
 78%|███████▊  | 51445/65536 [8:51:55<2:24:07,  1.63it/s]
 79%|███████▊  | 51446/65536 [8:51:56<2:28:56,  1.58it/s]
 79%|███████▊  | 51447/65536 [8:51:57<2:28:51,  1.58it/s]
 79%|███████▊  | 51448/65536 [8:51:57<2:24:03,  1.63it/s]
 79%|███████▊  | 51449/65536 [8:51:58<2:23:47,  1.63it/s]
 79%|███████▊  | 51450/65536 [8:51:58<2:24:46,  1.62it/s]
 79%|███████▊  | 51451/65536 [8:51:59<2:25:32,  1.61it/s]
 79%|███████▊  | 51452/65536 [8:52:00<2:31:13,  1.55it/s]
 79%|███████▊  | 51453/65536 [8:52:00<2:31:23,  1.55it/s]
 79%|███████▊  | 51454/65536 [8:52:01<2:29:36,  1.57it/s]
 79%|███████▊  | 51455/65536 [8:52:02<2:26:53,  1.60it/s]
 79%|███████▊  | 51456/65536 [8:52:02<2:27:20,  1.59it/s]
 79%|███████▊  | 51457/65536 [8:52:03<2:38:17,  1.48it/s]
 79%|███████▊  | 51458/65536 [8:52:04<2:38:19,  1.48it/s]
 79%|███████▊  | 51459/65536 [8:52:04<2:32:44,  1.54it/s]
 79%|███████▊  | 51460/65536 [8:52:05<2:28:21,  1.58it/s]
                                                         
{'loss': 1.5678, 'learning_rate': 2.9389616750336725e-07, 'epoch': 3176.54}
+
 79%|███████▊  | 51460/65536 [8:52:05<2:28:21,  1.58it/s]
 79%|██████��▊  | 51461/65536 [8:52:05<2:28:05,  1.58it/s]
 79%|███████▊  | 51462/65536 [8:52:06<2:25:14,  1.61it/s]
 79%|███████▊  | 51463/65536 [8:52:07<2:24:43,  1.62it/s]
 79%|███████▊  | 51464/65536 [8:52:07<2:23:36,  1.63it/s]
 79%|███████▊  | 51465/65536 [8:52:08<2:22:56,  1.64it/s]
 79%|███████▊  | 51466/65536 [8:52:08<2:21:02,  1.66it/s]
 79%|███████▊  | 51467/65536 [8:52:09<2:24:47,  1.62it/s]
 79%|███████▊  | 51468/65536 [8:52:10<2:30:01,  1.56it/s]
 79%|███████▊  | 51469/65536 [8:52:10<2:33:47,  1.52it/s]
 79%|███████▊  | 51470/65536 [8:52:11<2:30:09,  1.56it/s]
 79%|███████▊  | 51471/65536 [8:52:12<2:24:50,  1.62it/s]
 79%|███████▊  | 51472/65536 [8:52:12<2:22:16,  1.65it/s]
 79%|███████▊  | 51473/65536 [8:52:13<2:19:39,  1.68it/s]
 79%|███████▊  | 51474/65536 [8:52:13<2:20:55,  1.66it/s]
 79%|███████▊  | 51475/65536 [8:52:14<2:22:45,  1.64it/s]
 79%|███████▊  | 51476/65536 [8:52:15<2:26:34,  1.60it/s]
 79%|███████▊  | 51477/65536 [8:52:15<2:24:55,  1.62it/s]
 79%|███████▊  | 51478/65536 [8:52:16<2:24:52,  1.62it/s]
 79%|███████▊  | 51479/65536 [8:52:17<2:22:47,  1.64it/s]
 79%|███████▊  | 51480/65536 [8:52:17<2:25:12,  1.61it/s]
                                                         
{'loss': 1.5985, 'learning_rate': 2.9362066854414105e-07, 'epoch': 3177.78}
+
 79%|███████▊  | 51480/65536 [8:52:17<2:25:12,  1.61it/s]
 79%|███████▊  | 51481/65536 [8:52:18<2:23:18,  1.63it/s]
 79%|███████▊  | 51482/65536 [8:52:18<2:23:47,  1.63it/s]
 79%|███████▊  | 51483/65536 [8:52:19<2:23:28,  1.63it/s]
 79%|███████▊  | 51484/65536 [8:52:20<2:28:49,  1.57it/s]
 79%|███████▊  | 51485/65536 [8:52:20<2:26:43,  1.60it/s]
 79%|███████▊  | 51486/65536 [8:52:21<2:24:42,  1.62it/s]
 79%|███████▊  | 51487/65536 [8:52:21<2:25:05,  1.61it/s]
 79%|███████▊  | 51488/65536 [8:52:22<2:24:31,  1.62it/s]
 79%|███████▊  | 51489/65536 [8:52:23<2:22:07,  1.65it/s]
 79%|███████▊  | 51490/65536 [8:52:23<2:20:54,  1.66it/s]
 79%|███████▊  | 51491/65536 [8:52:24<2:21:40,  1.65it/s]
 79%|███████▊  | 51492/65536 [8:52:24<2:17:18,  1.70it/s]
 79%|███████▊  | 51493/65536 [8:52:25<2:17:18,  1.70it/s]
 79%|███████▊  | 51494/65536 [8:52:26<2:23:23,  1.63it/s]
 79%|███████▊  | 51495/65536 [8:52:26<2:24:05,  1.62it/s]
 79%|███████▊  | 51496/65536 [8:52:27<2:26:07,  1.60it/s]
 79%|███████▊  | 51497/65536 [8:52:28<2:27:29,  1.59it/s]
 79%|███████▊  | 51498/65536 [8:52:28<2:28:43,  1.57it/s]
 79%|███████▊  | 51499/65536 [8:52:29<2:25:41,  1.61it/s]
 79%|███████▊  | 51500/65536 [8:52:30<2:30:27,  1.55it/s]
                                                         
{'loss': 1.5859, 'learning_rate': 2.9334516958491486e-07, 'epoch': 3179.01}
+
 79%|███████▊  | 51500/65536 [8:52:30<2:30:27,  1.55it/s]
 79%|███████▊  | 51501/65536 [8:52:30<2:25:59,  1.60it/s]
 79%|███████▊  | 51502/65536 [8:52:31<2:25:45,  1.60it/s]
 79%|███████▊  | 51503/65536 [8:52:31<2:23:05,  1.63it/s]
 79%|███████▊  | 51504/65536 [8:52:32<2:22:25,  1.64it/s]
 79%|███████▊  | 51505/65536 [8:52:33<2:22:02,  1.65it/s]
 79%|███████▊  | 51506/65536 [8:52:33<2:21:10,  1.66it/s]
 79%|███████▊  | 51507/65536 [8:52:34<2:26:39,  1.59it/s]
 79%|███████▊  | 51508/65536 [8:52:35<2:33:52,  1.52it/s]
 79%|███████▊  | 51509/65536 [8:52:35<2:31:16,  1.55it/s]
 79%|███████▊  | 51510/65536 [8:52:36<2:28:48,  1.57it/s]
 79%|███████▊  | 51511/65536 [8:52:36<2:27:25,  1.59it/s]
 79%|███████▊  | 51512/65536 [8:52:37<2:25:23,  1.61it/s]
 79%|███████▊  | 51513/65536 [8:52:38<2:31:04,  1.55it/s]
 79%|███████▊  | 51514/65536 [8:52:38<2:27:12,  1.59it/s]
 79%|███████▊  | 51515/65536 [8:52:39<2:29:47,  1.56it/s]
 79%|███████▊  | 51516/65536 [8:52:40<2:28:11,  1.58it/s]
 79%|███████▊  | 51517/65536 [8:52:40<2:32:41,  1.53it/s]
 79%|███████▊  | 51518/65536 [8:52:41<2:26:52,  1.59it/s]
 79%|███████▊  | 51519/65536 [8:52:41<2:25:47,  1.60it/s]
 79%|███████▊  | 51520/65536 [8:52:42<2:21:42,  1.65it/s]
                                                         
{'loss': 1.5835, 'learning_rate': 2.930696706256887e-07, 'epoch': 3180.25}
+
 79%|███████▊  | 51520/65536 [8:52:42<2:21:42,  1.65it/s]
 79%|███████▊  | 51521/65536 [8:52:43<2:26:02,  1.60it/s]
 79%|███████▊  | 51522/65536 [8:52:43<2:22:58,  1.63it/s]
 79%|███████▊  | 51523/65536 [8:52:44<2:24:50,  1.61it/s]
 79%|███████▊  | 51524/65536 [8:52:44<2:22:29,  1.64it/s]
 79%|███████▊  | 51525/65536 [8:52:45<2:20:48,  1.66it/s]
 79%|███████▊  | 51526/65536 [8:52:46<2:25:09,  1.61it/s]
 79%|███████▊  | 51527/65536 [8:52:46<2:24:22,  1.62it/s]
 79%|███████▊  | 51528/65536 [8:52:47<2:26:06,  1.60it/s]
 79%|███████▊  | 51529/65536 [8:52:48<2:25:37,  1.60it/s]
 79%|███████▊  | 51530/65536 [8:52:48<2:24:53,  1.61it/s]
 79%|███████▊  | 51531/65536 [8:52:49<2:24:33,  1.61it/s]
 79%|███████▊  | 51532/65536 [8:52:49<2:21:12,  1.65it/s]
 79%|███████▊  | 51533/65536 [8:52:50<2:25:47,  1.60it/s]
 79%|███████▊  | 51534/65536 [8:52:51<2:22:41,  1.64it/s]
 79%|███████▊  | 51535/65536 [8:52:51<2:23:35,  1.63it/s]
 79%|███████▊  | 51536/65536 [8:52:52<2:28:50,  1.57it/s]
 79%|███████▊  | 51537/65536 [8:52:53<2:25:25,  1.60it/s]
 79%|███████▊  | 51538/65536 [8:52:53<2:23:57,  1.62it/s]
 79%|███████▊  | 51539/65536 [8:52:54<2:23:57,  1.62it/s]
 79%|███████▊  | 51540/65536 [8:52:54<2:21:14,  1.65it/s]
                                                         
{'loss': 1.5738, 'learning_rate': 2.927941716664626e-07, 'epoch': 3181.48}
+
 79%|███████▊  | 51540/65536 [8:52:54<2:21:14,  1.65it/s]
 79%|███████▊  | 51541/65536 [8:52:55<2:23:29,  1.63it/s]
 79%|███████▊  | 51542/65536 [8:52:56<2:22:56,  1.63it/s]
 79%|███████▊  | 51543/65536 [8:52:56<2:21:47,  1.64it/s]
 79%|███████▊  | 51544/65536 [8:52:57<2:21:54,  1.64it/s]
 79%|███████▊  | 51545/65536 [8:52:57<2:22:24,  1.64it/s]
 79%|███████▊  | 51546/65536 [8:52:58<2:22:10,  1.64it/s]
 79%|███████▊  | 51547/65536 [8:52:59<2:21:37,  1.65it/s]
 79%|███████▊  | 51548/65536 [8:52:59<2:28:00,  1.58it/s]
 79%|███████▊  | 51549/65536 [8:53:00<2:28:07,  1.57it/s]
 79%|███████▊  | 51550/65536 [8:53:01<2:26:54,  1.59it/s]
 79%|███████▊  | 51551/65536 [8:53:01<2:27:19,  1.58it/s]
 79%|███████▊  | 51552/65536 [8:53:02<2:24:43,  1.61it/s]
 79%|███████▊  | 51553/65536 [8:53:02<2:23:51,  1.62it/s]
 79%|███████▊  | 51554/65536 [8:53:03<2:27:53,  1.58it/s]
 79%|███████▊  | 51555/65536 [8:53:04<2:26:39,  1.59it/s]
 79%|███████▊  | 51556/65536 [8:53:04<2:22:58,  1.63it/s]
 79%|███████▊  | 51557/65536 [8:53:05<2:27:34,  1.58it/s]
 79%|███████▊  | 51558/65536 [8:53:06<2:30:40,  1.55it/s]
 79%|███████▊  | 51559/65536 [8:53:06<2:27:03,  1.58it/s]
 79%|███████▊  | 51560/65536 [8:53:07<2:26:13,  1.59it/s]
                                                         
{'loss': 1.5584, 'learning_rate': 2.9251867270723643e-07, 'epoch': 3182.72}
+
 79%|███████▊  | 51560/65536 [8:53:07<2:26:13,  1.59it/s]
 79%|███████▊  | 51561/65536 [8:53:07<2:24:33,  1.61it/s]
 79%|███████▊  | 51562/65536 [8:53:08<2:23:48,  1.62it/s]
 79%|███████▊  | 51563/65536 [8:53:09<2:22:55,  1.63it/s]
 79%|███████▊  | 51564/65536 [8:53:09<2:21:48,  1.64it/s]
 79%|███████▊  | 51565/65536 [8:53:10<2:27:35,  1.58it/s]
 79%|███████▊  | 51566/65536 [8:53:11<2:27:32,  1.58it/s]
 79%|███████▊  | 51567/65536 [8:53:11<2:28:11,  1.57it/s]
 79%|███████▊  | 51568/65536 [8:53:12<2:26:34,  1.59it/s]
 79%|███████▊  | 51569/65536 [8:53:12<2:24:12,  1.61it/s]
 79%|███████▊  | 51570/65536 [8:53:13<2:24:35,  1.61it/s]
 79%|███████▊  | 51571/65536 [8:53:14<2:27:21,  1.58it/s]
 79%|███████▊  | 51572/65536 [8:53:14<2:25:31,  1.60it/s]
 79%|███████▊  | 51573/65536 [8:53:15<2:21:05,  1.65it/s]
 79%|███████▊  | 51574/65536 [8:53:16<2:19:50,  1.66it/s]
 79%|███████▊  | 51575/65536 [8:53:16<2:18:57,  1.67it/s]
 79%|███████▊  | 51576/65536 [8:53:17<2:22:17,  1.64it/s]
 79%|███████▊  | 51577/65536 [8:53:17<2:19:32,  1.67it/s]
 79%|███████▊  | 51578/65536 [8:53:18<2:19:38,  1.67it/s]
 79%|███████▊  | 51579/65536 [8:53:19<2:20:40,  1.65it/s]
 79%|███████▊  | 51580/65536 [8:53:19<2:20:53,  1.65it/s]
                                                         
{'loss': 1.6035, 'learning_rate': 2.922431737480103e-07, 'epoch': 3183.95}
+
 79%|███████▊  | 51580/65536 [8:53:19<2:20:53,  1.65it/s]
 79%|███████▊  | 51581/65536 [8:53:20<2:28:52,  1.56it/s]
 79%|███████▊  | 51582/65536 [8:53:20<2:25:22,  1.60it/s]
 79%|███████▊  | 51583/65536 [8:53:21<2:25:07,  1.60it/s]
 79%|███████▊  | 51584/65536 [8:53:22<2:26:01,  1.59it/s]
 79%|███████▊  | 51585/65536 [8:53:22<2:25:06,  1.60it/s]
 79%|███████▊  | 51586/65536 [8:53:23<2:25:27,  1.60it/s]
 79%|███████▊  | 51587/65536 [8:53:24<2:24:28,  1.61it/s]
 79%|███████▊  | 51588/65536 [8:53:24<2:21:36,  1.64it/s]
 79%|███████▊  | 51589/65536 [8:53:25<2:20:51,  1.65it/s]
 79%|███████▊  | 51590/65536 [8:53:25<2:19:05,  1.67it/s]
 79%|███████▊  | 51591/65536 [8:53:26<2:23:14,  1.62it/s]
 79%|███████▊  | 51592/65536 [8:53:27<2:21:39,  1.64it/s]
 79%|███████▊  | 51593/65536 [8:53:27<2:21:53,  1.64it/s]
 79%|███████▊  | 51594/65536 [8:53:28<2:29:43,  1.55it/s]
 79%|███████▊  | 51595/65536 [8:53:29<2:29:50,  1.55it/s]
 79%|███████▊  | 51596/65536 [8:53:29<2:25:16,  1.60it/s]
 79%|███████▊  | 51597/65536 [8:53:30<2:22:57,  1.63it/s]
 79%|███████▊  | 51598/65536 [8:53:30<2:30:03,  1.55it/s]
 79%|███████▊  | 51599/65536 [8:53:31<2:29:27,  1.55it/s]
 79%|███████▊  | 51600/65536 [8:53:32<2:24:55,  1.60it/s]
                                                         
{'loss': 1.5679, 'learning_rate': 2.9196767478878415e-07, 'epoch': 3185.19}
+
 79%|███████▊  | 51600/65536 [8:53:32<2:24:55,  1.60it/s]
 79%|███████▊  | 51601/65536 [8:53:32<2:26:18,  1.59it/s]
 79%|███████▊  | 51602/65536 [8:53:33<2:23:12,  1.62it/s]
 79%|███████▊  | 51603/65536 [8:53:34<2:21:36,  1.64it/s]
 79%|███████▊  | 51604/65536 [8:53:34<2:18:25,  1.68it/s]
 79%|███████▊  | 51605/65536 [8:53:35<2:23:14,  1.62it/s]
 79%|███████▊  | 51606/65536 [8:53:35<2:20:27,  1.65it/s]
 79%|███████▊  | 51607/65536 [8:53:36<2:22:02,  1.63it/s]
 79%|███████▊  | 51608/65536 [8:53:37<2:25:29,  1.60it/s]
 79%|███████▊  | 51609/65536 [8:53:37<2:23:43,  1.62it/s]
 79%|███████▉  | 51610/65536 [8:53:38<2:23:10,  1.62it/s]
 79%|███████▉  | 51611/65536 [8:53:38<2:26:10,  1.59it/s]
 79%|███████▉  | 51612/65536 [8:53:39<2:24:58,  1.60it/s]
 79%|███████▉  | 51613/65536 [8:53:40<2:22:31,  1.63it/s]
 79%|███████▉  | 51614/65536 [8:53:40<2:25:59,  1.59it/s]
 79%|███████▉  | 51615/65536 [8:53:41<2:27:58,  1.57it/s]
 79%|███████▉  | 51616/65536 [8:53:42<2:21:23,  1.64it/s]
 79%|███████▉  | 51617/65536 [8:53:42<2:20:37,  1.65it/s]
 79%|███████▉  | 51618/65536 [8:53:43<2:20:38,  1.65it/s]
 79%|███████▉  | 51619/65536 [8:53:43<2:18:30,  1.67it/s]
 79%|███████▉  | 51620/65536 [8:53:44<2:24:43,  1.60it/s]
                                                         
{'loss': 1.5901, 'learning_rate': 2.916921758295579e-07, 'epoch': 3186.42}
+
 79%|███████▉  | 51620/65536 [8:53:44<2:24:43,  1.60it/s]
 79%|███████▉  | 51621/65536 [8:53:45<2:24:51,  1.60it/s]
 79%|███████▉  | 51622/65536 [8:53:45<2:22:58,  1.62it/s]
 79%|███████▉  | 51623/65536 [8:53:46<2:25:07,  1.60it/s]
 79%|███████▉  | 51624/65536 [8:53:47<2:25:00,  1.60it/s]
 79%|███████▉  | 51625/65536 [8:53:47<2:25:17,  1.60it/s]
 79%|███████▉  | 51626/65536 [8:53:48<2:24:30,  1.60it/s]
 79%|███████▉  | 51627/65536 [8:53:48<2:27:57,  1.57it/s]
 79%|███████▉  | 51628/65536 [8:53:49<2:26:22,  1.58it/s]
 79%|███████▉  | 51629/65536 [8:53:50<2:24:44,  1.60it/s]
 79%|███████▉  | 51630/65536 [8:53:50<2:30:42,  1.54it/s]
 79%|███████▉  | 51631/65536 [8:53:51<2:28:18,  1.56it/s]
 79%|███████▉  | 51632/65536 [8:53:52<2:22:12,  1.63it/s]
 79%|███████▉  | 51633/65536 [8:53:52<2:22:47,  1.62it/s]
 79%|███████▉  | 51634/65536 [8:53:53<2:19:52,  1.66it/s]
 79%|███████▉  | 51635/65536 [8:53:53<2:20:31,  1.65it/s]
 79%|███████▉  | 51636/65536 [8:53:54<2:19:03,  1.67it/s]
 79%|███████▉  | 51637/65536 [8:53:55<2:19:00,  1.67it/s]
 79%|███████▉  | 51638/65536 [8:53:55<2:18:54,  1.67it/s]
 79%|███████▉  | 51639/65536 [8:53:56<2:21:37,  1.64it/s]
 79%|███████▉  | 51640/65536 [8:53:56<2:27:36,  1.57it/s]
                                                         
{'loss': 1.558, 'learning_rate': 2.9141667687033176e-07, 'epoch': 3187.65}
+
 79%|███████▉  | 51640/65536 [8:53:56<2:27:36,  1.57it/s]
 79%|███████▉  | 51641/65536 [8:53:57<2:25:07,  1.60it/s]
 79%|███████▉  | 51642/65536 [8:53:58<2:26:46,  1.58it/s]
 79%|███████▉  | 51643/65536 [8:53:58<2:24:00,  1.61it/s]
 79%|███████▉  | 51644/65536 [8:53:59<2:28:00,  1.56it/s]
 79%|███████▉  | 51645/65536 [8:54:00<2:26:05,  1.58it/s]
 79%|███████▉  | 51646/65536 [8:54:00<2:29:02,  1.55it/s]
 79%|███████▉  | 51647/65536 [8:54:01<2:28:45,  1.56it/s]
 79%|███████▉  | 51648/65536 [8:54:01<2:23:53,  1.61it/s]
 79%|███████▉  | 51649/65536 [8:54:02<2:23:14,  1.62it/s]
 79%|███████▉  | 51650/65536 [8:54:03<2:21:22,  1.64it/s]
 79%|███████▉  | 51651/65536 [8:54:03<2:20:06,  1.65it/s]
 79%|███████▉  | 51652/65536 [8:54:04<2:27:31,  1.57it/s]
 79%|███████▉  | 51653/65536 [8:54:05<2:26:11,  1.58it/s]
 79%|███████▉  | 51654/65536 [8:54:05<2:23:53,  1.61it/s]
 79%|███████▉  | 51655/65536 [8:54:06<2:21:55,  1.63it/s]
 79%|███████▉  | 51656/65536 [8:54:06<2:22:37,  1.62it/s]
 79%|███████▉  | 51657/65536 [8:54:07<2:22:22,  1.62it/s]
 79%|███████▉  | 51658/65536 [8:54:08<2:27:35,  1.57it/s]
 79%|███████▉  | 51659/65536 [8:54:08<2:26:42,  1.58it/s]
 79%|███████▉  | 51660/65536 [8:54:09<2:22:33,  1.62it/s]
                                                         
{'loss': 1.5901, 'learning_rate': 2.9114117791110557e-07, 'epoch': 3188.89}
+
 79%|███████▉  | 51660/65536 [8:54:09<2:22:33,  1.62it/s]
 79%|███████▉  | 51661/65536 [8:54:10<2:19:35,  1.66it/s]
 79%|███████▉  | 51662/65536 [8:54:10<2:24:31,  1.60it/s]
 79%|███████▉  | 51663/65536 [8:54:11<2:21:40,  1.63it/s]
 79%|███████▉  | 51664/65536 [8:54:11<2:21:02,  1.64it/s]
 79%|███████▉  | 51665/65536 [8:54:12<2:21:48,  1.63it/s]
 79%|███████▉  | 51666/65536 [8:54:13<2:21:27,  1.63it/s]
 79%|███████▉  | 51667/65536 [8:54:13<2:23:31,  1.61it/s]
 79%|███████▉  | 51668/65536 [8:54:14<2:22:06,  1.63it/s]
 79%|███████▉  | 51669/65536 [8:54:14<2:21:25,  1.63it/s]
 79%|███████▉  | 51670/65536 [8:54:15<2:25:21,  1.59it/s]
 79%|███████▉  | 51671/65536 [8:54:16<2:28:18,  1.56it/s]
 79%|███████▉  | 51672/65536 [8:54:16<2:26:23,  1.58it/s]
 79%|███████▉  | 51673/65536 [8:54:17<2:21:28,  1.63it/s]
 79%|███████▉  | 51674/65536 [8:54:18<2:21:09,  1.64it/s]
 79%|███████▉  | 51675/65536 [8:54:18<2:21:20,  1.63it/s]
 79%|███████▉  | 51676/65536 [8:54:19<2:19:51,  1.65it/s]
 79%|███████▉  | 51677/65536 [8:54:19<2:19:10,  1.66it/s]
 79%|███████▉  | 51678/65536 [8:54:20<2:22:57,  1.62it/s]
 79%|███████▉  | 51679/65536 [8:54:21<2:27:45,  1.56it/s]
 79%|███████▉  | 51680/65536 [8:54:21<2:27:10,  1.57it/s]
                                                         
{'loss': 1.5902, 'learning_rate': 2.908656789518795e-07, 'epoch': 3190.12}
+
 79%|███████▉  | 51680/65536 [8:54:21<2:27:10,  1.57it/s]
 79%|███████▉  | 51681/65536 [8:54:22<2:28:06,  1.56it/s]
 79%|███████▉  | 51682/65536 [8:54:23<2:29:10,  1.55it/s]
 79%|███████▉  | 51683/65536 [8:54:23<2:23:30,  1.61it/s]
 79%|███████▉  | 51684/65536 [8:54:24<2:25:47,  1.58it/s]
 79%|███████▉  | 51685/65536 [8:54:24<2:23:15,  1.61it/s]
 79%|███████▉  | 51686/65536 [8:54:25<2:22:19,  1.62it/s]
 79%|███████▉  | 51687/65536 [8:54:26<2:21:56,  1.63it/s]
 79%|███████▉  | 51688/65536 [8:54:26<2:22:36,  1.62it/s]
 79%|███████▉  | 51689/65536 [8:54:27<2:24:24,  1.60it/s]
 79%|███████▉  | 51690/65536 [8:54:28<2:21:31,  1.63it/s]
 79%|███████▉  | 51691/65536 [8:54:28<2:20:12,  1.65it/s]
 79%|███████▉  | 51692/65536 [8:54:29<2:20:25,  1.64it/s]
 79%|███████▉  | 51693/65536 [8:54:29<2:20:00,  1.65it/s]
 79%|███████▉  | 51694/65536 [8:54:30<2:19:15,  1.66it/s]
 79%|███████▉  | 51695/65536 [8:54:31<2:27:10,  1.57it/s]
 79%|███████▉  | 51696/65536 [8:54:31<2:26:35,  1.57it/s]
 79%|███████▉  | 51697/65536 [8:54:32<2:23:24,  1.61it/s]
 79%|███████▉  | 51698/65536 [8:54:32<2:20:13,  1.64it/s]
 79%|███████▉  | 51699/65536 [8:54:33<2:18:10,  1.67it/s]
 79%|███████▉  | 51700/65536 [8:54:34<2:19:32,  1.65it/s]
                                                         
{'loss': 1.5672, 'learning_rate': 2.9059017999265334e-07, 'epoch': 3191.36}
+
 79%|███████▉  | 51700/65536 [8:54:34<2:19:32,  1.65it/s]
 79%|███████▉  | 51701/65536 [8:54:34<2:18:36,  1.66it/s]
 79%|███████▉  | 51702/65536 [8:54:35<2:20:44,  1.64it/s]
 79%|███████▉  | 51703/65536 [8:54:36<2:21:53,  1.62it/s]
 79%|█���█████▉  | 51704/65536 [8:54:36<2:22:43,  1.62it/s]
 79%|███████▉  | 51705/65536 [8:54:37<2:28:51,  1.55it/s]
 79%|███████▉  | 51706/65536 [8:54:37<2:24:00,  1.60it/s]
 79%|███████▉  | 51707/65536 [8:54:38<2:23:49,  1.60it/s]
 79%|███████▉  | 51708/65536 [8:54:39<2:23:27,  1.61it/s]
 79%|███████▉  | 51709/65536 [8:54:39<2:23:00,  1.61it/s]
 79%|███████▉  | 51710/65536 [8:54:40<2:25:21,  1.59it/s]
 79%|███████▉  | 51711/65536 [8:54:41<2:31:17,  1.52it/s]
 79%|███████▉  | 51712/65536 [8:54:41<2:27:35,  1.56it/s]
 79%|███████▉  | 51713/65536 [8:54:42<2:24:36,  1.59it/s]
 79%|███████▉  | 51714/65536 [8:54:43<2:30:09,  1.53it/s]
 79%|███████▉  | 51715/65536 [8:54:43<2:30:32,  1.53it/s]
 79%|███████▉  | 51716/65536 [8:54:44<2:25:54,  1.58it/s]
 79%|███████▉  | 51717/65536 [8:54:44<2:22:46,  1.61it/s]
 79%|███████▉  | 51718/65536 [8:54:45<2:23:20,  1.61it/s]
 79%|███████▉  | 51719/65536 [8:54:46<2:22:41,  1.61it/s]
 79%|███████▉  | 51720/65536 [8:54:46<2:20:10,  1.64it/s]
                                                         
{'loss': 1.5454, 'learning_rate': 2.903146810334272e-07, 'epoch': 3192.59}
+
 79%|███████▉  | 51720/65536 [8:54:46<2:20:10,  1.64it/s]
 79%|███████▉  | 51721/65536 [8:54:47<2:23:23,  1.61it/s]
 79%|███████▉  | 51722/65536 [8:54:47<2:21:56,  1.62it/s]
 79%|███████▉  | 51723/65536 [8:54:48<2:24:43,  1.59it/s]
 79%|███████▉  | 51724/65536 [8:54:49<2:22:08,  1.62it/s]
 79%|███████▉  | 51725/65536 [8:54:49<2:20:55,  1.63it/s]
 79%|███████▉  | 51726/65536 [8:54:50<2:20:22,  1.64it/s]
 79%|███████▉  | 51727/65536 [8:54:51<2:23:59,  1.60it/s]
 79%|███████▉  | 51728/65536 [8:54:51<2:21:34,  1.63it/s]
 79%|███████▉  | 51729/65536 [8:54:52<2:20:26,  1.64it/s]
 79%|███████▉  | 51730/65536 [8:54:52<2:22:02,  1.62it/s]
 79%|███████▉  | 51731/65536 [8:54:53<2:24:23,  1.59it/s]
 79%|███████▉  | 51732/65536 [8:54:54<2:23:00,  1.61it/s]
 79%|███████▉  | 51733/65536 [8:54:54<2:21:54,  1.62it/s]
 79%|███████▉  | 51734/65536 [8:54:55<2:22:02,  1.62it/s]
 79%|███████▉  | 51735/65536 [8:54:56<2:25:40,  1.58it/s]
 79%|███████▉  | 51736/65536 [8:54:56<2:23:50,  1.60it/s]
 79%|███████▉  | 51737/65536 [8:54:57<2:20:46,  1.63it/s]
 79%|███████▉  | 51738/65536 [8:54:57<2:21:43,  1.62it/s]
 79%|███████▉  | 51739/65536 [8:54:58<2:22:45,  1.61it/s]
 79%|███████▉  | 51740/65536 [8:54:59<2:22:31,  1.61it/s]
                                                         
{'loss': 1.568, 'learning_rate': 2.9003918207420106e-07, 'epoch': 3193.83}
+
 79%|███████▉  | 51740/65536 [8:54:59<2:22:31,  1.61it/s]
 79%|███████▉  | 51741/65536 [8:54:59<2:21:25,  1.63it/s]
 79%|███████▉  | 51742/65536 [8:55:00<2:21:51,  1.62it/s]
 79%|███████▉  | 51743/65536 [8:55:01<2:25:12,  1.58it/s]
 79%|███████▉  | 51744/65536 [8:55:01<2:23:47,  1.60it/s]
 79%|███████▉  | 51745/65536 [8:55:02<2:19:49,  1.64it/s]
 79%|███████▉  | 51746/65536 [8:55:02<2:19:46,  1.64it/s]
 79%|███████▉  | 51747/65536 [8:55:03<2:17:37,  1.67it/s]
 79%|███████▉  | 51748/65536 [8:55:03<2:17:24,  1.67it/s]
 79%|███████▉  | 51749/65536 [8:55:04<2:21:32,  1.62it/s]
 79%|███████▉  | 51750/65536 [8:55:05<2:20:57,  1.63it/s]
 79%|███████▉  | 51751/65536 [8:55:05<2:25:10,  1.58it/s]
 79%|███████▉  | 51752/65536 [8:55:06<2:24:29,  1.59it/s]
 79%|███████▉  | 51753/65536 [8:55:07<2:25:53,  1.57it/s]
 79%|███████▉  | 51754/65536 [8:55:07<2:25:29,  1.58it/s]
 79%|███████▉  | 51755/65536 [8:55:08<2:25:53,  1.57it/s]
 79%|███████▉  | 51756/65536 [8:55:09<2:24:16,  1.59it/s]
 79%|███████▉  | 51757/65536 [8:55:09<2:22:00,  1.62it/s]
 79%|███████▉  | 51758/65536 [8:55:10<2:21:25,  1.62it/s]
 79%|███████▉  | 51759/65536 [8:55:10<2:22:15,  1.61it/s]
 79%|███████▉  | 51760/65536 [8:55:11<2:26:27,  1.57it/s]
                                                         
{'loss': 1.5765, 'learning_rate': 2.897636831149748e-07, 'epoch': 3195.06}
+
 79%|███████▉  | 51760/65536 [8:55:11<2:26:27,  1.57it/s]
 79%|███████▉  | 51761/65536 [8:55:12<2:26:00,  1.57it/s]
 79%|███████▉  | 51762/65536 [8:55:12<2:31:07,  1.52it/s]
 79%|███████▉  | 51763/65536 [8:55:13<2:27:18,  1.56it/s]
 79%|███████▉  | 51764/65536 [8:55:14<2:26:35,  1.57it/s]
 79%|███████▉  | 51765/65536 [8:55:14<2:24:12,  1.59it/s]
 79%|███████▉  | 51766/65536 [8:55:15<2:23:12,  1.60it/s]
 79%|███████▉  | 51767/65536 [8:55:15<2:22:49,  1.61it/s]
 79%|███████▉  | 51768/65536 [8:55:16<2:22:26,  1.61it/s]
 79%|███████▉  | 51769/65536 [8:55:17<2:22:25,  1.61it/s]
 79%|███████▉  | 51770/65536 [8:55:17<2:20:07,  1.64it/s]
 79%|███████▉  | 51771/65536 [8:55:18<2:18:38,  1.65it/s]
 79%|███████▉  | 51772/65536 [8:55:19<2:19:10,  1.65it/s]
 79%|███████▉  | 51773/65536 [8:55:19<2:15:40,  1.69it/s]
 79%|███████▉  | 51774/65536 [8:55:20<2:15:35,  1.69it/s]
 79%|███████▉  | 51775/65536 [8:55:20<2:16:41,  1.68it/s]
 79%|███████▉  | 51776/65536 [8:55:21<2:25:53,  1.57it/s]
 79%|███████▉  | 51777/65536 [8:55:22<2:22:40,  1.61it/s]
 79%|███████▉  | 51778/65536 [8:55:22<2:20:24,  1.63it/s]
 79%|███████▉  | 51779/65536 [8:55:23<2:19:28,  1.64it/s]
 79%|███████▉  | 51780/65536 [8:55:23<2:21:39,  1.62it/s]
                                                         
{'loss': 1.5516, 'learning_rate': 2.8948818415574867e-07, 'epoch': 3196.3}
+
 79%|███████▉  | 51780/65536 [8:55:23<2:21:39,  1.62it/s]
 79%|███████▉  | 51781/65536 [8:55:24<2:21:07,  1.62it/s]
 79%|███████▉  | 51782/65536 [8:55:25<2:19:42,  1.64it/s]
 79%|███████▉  | 51783/65536 [8:55:25<2:22:21,  1.61it/s]
 79%|███████▉  | 51784/65536 [8:55:26<2:23:03,  1.60it/s]
 79%|███████▉  | 51785/65536 [8:55:27<2:20:58,  1.63it/s]
 79%|███████▉  | 51786/65536 [8:55:27<2:19:43,  1.64it/s]
 79%|███████▉  | 51787/65536 [8:55:28<2:20:47,  1.63it/s]
 79%|███████▉  | 51788/65536 [8:55:28<2:20:22,  1.63it/s]
 79%|███████▉  | 51789/65536 [8:55:29<2:20:57,  1.63it/s]
 79%|███████▉  | 51790/65536 [8:55:30<2:20:21,  1.63it/s]
 79%|███████▉  | 51791/65536 [8:55:30<2:22:07,  1.61it/s]
 79%|███████▉  | 51792/65536 [8:55:31<2:25:55,  1.57it/s]
 79%|███████▉  | 51793/65536 [8:55:32<2:31:06,  1.52it/s]
 79%|███████▉  | 51794/65536 [8:55:32<2:27:06,  1.56it/s]
 79%|███████▉  | 51795/65536 [8:55:33<2:27:14,  1.56it/s]
 79%|███████▉  | 51796/65536 [8:55:33<2:27:36,  1.55it/s]
 79%|███████▉  | 51797/65536 [8:55:34<2:22:51,  1.60it/s]
 79%|███████▉  | 51798/65536 [8:55:35<2:22:49,  1.60it/s]
 79%|███████▉  | 51799/65536 [8:55:35<2:21:49,  1.61it/s]
 79%|███████▉  | 51800/65536 [8:55:36<2:21:49,  1.61it/s]
                                                         
{'loss': 1.5751, 'learning_rate': 2.8921268519652253e-07, 'epoch': 3197.53}
+
 79%|███████▉  | 51800/65536 [8:55:36<2:21:49,  1.61it/s]
 79%|███████▉  | 51801/65536 [8:55:37<2:20:39,  1.63it/s]
 79%|███████▉  | 51802/65536 [8:55:37<2:19:30,  1.64it/s]
 79%|███████▉  | 51803/65536 [8:55:38<2:18:35,  1.65it/s]
 79%|███████▉  | 51804/65536 [8:55:38<2:18:43,  1.65it/s]
 79%|███████▉  | 51805/65536 [8:55:39<2:20:28,  1.63it/s]
 79%|███████▉  | 51806/65536 [8:55:40<2:19:46,  1.64it/s]
 79%|███████▉  | 51807/65536 [8:55:40<2:17:16,  1.67it/s]
 79%|███████▉  | 51808/65536 [8:55:41<2:23:23,  1.60it/s]
 79%|███████▉  | 51809/65536 [8:55:41<2:21:52,  1.61it/s]
 79%|███████▉  | 51810/65536 [8:55:42<2:27:09,  1.55it/s]
 79%|███████▉  | 51811/65536 [8:55:43<2:23:19,  1.60it/s]
 79%|███████▉  | 51812/65536 [8:55:43<2:25:04,  1.58it/s]
 79%|███████▉  | 51813/65536 [8:55:44<2:23:43,  1.59it/s]
 79%|███████▉  | 51814/65536 [8:55:45<2:21:32,  1.62it/s]
 79%|███████▉  | 51815/65536 [8:55:45<2:21:20,  1.62it/s]
 79%|███████▉  | 51816/65536 [8:55:46<2:22:43,  1.60it/s]
 79%|███████▉  | 51817/65536 [8:55:46<2:20:40,  1.63it/s]
 79%|███████▉  | 51818/65536 [8:55:47<2:16:55,  1.67it/s]
 79%|███████▉  | 51819/65536 [8:55:48<2:15:11,  1.69it/s]
 79%|███████▉  | 51820/65536 [8:55:48<2:13:56,  1.71it/s]
                                                         
{'loss': 1.6064, 'learning_rate': 2.889371862372964e-07, 'epoch': 3198.77}
+
 79%|███████▉  | 51820/65536 [8:55:48<2:13:56,  1.71it/s]
 79%|███████▉  | 51821/65536 [8:55:49<2:16:03,  1.68it/s]
 79%|███████▉  | 51822/65536 [8:55:49<2:19:45,  1.64it/s]
 79%|███████▉  | 51823/65536 [8:55:50<2:23:17,  1.59it/s]
 79%|███████▉  | 51824/65536 [8:55:51<2:25:55,  1.57it/s]
 79%|███████▉  | 51825/65536 [8:55:51<2:26:00,  1.57it/s]
 79%|███████▉  | 51826/65536 [8:55:52<2:25:49,  1.57it/s]
 79%|███████▉  | 51827/65536 [8:55:53<2:21:20,  1.62it/s]
 79%|███████▉  | 51828/65536 [8:55:53<2:23:38,  1.59it/s]
 79%|███████▉  | 51829/65536 [8:55:54<2:21:41,  1.61it/s]
 79%|███████▉  | 51830/65536 [8:55:54<2:21:04,  1.62it/s]
 79%|███████▉  | 51831/65536 [8:55:55<2:22:15,  1.61it/s]
 79%|███████▉  | 51832/65536 [8:55:56<2:22:02,  1.61it/s]
 79%|███████▉  | 51833/65536 [8:55:56<2:21:36,  1.61it/s]
 79%|███████▉  | 51834/65536 [8:55:57<2:23:50,  1.59it/s]
 79%|███████▉  | 51835/65536 [8:55:58<2:21:25,  1.61it/s]
 79%|███████▉  | 51836/65536 [8:55:58<2:18:09,  1.65it/s]
 79%|███████▉  | 51837/65536 [8:55:59<2:18:36,  1.65it/s]
 79%|███████▉  | 51838/65536 [8:55:59<2:19:55,  1.63it/s]
 79%|███████▉  | 51839/65536 [8:56:00<2:20:56,  1.62it/s]
 79%|███████▉  | 51840/65536 [8:56:01<2:20:12,  1.63it/s]
                                                         
{'loss': 1.5567, 'learning_rate': 2.886616872780703e-07, 'epoch': 3200.0}
+
 79%|███████▉  | 51840/65536 [8:56:01<2:20:12,  1.63it/s]
 79%|███████▉  | 51841/65536 [8:56:01<2:25:54,  1.56it/s]
 79%|███████▉  | 51842/65536 [8:56:02<2:23:48,  1.59it/s]
 79%|███████▉  | 51843/65536 [8:56:02<2:21:27,  1.61it/s]
 79%|███████▉  | 51844/65536 [8:56:03<2:22:47,  1.60it/s]
 79%|███████▉  | 51845/65536 [8:56:04<2:21:19,  1.61it/s]
 79%|███████▉  | 51846/65536 [8:56:04<2:22:51,  1.60it/s]
 79%|███████▉  | 51847/65536 [8:56:05<2:19:34,  1.63it/s]
 79%|███████▉  | 51848/65536 [8:56:06<2:22:13,  1.60it/s]
 79%|███████▉  | 51849/65536 [8:56:06<2:22:34,  1.60it/s]
 79%|███████▉  | 51850/65536 [8:56:07<2:18:08,  1.65it/s]
 79%|███████▉  | 51851/65536 [8:56:07<2:19:14,  1.64it/s]
 79%|███████▉  | 51852/65536 [8:56:08<2:17:25,  1.66it/s]
 79%|███████▉  | 51853/65536 [8:56:09<2:15:42,  1.68it/s]
 79%|███████▉  | 51854/65536 [8:56:09<2:17:53,  1.65it/s]
 79%|███████▉  | 51855/65536 [8:56:10<2:19:18,  1.64it/s]
 79%|███████▉  | 51856/65536 [8:56:10<2:19:50,  1.63it/s]
 79%|███████▉  | 51857/65536 [8:56:11<2:26:57,  1.55it/s]
 79%|███████▉  | 51858/65536 [8:56:12<2:24:09,  1.58it/s]
 79%|███████▉  | 51859/65536 [8:56:12<2:23:51,  1.58it/s]
 79%|███████▉  | 51860/65536 [8:56:13<2:24:42,  1.58it/s]
                                                         
{'loss': 1.5783, 'learning_rate': 2.8838618831884416e-07, 'epoch': 3201.23}
+
 79%|███████▉  | 51860/65536 [8:56:13<2:24:42,  1.58it/s]
 79%|███████▉  | 51861/65536 [8:56:14<2:24:28,  1.58it/s]
 79%|███████▉  | 51862/65536 [8:56:14<2:24:07,  1.58it/s]
 79%|███████▉  | 51863/65536 [8:56:15<2:20:58,  1.62it/s]
 79%|███████▉  | 51864/65536 [8:56:16<2:20:26,  1.62it/s]
 79%|███████▉  | 51865/65536 [8:56:16<2:18:47,  1.64it/s]
 79%|███████▉  | 51866/65536 [8:56:17<2:19:37,  1.63it/s]
 79%|███████▉  | 51867/65536 [8:56:17<2:20:16,  1.62it/s]
 79%|███████▉  | 51868/65536 [8:56:18<2:19:24,  1.63it/s]
 79%|███████▉  | 51869/65536 [8:56:19<2:21:00,  1.62it/s]
 79%|███████▉  | 51870/65536 [8:56:19<2:22:18,  1.60it/s]
 79%|███████▉  | 51871/65536 [8:56:20<2:21:44,  1.61it/s]
 79%|███████▉  | 51872/65536 [8:56:20<2:19:45,  1.63it/s]
 79%|███████▉  | 51873/65536 [8:56:21<2:24:57,  1.57it/s]
 79%|███████▉  | 51874/65536 [8:56:22<2:23:01,  1.59it/s]
 79%|███████▉  | 51875/65536 [8:56:22<2:21:05,  1.61it/s]
 79%|███████▉  | 51876/65536 [8:56:23<2:21:36,  1.61it/s]
 79%|███████▉  | 51877/65536 [8:56:24<2:19:52,  1.63it/s]
 79%|███████▉  | 51878/65536 [8:56:24<2:18:47,  1.64it/s]
 79%|███████▉  | 51879/65536 [8:56:25<2:18:17,  1.65it/s]
 79%|███████▉  | 51880/65536 [8:56:25<2:19:26,  1.63it/s]
                                                         
{'loss': 1.5631, 'learning_rate': 2.88110689359618e-07, 'epoch': 3202.47}
+
 79%|███████▉  | 51880/65536 [8:56:25<2:19:26,  1.63it/s]
 79%|███████▉  | 51881/65536 [8:56:26<2:19:53,  1.63it/s]
 79%|███████▉  | 51882/65536 [8:56:27<2:22:15,  1.60it/s]
 79%|███████▉  | 51883/65536 [8:56:27<2:23:56,  1.58it/s]
 79%|███████▉  | 51884/65536 [8:56:28<2:22:11,  1.60it/s]
 79%|███████▉  | 51885/65536 [8:56:28<2:19:10,  1.63it/s]
 79%|███████▉  | 51886/65536 [8:56:29<2:21:11,  1.61it/s]
 79%|███████▉  | 51887/65536 [8:56:30<2:20:26,  1.62it/s]
 79%|███████▉  | 51888/65536 [8:56:30<2:21:17,  1.61it/s]
 79%|███████▉  | 51889/65536 [8:56:31<2:25:27,  1.56it/s]
 79%|███████▉  | 51890/65536 [8:56:32<2:21:13,  1.61it/s]
 79%|███████▉  | 51891/65536 [8:56:32<2:18:21,  1.64it/s]
 79%|███████▉  | 51892/65536 [8:56:33<2:21:31,  1.61it/s]
 79%|███████▉  | 51893/65536 [8:56:34<2:25:34,  1.56it/s]
 79%|███████▉  | 51894/65536 [8:56:34<2:23:39,  1.58it/s]
 79%|███████▉  | 51895/65536 [8:56:35<2:20:44,  1.62it/s]
 79%|███████▉  | 51896/65536 [8:56:35<2:18:46,  1.64it/s]
 79%|███████▉  | 51897/65536 [8:56:36<2:21:57,  1.60it/s]
 79%|███████▉  | 51898/65536 [8:56:37<2:23:32,  1.58it/s]
 79%|███████▉  | 51899/65536 [8:56:37<2:20:04,  1.62it/s]
 79%|███████▉  | 51900/65536 [8:56:38<2:20:33,  1.62it/s]
                                                         
{'loss': 1.531, 'learning_rate': 2.8783519040039177e-07, 'epoch': 3203.7}
+
 79%|███████▉  | 51900/65536 [8:56:38<2:20:33,  1.62it/s]
 79%|███████▉  | 51901/65536 [8:56:38<2:20:38,  1.62it/s]
 79%|███████▉  | 51902/65536 [8:56:39<2:19:42,  1.63it/s]
 79%|███████▉  | 51903/65536 [8:56:40<2:18:41,  1.64it/s]
 79%|███████▉  | 51904/65536 [8:56:40<2:21:47,  1.60it/s]
 79%|███████▉  | 51905/65536 [8:56:41<2:27:29,  1.54it/s]
 79%|███████▉  | 51906/65536 [8:56:42<2:26:53,  1.55it/s]
 79%|███████▉  | 51907/65536 [8:56:42<2:23:41,  1.58it/s]
 79%|███████▉  | 51908/65536 [8:56:43<2:24:22,  1.57it/s]
 79%|███████▉  | 51909/65536 [8:56:44<2:25:14,  1.56it/s]
 79%|███████▉  | 51910/65536 [8:56:44<2:26:15,  1.55it/s]
 79%|███████▉  | 51911/65536 [8:56:45<2:22:46,  1.59it/s]
 79%|███████▉  | 51912/65536 [8:56:45<2:21:01,  1.61it/s]
 79%|███████▉  | 51913/65536 [8:56:46<2:20:08,  1.62it/s]
 79%|███████▉  | 51914/65536 [8:56:47<2:18:32,  1.64it/s]
 79%|███████▉  | 51915/65536 [8:56:47<2:20:13,  1.62it/s]
 79%|███████▉  | 51916/65536 [8:56:48<2:19:08,  1.63it/s]
 79%|███████▉  | 51917/65536 [8:56:49<2:22:01,  1.60it/s]
 79%|███████▉  | 51918/65536 [8:56:49<2:20:21,  1.62it/s]
 79%|███████▉  | 51919/65536 [8:56:50<2:19:17,  1.63it/s]
 79%|███████▉  | 51920/65536 [8:56:50<2:18:32,  1.64it/s]
                                                         
{'loss': 1.5749, 'learning_rate': 2.8755969144116563e-07, 'epoch': 3204.94}
+
 79%|███████▉  | 51920/65536 [8:56:50<2:18:32,  1.64it/s]
 79%|███████▉  | 51921/65536 [8:56:51<2:17:31,  1.65it/s]
 79%|███████▉  | 51922/65536 [8:56:52<2:21:39,  1.60it/s]
 79%|███████▉  | 51923/65536 [8:56:52<2:20:03,  1.62it/s]
 79%|███████▉  | 51924/65536 [8:56:53<2:20:14,  1.62it/s]
 79%|███████▉  | 51925/65536 [8:56:53<2:20:40,  1.61it/s]
 79%|███████▉  | 51926/65536 [8:56:54<2:22:17,  1.59it/s]
 79%|███████▉  | 51927/65536 [8:56:55<2:20:42,  1.61it/s]
 79%|███████▉  | 51928/65536 [8:56:55<2:22:08,  1.60it/s]
 79%|███████▉  | 51929/65536 [8:56:56<2:21:16,  1.61it/s]
 79%|███████▉  | 51930/65536 [8:56:57<2:18:43,  1.63it/s]
 79%|███████▉  | 51931/65536 [8:56:57<2:17:42,  1.65it/s]
 79%|███████▉  | 51932/65536 [8:56:58<2:15:17,  1.68it/s]
 79%|███████▉  | 51933/65536 [8:56:58<2:15:50,  1.67it/s]
 79%|███████▉  | 51934/65536 [8:56:59<2:17:09,  1.65it/s]
 79%|███████▉  | 51935/65536 [8:57:00<2:21:49,  1.60it/s]
 79%|███████▉  | 51936/65536 [8:57:00<2:18:58,  1.63it/s]
 79%|███████▉  | 51937/65536 [8:57:01<2:21:53,  1.60it/s]
 79%|███████▉  | 51938/65536 [8:57:02<2:25:38,  1.56it/s]
 79%|███████▉  | 51939/65536 [8:57:02<2:20:38,  1.61it/s]
 79%|███████▉  | 51940/65536 [8:57:03<2:19:18,  1.63it/s]
                                                         
{'loss': 1.5879, 'learning_rate': 2.872841924819395e-07, 'epoch': 3206.17}
+
 79%|███████▉  | 51940/65536 [8:57:03<2:19:18,  1.63it/s]
 79%|███████▉  | 51941/65536 [8:57:03<2:19:41,  1.62it/s]
 79%|███████▉  | 51942/65536 [8:57:04<2:17:20,  1.65it/s]
 79%|███████▉  | 51943/65536 [8:57:04<2:17:41,  1.65it/s]
 79%|███████▉  | 51944/65536 [8:57:05<2:18:44,  1.63it/s]
 79%|███████▉  | 51945/65536 [8:57:06<2:22:05,  1.59it/s]
 79%|███████▉  | 51946/65536 [8:57:06<2:23:45,  1.58it/s]
 79%|███████▉  | 51947/65536 [8:57:07<2:22:34,  1.59it/s]
 79%|███████▉  | 51948/65536 [8:57:08<2:21:25,  1.60it/s]
 79%|███████▉  | 51949/65536 [8:57:08<2:21:40,  1.60it/s]
 79%|███████▉  | 51950/65536 [8:57:09<2:17:45,  1.64it/s]
 79%|███████▉  | 51951/65536 [8:57:09<2:16:50,  1.65it/s]
 79%|███████▉  | 51952/65536 [8:57:10<2:19:54,  1.62it/s]
 79%|███████▉  | 51953/65536 [8:57:11<2:19:04,  1.63it/s]
 79%|███████▉  | 51954/65536 [8:57:11<2:23:27,  1.58it/s]
 79%|███████▉  | 51955/65536 [8:57:12<2:21:49,  1.60it/s]
 79%|███████▉  | 51956/65536 [8:57:13<2:21:35,  1.60it/s]
 79%|███████▉  | 51957/65536 [8:57:13<2:22:27,  1.59it/s]
 79%|███████▉  | 51958/65536 [8:57:14<2:23:49,  1.57it/s]
 79%|███████▉  | 51959/65536 [8:57:15<2:23:45,  1.57it/s]
 79%|███████▉  | 51960/65536 [8:57:15<2:28:56,  1.52it/s]
                                                         
{'loss': 1.5604, 'learning_rate': 2.8700869352271335e-07, 'epoch': 3207.41}
+
 79%|███████▉  | 51960/65536 [8:57:15<2:28:56,  1.52it/s]
 79%|███████▉  | 51961/65536 [8:57:16<2:31:33,  1.49it/s]
 79%|███████▉  | 51962/65536 [8:57:17<2:29:22,  1.51it/s]
 79%|███████▉  | 51963/65536 [8:57:17<2:27:07,  1.54it/s]
 79%|███████▉  | 51964/65536 [8:57:18<2:23:39,  1.57it/s]
 79%|███████▉  | 51965/65536 [8:57:18<2:24:07,  1.57it/s]
 79%|███████▉  | 51966/65536 [8:57:19<2:21:21,  1.60it/s]
 79%|███████▉  | 51967/65536 [8:57:20<2:18:58,  1.63it/s]
 79%|███████▉  | 51968/65536 [8:57:20<2:20:33,  1.61it/s]
 79%|███████▉  | 51969/65536 [8:57:21<2:20:11,  1.61it/s]
 79%|███████▉  | 51970/65536 [8:57:22<2:23:02,  1.58it/s]
 79%|███████▉  | 51971/65536 [8:57:22<2:22:23,  1.59it/s]
 79%|███████▉  | 51972/65536 [8:57:23<2:21:42,  1.60it/s]
 79%|███████▉  | 51973/65536 [8:57:23<2:21:14,  1.60it/s]
 79%|███████▉  | 51974/65536 [8:57:24<2:21:14,  1.60it/s]
 79%|███████▉  | 51975/65536 [8:57:25<2:18:24,  1.63it/s]
 79%|███████▉  | 51976/65536 [8:57:25<2:18:24,  1.63it/s]
 79%|███████▉  | 51977/65536 [8:57:26<2:21:50,  1.59it/s]
 79%|███████▉  | 51978/65536 [8:57:27<2:19:47,  1.62it/s]
 79%|███████▉  | 51979/65536 [8:57:27<2:16:53,  1.65it/s]
 79%|███████▉  | 51980/65536 [8:57:28<2:17:45,  1.64it/s]
                                                         
{'loss': 1.5599, 'learning_rate': 2.867331945634872e-07, 'epoch': 3208.64}
+
 79%|███████▉  | 51980/65536 [8:57:28<2:17:45,  1.64it/s]
 79%|███████▉  | 51981/65536 [8:57:28<2:20:36,  1.61it/s]
 79%|███████▉  | 51982/65536 [8:57:29<2:20:17,  1.61it/s]
 79%|███████▉  | 51983/65536 [8:57:30<2:17:41,  1.64it/s]
 79%|███████▉  | 51984/65536 [8:57:30<2:13:55,  1.69it/s]
 79%|███████▉  | 51985/65536 [8:57:31<2:15:13,  1.67it/s]
 79%|███████▉  | 51986/65536 [8:57:31<2:19:24,  1.62it/s]
 79%|███████▉  | 51987/65536 [8:57:32<2:17:45,  1.64it/s]
 79%|███████▉  | 51988/65536 [8:57:33<2:15:56,  1.66it/s]
 79%|███████▉  | 51989/65536 [8:57:33<2:15:52,  1.66it/s]
 79%|███████▉  | 51990/65536 [8:57:34<2:17:42,  1.64it/s]
 79%|███████▉  | 51991/65536 [8:57:34<2:17:11,  1.65it/s]
 79%|███████▉  | 51992/65536 [8:57:35<2:14:48,  1.67it/s]
 79%|███████▉  | 51993/65536 [8:57:36<2:18:02,  1.64it/s]
 79%|███████▉  | 51994/65536 [8:57:36<2:16:06,  1.66it/s]
 79%|███████▉  | 51995/65536 [8:57:37<2:17:03,  1.65it/s]
 79%|███████▉  | 51996/65536 [8:57:37<2:15:36,  1.66it/s]
 79%|███████▉  | 51997/65536 [8:57:38<2:18:02,  1.63it/s]
 79%|███████▉  | 51998/65536 [8:57:39<2:17:29,  1.64it/s]
 79%|███████▉  | 51999/65536 [8:57:39<2:17:56,  1.64it/s]
 79%|███████▉  | 52000/65536 [8:57:40<2:22:59,  1.58it/s]
                                                         
{'loss': 1.5697, 'learning_rate': 2.8645769560426106e-07, 'epoch': 3209.88}
+
 79%|███████▉  | 52000/65536 [8:57:40<2:22:59,  1.58it/s]
 79%|███████▉  | 52001/65536 [8:57:41<2:23:30,  1.57it/s]
 79%|███████▉  | 52002/65536 [8:57:41<2:23:57,  1.57it/s]
 79%|███████▉  | 52003/65536 [8:57:42<2:27:56,  1.52it/s]
 79%|███████▉  | 52004/65536 [8:57:43<2:27:12,  1.53it/s]
 79%|███████▉  | 52005/65536 [8:57:43<2:23:12,  1.57it/s]
 79%|███████▉  | 52006/65536 [8:57:44<2:23:28,  1.57it/s]
 79%|███████▉  | 52007/65536 [8:57:44<2:20:37,  1.60it/s]
 79%|███████▉  | 52008/65536 [8:57:45<2:15:09,  1.67it/s]
 79%|███████▉  | 52009/65536 [8:57:46<2:17:08,  1.64it/s]
 79%|███████▉  | 52010/65536 [8:57:46<2:17:22,  1.64it/s]
 79%|███████▉  | 52011/65536 [8:57:47<2:15:10,  1.67it/s]
 79%|███████▉  | 52012/65536 [8:57:47<2:19:39,  1.61it/s]
 79%|███████▉  | 52013/65536 [8:57:48<2:18:36,  1.63it/s]
 79%|███████▉  | 52014/65536 [8:57:49<2:18:33,  1.63it/s]
 79%|███████▉  | 52015/65536 [8:57:49<2:21:36,  1.59it/s]
 79%|███████▉  | 52016/65536 [8:57:50<2:22:05,  1.59it/s]
 79%|███████▉  | 52017/65536 [8:57:51<2:21:06,  1.60it/s]
 79%|███████▉  | 52018/65536 [8:57:51<2:23:04,  1.57it/s]
 79%|███████▉  | 52019/65536 [8:57:52<2:27:59,  1.52it/s]
 79%|███████▉  | 52020/65536 [8:57:53<2:24:43,  1.56it/s]
                                                         
{'loss': 1.5664, 'learning_rate': 2.8618219664503487e-07, 'epoch': 3211.11}
+
 79%|███████▉  | 52020/65536 [8:57:53<2:24:43,  1.56it/s]
 79%|███████▉  | 52021/65536 [8:57:53<2:21:50,  1.59it/s]
 79%|███████▉  | 52022/65536 [8:57:54<2:20:19,  1.61it/s]
 79%|███████▉  | 52023/65536 [8:57:54<2:20:45,  1.60it/s]
 79%|███████▉  | 52024/65536 [8:57:55<2:16:58,  1.64it/s]
 79%|███████▉  | 52025/65536 [8:57:56<2:16:09,  1.65it/s]
 79%|███████▉  | 52026/65536 [8:57:56<2:18:35,  1.62it/s]
 79%|███████▉  | 52027/65536 [8:57:57<2:17:10,  1.64it/s]
 79%|███████▉  | 52028/65536 [8:57:57<2:18:42,  1.62it/s]
 79%|███████▉  | 52029/65536 [8:57:58<2:18:31,  1.63it/s]
 79%|███████▉  | 52030/65536 [8:57:59<2:22:27,  1.58it/s]
 79%|███████▉  | 52031/65536 [8:57:59<2:21:05,  1.60it/s]
 79%|███████▉  | 52032/65536 [8:58:00<2:20:05,  1.61it/s]
 79%|███████▉  | 52033/65536 [8:58:01<2:19:00,  1.62it/s]
 79%|███████▉  | 52034/65536 [8:58:01<2:20:33,  1.60it/s]
 79%|███████▉  | 52035/65536 [8:58:02<2:21:39,  1.59it/s]
 79%|███████▉  | 52036/65536 [8:58:02<2:24:23,  1.56it/s]
 79%|███████▉  | 52037/65536 [8:58:03<2:22:45,  1.58it/s]
 79%|███████▉  | 52038/65536 [8:58:04<2:21:52,  1.59it/s]
 79%|███████▉  | 52039/65536 [8:58:04<2:21:26,  1.59it/s]
 79%|███████▉  | 52040/65536 [8:58:05<2:18:03,  1.63it/s]
                                                         
{'loss': 1.5831, 'learning_rate': 2.8590669768580873e-07, 'epoch': 3212.35}
+
 79%|███████▉  | 52040/65536 [8:58:05<2:18:03,  1.63it/s]
 79%|███████▉  | 52041/65536 [8:58:05<2:16:32,  1.65it/s]
 79%|███████▉  | 52042/65536 [8:58:06<2:15:48,  1.66it/s]
 79%|███████▉  | 52043/65536 [8:58:07<2:14:28,  1.67it/s]
 79%|███████▉  | 52044/65536 [8:58:07<2:19:59,  1.61it/s]
 79%|███████▉  | 52045/65536 [8:58:08<2:18:33,  1.62it/s]
 79%|███████▉  | 52046/65536 [8:58:09<2:20:48,  1.60it/s]
 79%|███████▉  | 52047/65536 [8:58:09<2:19:37,  1.61it/s]
 79%|███████▉  | 52048/65536 [8:58:10<2:18:04,  1.63it/s]
 79%|███████▉  | 52049/65536 [8:58:10<2:17:09,  1.64it/s]
 79%|███████▉  | 52050/65536 [8:58:11<2:17:02,  1.64it/s]
 79%|███████▉  | 52051/65536 [8:58:12<2:20:48,  1.60it/s]
 79%|███████▉  | 52052/65536 [8:58:12<2:22:45,  1.57it/s]
 79%|███████▉  | 52053/65536 [8:58:13<2:20:27,  1.60it/s]
 79%|███████▉  | 52054/65536 [8:58:14<2:21:44,  1.59it/s]
 79%|███████▉  | 52055/65536 [8:58:14<2:22:32,  1.58it/s]
 79%|███████▉  | 52056/65536 [8:58:15<2:19:27,  1.61it/s]
 79%|███████▉  | 52057/65536 [8:58:15<2:22:07,  1.58it/s]
 79%|███████▉  | 52058/65536 [8:58:16<2:18:12,  1.63it/s]
 79%|███████▉  | 52059/65536 [8:58:17<2:17:10,  1.64it/s]
 79%|███████▉  | 52060/65536 [8:58:17<2:17:06,  1.64it/s]
                                                         
{'loss': 1.5901, 'learning_rate': 2.856311987265826e-07, 'epoch': 3213.58}
+
 79%|███████▉  | 52060/65536 [8:58:17<2:17:06,  1.64it/s]
 79%|███████▉  | 52061/65536 [8:58:18<2:16:27,  1.65it/s]
 79%|███████▉  | 52062/65536 [8:58:18<2:15:16,  1.66it/s]
 79%|███████▉  | 52063/65536 [8:58:19<2:15:42,  1.65it/s]
 79%|███████▉  | 52064/65536 [8:58:20<2:17:26,  1.63it/s]
 79%|███████▉  | 52065/65536 [8:58:20<2:18:15,  1.62it/s]
 79%|███████▉  | 52066/65536 [8:58:21<2:18:06,  1.63it/s]
 79%|███████▉  | 52067/65536 [8:58:22<2:25:49,  1.54it/s]
 79%|███████���  | 52068/65536 [8:58:22<2:25:01,  1.55it/s]
 79%|███████▉  | 52069/65536 [8:58:23<2:24:54,  1.55it/s]
 79%|███████▉  | 52070/65536 [8:58:24<2:21:43,  1.58it/s]
 79%|███████▉  | 52071/65536 [8:58:24<2:21:26,  1.59it/s]
 79%|███████▉  | 52072/65536 [8:58:25<2:18:02,  1.63it/s]
 79%|███████▉  | 52073/65536 [8:58:25<2:18:51,  1.62it/s]
 79%|███████▉  | 52074/65536 [8:58:26<2:20:15,  1.60it/s]
 79%|███████▉  | 52075/65536 [8:58:27<2:22:35,  1.57it/s]
 79%|███████▉  | 52076/65536 [8:58:27<2:20:31,  1.60it/s]
 79%|███████▉  | 52077/65536 [8:58:28<2:18:02,  1.62it/s]
 79%|███████▉  | 52078/65536 [8:58:29<2:21:23,  1.59it/s]
 79%|███████▉  | 52079/65536 [8:58:29<2:22:51,  1.57it/s]
 79%|███████▉  | 52080/65536 [8:58:30<2:20:01,  1.60it/s]
                                                         
{'loss': 1.5426, 'learning_rate': 2.8535569976735645e-07, 'epoch': 3214.81}
+
 79%|███████▉  | 52080/65536 [8:58:30<2:20:01,  1.60it/s]
 79%|███████▉  | 52081/65536 [8:58:30<2:18:40,  1.62it/s]
 79%|███████▉  | 52082/65536 [8:58:31<2:16:12,  1.65it/s]
 79%|███████▉  | 52083/65536 [8:58:32<2:16:09,  1.65it/s]
 79%|███████▉  | 52084/65536 [8:58:32<2:19:29,  1.61it/s]
 79%|███████▉  | 52085/65536 [8:58:33<2:19:04,  1.61it/s]
 79%|███████▉  | 52086/65536 [8:58:33<2:17:13,  1.63it/s]
 79%|███████▉  | 52087/65536 [8:58:34<2:18:05,  1.62it/s]
 79%|███████▉  | 52088/65536 [8:58:35<2:17:37,  1.63it/s]
 79%|███████▉  | 52089/65536 [8:58:35<2:18:06,  1.62it/s]
 79%|███████▉  | 52090/65536 [8:58:36<2:19:29,  1.61it/s]
 79%|███████▉  | 52091/65536 [8:58:37<2:16:37,  1.64it/s]
 79%|███████▉  | 52092/65536 [8:58:37<2:17:29,  1.63it/s]
 79%|███████▉  | 52093/65536 [8:58:38<2:16:44,  1.64it/s]
 79%|███████▉  | 52094/65536 [8:58:38<2:20:16,  1.60it/s]
 79%|███████▉  | 52095/65536 [8:58:39<2:20:13,  1.60it/s]
 79%|███████▉  | 52096/65536 [8:58:40<2:18:15,  1.62it/s]
 79%|███████▉  | 52097/65536 [8:58:40<2:16:59,  1.63it/s]
 79%|███████▉  | 52098/65536 [8:58:41<2:18:09,  1.62it/s]
 79%|███████▉  | 52099/65536 [8:58:41<2:19:33,  1.60it/s]
 79%|███████▉  | 52100/65536 [8:58:42<2:24:48,  1.55it/s]
                                                         
{'loss': 1.5985, 'learning_rate': 2.8508020080813025e-07, 'epoch': 3216.05}
+
 79%|███████▉  | 52100/65536 [8:58:42<2:24:48,  1.55it/s]
 79%|███████▉  | 52101/65536 [8:58:43<2:21:07,  1.59it/s]
 80%|███████▉  | 52102/65536 [8:58:43<2:19:08,  1.61it/s]
 80%|███████▉  | 52103/65536 [8:58:44<2:16:16,  1.64it/s]
 80%|███████▉  | 52104/65536 [8:58:45<2:15:55,  1.65it/s]
 80%|███████▉  | 52105/65536 [8:58:45<2:21:42,  1.58it/s]
 80%|███████▉  | 52106/65536 [8:58:46<2:23:31,  1.56it/s]
 80%|███████▉  | 52107/65536 [8:58:47<2:22:18,  1.57it/s]
 80%|███████▉  | 52108/65536 [8:58:47<2:20:09,  1.60it/s]
 80%|███████▉  | 52109/65536 [8:58:48<2:18:04,  1.62it/s]
 80%|███████▉  | 52110/65536 [8:58:48<2:18:55,  1.61it/s]
 80%|███████▉  | 52111/65536 [8:58:49<2:17:53,  1.62it/s]
 80%|███████▉  | 52112/65536 [8:58:50<2:19:03,  1.61it/s]
 80%|███████▉  | 52113/65536 [8:58:50<2:21:40,  1.58it/s]
 80%|███████▉  | 52114/65536 [8:58:51<2:20:56,  1.59it/s]
 80%|███████▉  | 52115/65536 [8:58:52<2:20:53,  1.59it/s]
 80%|███████▉  | 52116/65536 [8:58:52<2:24:50,  1.54it/s]
 80%|███████▉  | 52117/65536 [8:58:53<2:22:22,  1.57it/s]
 80%|███████▉  | 52118/65536 [8:58:54<2:25:48,  1.53it/s]
 80%|███████▉  | 52119/65536 [8:58:54<2:23:05,  1.56it/s]
 80%|███████▉  | 52120/65536 [8:58:55<2:19:58,  1.60it/s]
                                                         
{'loss': 1.5181, 'learning_rate': 2.8480470184890416e-07, 'epoch': 3217.28}
+
 80%|███████▉  | 52120/65536 [8:58:55<2:19:58,  1.60it/s]
 80%|███████▉  | 52121/65536 [8:58:55<2:22:56,  1.56it/s]
 80%|███████▉  | 52122/65536 [8:58:56<2:22:43,  1.57it/s]
 80%|███████▉  | 52123/65536 [8:58:57<2:21:22,  1.58it/s]
 80%|███████▉  | 52124/65536 [8:58:57<2:23:05,  1.56it/s]
 80%|███████▉  | 52125/65536 [8:58:58<2:21:48,  1.58it/s]
 80%|███████▉  | 52126/65536 [8:58:59<2:19:15,  1.60it/s]
 80%|███████▉  | 52127/65536 [8:58:59<2:17:59,  1.62it/s]
 80%|███████▉  | 52128/65536 [8:59:00<2:16:42,  1.63it/s]
 80%|███████▉  | 52129/65536 [8:59:00<2:15:06,  1.65it/s]
 80%|███████▉  | 52130/65536 [8:59:01<2:14:55,  1.66it/s]
 80%|███████▉  | 52131/65536 [8:59:02<2:15:48,  1.65it/s]
 80%|███████▉  | 52132/65536 [8:59:02<2:22:41,  1.57it/s]
 80%|███████▉  | 52133/65536 [8:59:03<2:23:43,  1.55it/s]
 80%|███████▉  | 52134/65536 [8:59:04<2:23:00,  1.56it/s]
 80%|███████▉  | 52135/65536 [8:59:04<2:21:11,  1.58it/s]
 80%|███████▉  | 52136/65536 [8:59:05<2:19:31,  1.60it/s]
 80%|███████▉  | 52137/65536 [8:59:05<2:17:59,  1.62it/s]
 80%|███████▉  | 52138/65536 [8:59:06<2:17:37,  1.62it/s]
 80%|███████▉  | 52139/65536 [8:59:07<2:19:12,  1.60it/s]
 80%|███████▉  | 52140/65536 [8:59:07<2:21:10,  1.58it/s]
                                                         
{'loss': 1.5434, 'learning_rate': 2.84529202889678e-07, 'epoch': 3218.52}
+
 80%|███████▉  | 52140/65536 [8:59:07<2:21:10,  1.58it/s]
 80%|███████▉  | 52141/65536 [8:59:08<2:21:07,  1.58it/s]
 80%|███████▉  | 52142/65536 [8:59:08<2:19:30,  1.60it/s]
 80%|███████▉  | 52143/65536 [8:59:09<2:16:21,  1.64it/s]
 80%|███████▉  | 52144/65536 [8:59:10<2:13:16,  1.67it/s]
 80%|███████▉  | 52145/65536 [8:59:10<2:12:16,  1.69it/s]
 80%|███████▉  | 52146/65536 [8:59:11<2:18:50,  1.61it/s]
 80%|███████▉  | 52147/65536 [8:59:11<2:15:50,  1.64it/s]
 80%|███████▉  | 52148/65536 [8:59:12<2:20:09,  1.59it/s]
 80%|███████▉  | 52149/65536 [8:59:13<2:17:57,  1.62it/s]
 80%|███████▉  | 52150/65536 [8:59:13<2:17:28,  1.62it/s]
 80%|███████▉  | 52151/65536 [8:59:14<2:15:17,  1.65it/s]
 80%|███████▉  | 52152/65536 [8:59:15<2:14:33,  1.66it/s]
 80%|███████▉  | 52153/65536 [8:59:15<2:18:30,  1.61it/s]
 80%|███████▉  | 52154/65536 [8:59:16<2:19:29,  1.60it/s]
 80%|███████▉  | 52155/65536 [8:59:16<2:17:54,  1.62it/s]
 80%|███████▉  | 52156/65536 [8:59:17<2:15:43,  1.64it/s]
 80%|███████▉  | 52157/65536 [8:59:18<2:20:22,  1.59it/s]
 80%|███████▉  | 52158/65536 [8:59:18<2:17:52,  1.62it/s]
 80%|███████▉  | 52159/65536 [8:59:19<2:16:09,  1.64it/s]
 80%|███████▉  | 52160/65536 [8:59:19<2:15:09,  1.65it/s]
                                                         
{'loss': 1.5892, 'learning_rate': 2.842537039304518e-07, 'epoch': 3219.75}
+
 80%|███████▉  | 52160/65536 [8:59:19<2:15:09,  1.65it/s]
 80%|███████▉  | 52161/65536 [8:59:20<2:19:55,  1.59it/s]
 80%|███████▉  | 52162/65536 [8:59:21<2:20:13,  1.59it/s]
 80%|███████▉  | 52163/65536 [8:59:21<2:18:21,  1.61it/s]
 80%|███████▉  | 52164/65536 [8:59:22<2:17:06,  1.63it/s]
 80%|███████▉  | 52165/65536 [8:59:23<2:22:52,  1.56it/s]
 80%|███████▉  | 52166/65536 [8:59:23<2:25:13,  1.53it/s]
 80%|███████▉  | 52167/65536 [8:59:24<2:22:13,  1.57it/s]
 80%|███████▉  | 52168/65536 [8:59:25<2:23:27,  1.55it/s]
 80%|███████▉  | 52169/65536 [8:59:25<2:20:35,  1.58it/s]
 80%|███████▉  | 52170/65536 [8:59:26<2:19:34,  1.60it/s]
 80%|███████▉  | 52171/65536 [8:59:26<2:18:58,  1.60it/s]
 80%|███████▉  | 52172/65536 [8:59:27<2:18:28,  1.61it/s]
 80%|███████▉  | 52173/65536 [8:59:28<2:18:12,  1.61it/s]
 80%|███████▉  | 52174/65536 [8:59:28<2:17:26,  1.62it/s]
 80%|███████▉  | 52175/65536 [8:59:29<2:17:03,  1.62it/s]
 80%|███████▉  | 52176/65536 [8:59:30<2:15:55,  1.64it/s]
 80%|███████▉  | 52177/65536 [8:59:30<2:18:09,  1.61it/s]
 80%|███████▉  | 52178/65536 [8:59:31<2:18:01,  1.61it/s]
 80%|███████▉  | 52179/65536 [8:59:31<2:16:53,  1.63it/s]
 80%|███████▉  | 52180/65536 [8:59:32<2:16:00,  1.64it/s]
                                                         
{'loss': 1.5584, 'learning_rate': 2.8397820497122564e-07, 'epoch': 3220.99}
+
 80%|███████▉  | 52180/65536 [8:59:32<2:16:00,  1.64it/s]
 80%|███████▉  | 52181/65536 [8:59:33<2:20:08,  1.59it/s]
 80%|███████▉  | 52182/65536 [8:59:33<2:19:19,  1.60it/s]
 80%|███████▉  | 52183/65536 [8:59:34<2:18:16,  1.61it/s]
 80%|███████▉  | 52184/65536 [8:59:35<2:17:51,  1.61it/s]
 80%|███████▉  | 52185/65536 [8:59:35<2:20:19,  1.59it/s]
 80%|███████▉  | 52186/65536 [8:59:36<2:21:33,  1.57it/s]
 80%|███████▉  | 52187/65536 [8:59:36<2:18:17,  1.61it/s]
 80%|███████▉  | 52188/65536 [8:59:37<2:17:11,  1.62it/s]
 80%|███████▉  | 52189/65536 [8:59:38<2:16:40,  1.63it/s]
 80%|███████▉  | 52190/65536 [8:59:38<2:21:09,  1.58it/s]
 80%|███████▉  | 52191/65536 [8:59:39<2:21:40,  1.57it/s]
 80%|███████▉  | 52192/65536 [8:59:40<2:23:33,  1.55it/s]
 80%|███████▉  | 52193/65536 [8:59:40<2:20:01,  1.59it/s]
 80%|███████▉  | 52194/65536 [8:59:41<2:20:11,  1.59it/s]
 80%|███████▉  | 52195/65536 [8:59:41<2:20:42,  1.58it/s]
 80%|███████▉  | 52196/65536 [8:59:42<2:15:54,  1.64it/s]
 80%|███████▉  | 52197/65536 [8:59:43<2:19:16,  1.60it/s]
 80%|███████▉  | 52198/65536 [8:59:43<2:19:10,  1.60it/s]
 80%|███████▉  | 52199/65536 [8:59:44<2:18:18,  1.61it/s]
 80%|███████▉  | 52200/65536 [8:59:45<2:19:18,  1.60it/s]
                                                         
{'loss': 1.5746, 'learning_rate': 2.837027060119995e-07, 'epoch': 3222.22}
+
 80%|███████▉  | 52200/65536 [8:59:45<2:19:18,  1.60it/s]
 80%|███████▉  | 52201/65536 [8:59:45<2:19:37,  1.59it/s]
 80%|███████▉  | 52202/65536 [8:59:46<2:18:03,  1.61it/s]
 80%|███████▉  | 52203/65536 [8:59:46<2:21:02,  1.58it/s]
 80%|███████▉  | 52204/65536 [8:59:47<2:18:58,  1.60it/s]
 80%|███████▉  | 52205/65536 [8:59:48<2:17:40,  1.61it/s]
 80%|███████▉  | 52206/65536 [8:59:48<2:14:40,  1.65it/s]
 80%|███████▉  | 52207/65536 [8:59:49<2:17:45,  1.61it/s]
 80%|███████▉  | 52208/65536 [8:59:50<2:20:48,  1.58it/s]
 80%|███████▉  | 52209/65536 [8:59:50<2:15:47,  1.64it/s]
 80%|███████▉  | 52210/65536 [8:59:51<2:15:36,  1.64it/s]
 80%|███████▉  | 52211/65536 [8:59:51<2:13:30,  1.66it/s]
 80%|███████▉  | 52212/65536 [8:59:52<2:14:02,  1.66it/s]
 80%|███████▉  | 52213/65536 [8:59:53<2:18:59,  1.60it/s]
 80%|███████▉  | 52214/65536 [8:59:53<2:23:20,  1.55it/s]
 80%|███████▉  | 52215/65536 [8:59:54<2:20:44,  1.58it/s]
 80%|███████▉  | 52216/65536 [8:59:55<2:18:12,  1.61it/s]
 80%|███████▉  | 52217/65536 [8:59:55<2:19:27,  1.59it/s]
 80%|███████▉  | 52218/65536 [8:59:56<2:16:36,  1.62it/s]
 80%|███████▉  | 52219/65536 [8:59:56<2:18:00,  1.61it/s]
 80%|███████▉  | 52220/65536 [8:59:57<2:14:24,  1.65it/s]
                                                         
{'loss': 1.5991, 'learning_rate': 2.8342720705277335e-07, 'epoch': 3223.46}
+
 80%|███████▉  | 52220/65536 [8:59:57<2:14:24,  1.65it/s]
 80%|███████▉  | 52221/65536 [8:59:58<2:16:52,  1.62it/s]
 80%|███████▉  | 52222/65536 [8:59:58<2:18:59,  1.60it/s]
 80%|███████▉  | 52223/65536 [8:59:59<2:16:26,  1.63it/s]
 80%|███████▉  | 52224/65536 [8:59:59<2:17:47,  1.61it/s]
 80%|███████▉  | 52225/65536 [9:00:00<2:15:15,  1.64it/s]
 80%|███████▉  | 52226/65536 [9:00:01<2:15:02,  1.64it/s]
 80%|███████▉  | 52227/65536 [9:00:01<2:12:24,  1.68it/s]
 80%|███████▉  | 52228/65536 [9:00:02<2:12:42,  1.67it/s]
 80%|███████▉  | 52229/65536 [9:00:03<2:19:22,  1.59it/s]
 80%|███████▉  | 52230/65536 [9:00:03<2:18:57,  1.60it/s]
 80%|███████▉  | 52231/65536 [9:00:04<2:18:25,  1.60it/s]
 80%|███████▉  | 52232/65536 [9:00:04<2:18:06,  1.61it/s]
 80%|███████▉  | 52233/65536 [9:00:05<2:15:43,  1.63it/s]
 80%|███████▉  | 52234/65536 [9:00:06<2:16:06,  1.63it/s]
 80%|███████▉  | 52235/65536 [9:00:06<2:16:21,  1.63it/s]
 80%|███████▉  | 52236/65536 [9:00:07<2:21:35,  1.57it/s]
 80%|███████▉  | 52237/65536 [9:00:08<2:23:19,  1.55it/s]
 80%|███████▉  | 52238/65536 [9:00:08<2:19:29,  1.59it/s]
 80%|███████▉  | 52239/65536 [9:00:09<2:20:33,  1.58it/s]
 80%|███████▉  | 52240/65536 [9:00:09<2:19:13,  1.59it/s]
                                                         
{'loss': 1.5501, 'learning_rate': 2.831517080935472e-07, 'epoch': 3224.69}
+
 80%|███████▉  | 52240/65536 [9:00:09<2:19:13,  1.59it/s]
 80%|███████▉  | 52241/65536 [9:00:10<2:19:25,  1.59it/s]
 80%|███████▉  | 52242/65536 [9:00:11<2:19:39,  1.59it/s]
 80%|███████▉  | 52243/65536 [9:00:11<2:16:26,  1.62it/s]
 80%|███████▉  | 52244/65536 [9:00:12<2:16:04,  1.63it/s]
 80%|███████▉  | 52245/65536 [9:00:12<2:13:31,  1.66it/s]
 80%|███████▉  | 52246/65536 [9:00:13<2:22:06,  1.56it/s]
 80%|███████▉  | 52247/65536 [9:00:14<2:24:00,  1.54it/s]
 80%|███████▉  | 52248/65536 [9:00:14<2:20:28,  1.58it/s]
 80%|███████▉  | 52249/65536 [9:00:15<2:16:35,  1.62it/s]
 80%|███████▉  | 52250/65536 [9:00:16<2:15:53,  1.63it/s]
 80%|███████▉  | 52251/65536 [9:00:16<2:16:10,  1.63it/s]
 80%|███████▉  | 52252/65536 [9:00:17<2:15:25,  1.63it/s]
 80%|███████▉  | 52253/65536 [9:00:18<2:20:21,  1.58it/s]
 80%|███████▉  | 52254/65536 [9:00:18<2:18:06,  1.60it/s]
 80%|███████▉  | 52255/65536 [9:00:19<2:14:30,  1.65it/s]
 80%|███████▉  | 52256/65536 [9:00:19<2:13:20,  1.66it/s]
 80%|███████▉  | 52257/65536 [9:00:20<2:13:15,  1.66it/s]
 80%|███████▉  | 52258/65536 [9:00:21<2:13:47,  1.65it/s]
 80%|███████▉  | 52259/65536 [9:00:21<2:13:30,  1.66it/s]
 80%|███████▉  | 52260/65536 [9:00:22<2:18:23,  1.60it/s]
                                                         
{'loss': 1.5744, 'learning_rate': 2.8287620913432107e-07, 'epoch': 3225.93}
+
 80%|███████▉  | 52260/65536 [9:00:22<2:18:23,  1.60it/s]
 80%|███████▉  | 52261/65536 [9:00:22<2:17:22,  1.61it/s]
 80%|███████▉  | 52262/65536 [9:00:23<2:21:25,  1.56it/s]
 80%|███████▉  | 52263/65536 [9:00:24<2:19:26,  1.59it/s]
 80%|███████▉  | 52264/65536 [9:00:24<2:17:29,  1.61it/s]
 80%|███████▉  | 52265/65536 [9:00:25<2:14:01,  1.65it/s]
 80%|███████▉  | 52266/65536 [9:00:26<2:19:27,  1.59it/s]
 80%|███████▉  | 52267/65536 [9:00:26<2:20:55,  1.57it/s]
 80%|███████▉  | 52268/65536 [9:00:27<2:18:47,  1.59it/s]
 80%|███████▉  | 52269/65536 [9:00:27<2:14:22,  1.65it/s]
 80%|███████▉  | 52270/65536 [9:00:28<2:15:15,  1.63it/s]
 80%|███████▉  | 52271/65536 [9:00:29<2:15:16,  1.63it/s]
 80%|███████▉  | 52272/65536 [9:00:29<2:15:00,  1.64it/s]
 80%|███████▉  | 52273/65536 [9:00:30<2:16:10,  1.62it/s]
 80%|███████▉  | 52274/65536 [9:00:30<2:16:38,  1.62it/s]
 80%|███████▉  | 52275/65536 [9:00:31<2:19:44,  1.58it/s]
 80%|███████▉  | 52276/65536 [9:00:32<2:17:07,  1.61it/s]
 80%|███████▉  | 52277/65536 [9:00:32<2:16:27,  1.62it/s]
 80%|███████▉  | 52278/65536 [9:00:33<2:20:19,  1.57it/s]
 80%|███████▉  | 52279/65536 [9:00:34<2:20:33,  1.57it/s]
 80%|███████▉  | 52280/65536 [9:00:34<2:21:58,  1.56it/s]
                                                         
{'loss': 1.5674, 'learning_rate': 2.8260071017509493e-07, 'epoch': 3227.16}
+
 80%|███████▉  | 52280/65536 [9:00:34<2:21:58,  1.56it/s]
 80%|███████▉  | 52281/65536 [9:00:35<2:20:01,  1.58it/s]
 80%|███████▉  | 52282/65536 [9:00:36<2:19:07,  1.59it/s]
 80%|███████▉  | 52283/65536 [9:00:36<2:17:28,  1.61it/s]
 80%|███████▉  | 52284/65536 [9:00:37<2:16:37,  1.62it/s]
 80%|███████▉  | 52285/65536 [9:00:37<2:18:54,  1.59it/s]
 80%|███████▉  | 52286/65536 [9:00:38<2:16:58,  1.61it/s]
 80%|███████▉  | 52287/65536 [9:00:39<2:18:29,  1.59it/s]
 80%|███████▉  | 52288/65536 [9:00:39<2:14:36,  1.64it/s]
 80%|███████▉  | 52289/65536 [9:00:40<2:15:08,  1.63it/s]
 80%|███████▉  | 52290/65536 [9:00:40<2:13:07,  1.66it/s]
 80%|███████▉  | 52291/65536 [9:00:41<2:15:07,  1.63it/s]
 80%|███████▉  | 52292/65536 [9:00:42<2:13:25,  1.65it/s]
 80%|███████▉  | 52293/65536 [9:00:42<2:12:00,  1.67it/s]
 80%|███████▉  | 52294/65536 [9:00:43<2:15:03,  1.63it/s]
 80%|███████▉  | 52295/65536 [9:00:43<2:14:34,  1.64it/s]
 80%|███████▉  | 52296/65536 [9:00:44<2:13:37,  1.65it/s]
 80%|███████▉  | 52297/65536 [9:00:45<2:15:56,  1.62it/s]
 80%|███████▉  | 52298/65536 [9:00:45<2:13:59,  1.65it/s]
 80%|███████▉  | 52299/65536 [9:00:46<2:16:53,  1.61it/s]
 80%|███████▉  | 52300/65536 [9:00:47<2:17:31,  1.60it/s]
                                                         
{'loss': 1.5572, 'learning_rate': 2.823252112158687e-07, 'epoch': 3228.4}
+
 80%|███████▉  | 52300/65536 [9:00:47<2:17:31,  1.60it/s]
 80%|███████▉  | 52301/65536 [9:00:47<2:17:11,  1.61it/s]
 80%|███████▉  | 52302/65536 [9:00:48<2:19:24,  1.58it/s]
 80%|███████▉  | 52303/65536 [9:00:48<2:19:52,  1.58it/s]
 80%|███████▉  | 52304/65536 [9:00:49<2:16:59,  1.61it/s]
 80%|███████▉  | 52305/65536 [9:00:50<2:15:30,  1.63it/s]
 80%|███████▉  | 52306/65536 [9:00:50<2:16:00,  1.62it/s]
 80%|███████▉  | 52307/65536 [9:00:51<2:15:01,  1.63it/s]
 80%|███████▉  | 52308/65536 [9:00:52<2:16:09,  1.62it/s]
 80%|███████▉  | 52309/65536 [9:00:52<2:18:12,  1.60it/s]
 80%|███████▉  | 52310/65536 [9:00:53<2:26:15,  1.51it/s]
 80%|██���████▉  | 52311/65536 [9:00:54<2:21:13,  1.56it/s]
 80%|███████▉  | 52312/65536 [9:00:54<2:20:39,  1.57it/s]
 80%|███████▉  | 52313/65536 [9:00:55<2:19:49,  1.58it/s]
 80%|███████▉  | 52314/65536 [9:00:55<2:22:19,  1.55it/s]
 80%|███████▉  | 52315/65536 [9:00:56<2:18:09,  1.59it/s]
 80%|███████▉  | 52316/65536 [9:00:57<2:17:47,  1.60it/s]
 80%|███████▉  | 52317/65536 [9:00:57<2:15:58,  1.62it/s]
 80%|███████▉  | 52318/65536 [9:00:58<2:12:58,  1.66it/s]
 80%|███████▉  | 52319/65536 [9:00:58<2:11:22,  1.68it/s]
 80%|███████▉  | 52320/65536 [9:00:59<2:13:13,  1.65it/s]
                                                         
{'loss': 1.5803, 'learning_rate': 2.8204971225664254e-07, 'epoch': 3229.63}
+
 80%|███████▉  | 52320/65536 [9:00:59<2:13:13,  1.65it/s]
 80%|███████▉  | 52321/65536 [9:01:00<2:14:49,  1.63it/s]
 80%|███████▉  | 52322/65536 [9:01:00<2:12:18,  1.66it/s]
 80%|███████▉  | 52323/65536 [9:01:01<2:10:13,  1.69it/s]
 80%|███████▉  | 52324/65536 [9:01:01<2:13:58,  1.64it/s]
 80%|███████▉  | 52325/65536 [9:01:02<2:17:30,  1.60it/s]
 80%|███████▉  | 52326/65536 [9:01:03<2:15:40,  1.62it/s]
 80%|███████▉  | 52327/65536 [9:01:03<2:19:59,  1.57it/s]
 80%|███████▉  | 52328/65536 [9:01:04<2:22:49,  1.54it/s]
 80%|███████▉  | 52329/65536 [9:01:05<2:20:45,  1.56it/s]
 80%|███████▉  | 52330/65536 [9:01:05<2:17:50,  1.60it/s]
 80%|███████▉  | 52331/65536 [9:01:06<2:18:20,  1.59it/s]
 80%|███████▉  | 52332/65536 [9:01:07<2:17:07,  1.60it/s]
 80%|███████▉  | 52333/65536 [9:01:07<2:14:57,  1.63it/s]
 80%|███████▉  | 52334/65536 [9:01:08<2:11:53,  1.67it/s]
 80%|███████▉  | 52335/65536 [9:01:08<2:12:55,  1.66it/s]
 80%|███████▉  | 52336/65536 [9:01:09<2:13:24,  1.65it/s]
 80%|███████▉  | 52337/65536 [9:01:10<2:14:10,  1.64it/s]
 80%|███████▉  | 52338/65536 [9:01:10<2:15:03,  1.63it/s]
 80%|███████▉  | 52339/65536 [9:01:11<2:19:36,  1.58it/s]
 80%|███████▉  | 52340/65536 [9:01:11<2:15:54,  1.62it/s]
                                                         
{'loss': 1.5868, 'learning_rate': 2.817742132974164e-07, 'epoch': 3230.86}
+
 80%|███████▉  | 52340/65536 [9:01:11<2:15:54,  1.62it/s]
 80%|███████▉  | 52341/65536 [9:01:12<2:16:21,  1.61it/s]
 80%|███████▉  | 52342/65536 [9:01:13<2:14:34,  1.63it/s]
 80%|███████▉  | 52343/65536 [9:01:13<2:18:01,  1.59it/s]
 80%|███████▉  | 52344/65536 [9:01:14<2:15:05,  1.63it/s]
 80%|███████▉  | 52345/65536 [9:01:15<2:18:12,  1.59it/s]
 80%|███████▉  | 52346/65536 [9:01:15<2:17:22,  1.60it/s]
 80%|███████▉  | 52347/65536 [9:01:16<2:18:24,  1.59it/s]
 80%|███████▉  | 52348/65536 [9:01:16<2:14:10,  1.64it/s]
 80%|███████▉  | 52349/65536 [9:01:17<2:19:33,  1.57it/s]
 80%|███████▉  | 52350/65536 [9:01:18<2:17:49,  1.59it/s]
 80%|███████▉  | 52351/65536 [9:01:18<2:17:24,  1.60it/s]
 80%|███████▉  | 52352/65536 [9:01:19<2:17:56,  1.59it/s]
 80%|███████▉  | 52353/65536 [9:01:20<2:16:57,  1.60it/s]
 80%|███████▉  | 52354/65536 [9:01:20<2:15:33,  1.62it/s]
 80%|███████▉  | 52355/65536 [9:01:21<2:17:24,  1.60it/s]
 80%|███████▉  | 52356/65536 [9:01:21<2:17:34,  1.60it/s]
 80%|███████▉  | 52357/65536 [9:01:22<2:13:17,  1.65it/s]
 80%|███████▉  | 52358/65536 [9:01:23<2:12:36,  1.66it/s]
 80%|███████▉  | 52359/65536 [9:01:23<2:16:08,  1.61it/s]
 80%|███████▉  | 52360/65536 [9:01:24<2:14:42,  1.63it/s]
                                                         
{'loss': 1.5713, 'learning_rate': 2.8149871433819026e-07, 'epoch': 3232.1}
+
 80%|███████▉  | 52360/65536 [9:01:24<2:14:42,  1.63it/s]
 80%|███████▉  | 52361/65536 [9:01:24<2:15:12,  1.62it/s]
 80%|███████▉  | 52362/65536 [9:01:25<2:14:54,  1.63it/s]
 80%|███████▉  | 52363/65536 [9:01:26<2:16:47,  1.61it/s]
 80%|███████▉  | 52364/65536 [9:01:26<2:17:10,  1.60it/s]
 80%|███████▉  | 52365/65536 [9:01:27<2:19:54,  1.57it/s]
 80%|███████▉  | 52366/65536 [9:01:28<2:17:36,  1.60it/s]
 80%|███████▉  | 52367/65536 [9:01:28<2:18:19,  1.59it/s]
 80%|███████▉  | 52368/65536 [9:01:29<2:16:43,  1.61it/s]
 80%|███████▉  | 52369/65536 [9:01:29<2:13:27,  1.64it/s]
 80%|███████▉  | 52370/65536 [9:01:30<2:14:57,  1.63it/s]
 80%|███████▉  | 52371/65536 [9:01:31<2:14:19,  1.63it/s]
 80%|███████▉  | 52372/65536 [9:01:31<2:14:49,  1.63it/s]
 80%|███████▉  | 52373/65536 [9:01:32<2:14:14,  1.63it/s]
 80%|███████▉  | 52374/65536 [9:01:32<2:14:23,  1.63it/s]
 80%|███████▉  | 52375/65536 [9:01:33<2:20:55,  1.56it/s]
 80%|███████▉  | 52376/65536 [9:01:34<2:16:54,  1.60it/s]
 80%|███████▉  | 52377/65536 [9:01:34<2:14:39,  1.63it/s]
 80%|███████▉  | 52378/65536 [9:01:35<2:15:03,  1.62it/s]
 80%|███████▉  | 52379/65536 [9:01:36<2:19:00,  1.58it/s]
 80%|███████▉  | 52380/65536 [9:01:36<2:19:14,  1.57it/s]
                                                         
{'loss': 1.5832, 'learning_rate': 2.812232153789641e-07, 'epoch': 3233.33}
+
 80%|███████▉  | 52380/65536 [9:01:36<2:19:14,  1.57it/s]
 80%|███████▉  | 52381/65536 [9:01:37<2:20:19,  1.56it/s]
 80%|███████▉  | 52382/65536 [9:01:38<2:15:00,  1.62it/s]
 80%|███████▉  | 52383/65536 [9:01:38<2:16:18,  1.61it/s]
 80%|███████▉  | 52384/65536 [9:01:39<2:15:07,  1.62it/s]
 80%|███████▉  | 52385/65536 [9:01:39<2:14:07,  1.63it/s]
 80%|███████▉  | 52386/65536 [9:01:40<2:13:22,  1.64it/s]
 80%|███████▉  | 52387/65536 [9:01:41<2:14:26,  1.63it/s]
 80%|███████▉  | 52388/65536 [9:01:41<2:14:14,  1.63it/s]
 80%|███████▉  | 52389/65536 [9:01:42<2:14:55,  1.62it/s]
 80%|███████▉  | 52390/65536 [9:01:42<2:17:43,  1.59it/s]
 80%|███████▉  | 52391/65536 [9:01:43<2:22:00,  1.54it/s]
 80%|███████▉  | 52392/65536 [9:01:44<2:18:08,  1.59it/s]
 80%|███████▉  | 52393/65536 [9:01:44<2:17:03,  1.60it/s]
 80%|███████▉  | 52394/65536 [9:01:45<2:14:30,  1.63it/s]
 80%|███████▉  | 52395/65536 [9:01:46<2:14:57,  1.62it/s]
 80%|███████▉  | 52396/65536 [9:01:46<2:13:05,  1.65it/s]
 80%|███████▉  | 52397/65536 [9:01:47<2:12:01,  1.66it/s]
 80%|███████▉  | 52398/65536 [9:01:47<2:14:19,  1.63it/s]
 80%|███████▉  | 52399/65536 [9:01:48<2:15:54,  1.61it/s]
 80%|███████▉  | 52400/65536 [9:01:49<2:12:55,  1.65it/s]
                                                         
{'loss': 1.574, 'learning_rate': 2.80947716419738e-07, 'epoch': 3234.57}
+
 80%|███████▉  | 52400/65536 [9:01:49<2:12:55,  1.65it/s]
 80%|███████▉  | 52401/65536 [9:01:49<2:12:15,  1.66it/s]
 80%|███████▉  | 52402/65536 [9:01:50<2:13:25,  1.64it/s]
 80%|███████▉  | 52403/65536 [9:01:50<2:15:27,  1.62it/s]
 80%|███████▉  | 52404/65536 [9:01:51<2:15:23,  1.62it/s]
 80%|███████▉  | 52405/65536 [9:01:52<2:20:58,  1.55it/s]
 80%|███████▉  | 52406/65536 [9:01:52<2:17:09,  1.60it/s]
 80%|███████▉  | 52407/65536 [9:01:53<2:18:49,  1.58it/s]
 80%|███████▉  | 52408/65536 [9:01:54<2:24:02,  1.52it/s]
 80%|███████▉  | 52409/65536 [9:01:54<2:19:42,  1.57it/s]
 80%|███████▉  | 52410/65536 [9:01:55<2:18:44,  1.58it/s]
 80%|███████▉  | 52411/65536 [9:01:56<2:23:14,  1.53it/s]
 80%|███████▉  | 52412/65536 [9:01:56<2:24:39,  1.51it/s]
 80%|███████▉  | 52413/65536 [9:01:57<2:30:36,  1.45it/s]
 80%|███████▉  | 52414/65536 [9:01:58<2:26:02,  1.50it/s]
 80%|███████▉  | 52415/65536 [9:01:58<2:22:46,  1.53it/s]
 80%|███████▉  | 52416/65536 [9:01:59<2:18:42,  1.58it/s]
 80%|███████▉  | 52417/65536 [9:02:00<2:16:08,  1.61it/s]
 80%|███████▉  | 52418/65536 [9:02:00<2:13:33,  1.64it/s]
 80%|███████▉  | 52419/65536 [9:02:01<2:15:12,  1.62it/s]
 80%|███████▉  | 52420/65536 [9:02:01<2:14:15,  1.63it/s]
                                                         
{'loss': 1.5573, 'learning_rate': 2.8067221746051184e-07, 'epoch': 3235.8}
+
 80%|███████▉  | 52420/65536 [9:02:01<2:14:15,  1.63it/s]
 80%|███████▉  | 52421/65536 [9:02:02<2:20:18,  1.56it/s]
 80%|███████▉  | 52422/65536 [9:02:03<2:20:18,  1.56it/s]
 80%|███████▉  | 52423/65536 [9:02:03<2:18:02,  1.58it/s]
 80%|███████▉  | 52424/65536 [9:02:04<2:20:51,  1.55it/s]
 80%|███████▉  | 52425/65536 [9:02:05<2:19:33,  1.57it/s]
 80%|███████▉  | 52426/65536 [9:02:05<2:18:12,  1.58it/s]
 80%|███████▉  | 52427/65536 [9:02:06<2:18:45,  1.57it/s]
 80%|███████▉  | 52428/65536 [9:02:06<2:16:30,  1.60it/s]
 80%|████████  | 52429/65536 [9:02:07<2:16:31,  1.60it/s]
 80%|████████  | 52430/65536 [9:02:08<2:14:59,  1.62it/s]
 80%|████████  | 52431/65536 [9:02:08<2:13:13,  1.64it/s]
 80%|████████  | 52432/65536 [9:02:09<2:15:38,  1.61it/s]
 80%|████████  | 52433/65536 [9:02:10<2:18:58,  1.57it/s]
 80%|████████  | 52434/65536 [9:02:10<2:20:39,  1.55it/s]
 80%|████████  | 52435/65536 [9:02:11<2:19:41,  1.56it/s]
 80%|████████  | 52436/65536 [9:02:11<2:16:16,  1.60it/s]
 80%|████████  | 52437/65536 [9:02:12<2:10:51,  1.67it/s]
 80%|████████  | 52438/65536 [9:02:13<2:14:07,  1.63it/s]
 80%|████████  | 52439/65536 [9:02:13<2:13:38,  1.63it/s]
 80%|████████  | 52440/65536 [9:02:14<2:18:23,  1.58it/s]
                                                         
{'loss': 1.522, 'learning_rate': 2.803967185012856e-07, 'epoch': 3237.04}
+
 80%|████████  | 52440/65536 [9:02:14<2:18:23,  1.58it/s]
 80%|████████  | 52441/65536 [9:02:15<2:15:02,  1.62it/s]
 80%|████████  | 52442/65536 [9:02:15<2:13:13,  1.64it/s]
 80%|████████  | 52443/65536 [9:02:16<2:11:39,  1.66it/s]
 80%|████████  | 52444/65536 [9:02:16<2:14:57,  1.62it/s]
 80%|████████  | 52445/65536 [9:02:17<2:12:38,  1.64it/s]
 80%|████████  | 52446/65536 [9:02:18<2:09:41,  1.68it/s]
 80%|████████  | 52447/65536 [9:02:18<2:15:50,  1.61it/s]
 80%|████████  | 52448/65536 [9:02:19<2:16:08,  1.60it/s]
 80%|████████  | 52449/65536 [9:02:19<2:17:03,  1.59it/s]
 80%|████████  | 52450/65536 [9:02:20<2:15:55,  1.60it/s]
 80%|████████  | 52451/65536 [9:02:21<2:15:04,  1.61it/s]
 80%|████████  | 52452/65536 [9:02:21<2:17:09,  1.59it/s]
 80%|████████  | 52453/65536 [9:02:22<2:16:13,  1.60it/s]
 80%|████████  | 52454/65536 [9:02:23<2:17:37,  1.58it/s]
 80%|████████  | 52455/65536 [9:02:23<2:13:43,  1.63it/s]
 80%|████████  | 52456/65536 [9:02:24<2:17:38,  1.58it/s]
 80%|████████  | 52457/65536 [9:02:24<2:15:58,  1.60it/s]
 80%|████████  | 52458/65536 [9:02:25<2:14:03,  1.63it/s]
 80%|████████  | 52459/65536 [9:02:26<2:13:37,  1.63it/s]
 80%|████████  | 52460/65536 [9:02:26<2:13:32,  1.63it/s]
                                                         
{'loss': 1.5786, 'learning_rate': 2.8012121954205945e-07, 'epoch': 3238.27}
+
 80%|████████  | 52460/65536 [9:02:26<2:13:32,  1.63it/s]
 80%|████████  | 52461/65536 [9:02:27<2:17:02,  1.59it/s]
 80%|████████  | 52462/65536 [9:02:28<2:22:37,  1.53it/s]
 80%|████████  | 52463/65536 [9:02:28<2:18:43,  1.57it/s]
 80%|████████  | 52464/65536 [9:02:29<2:17:11,  1.59it/s]
 80%|████████  | 52465/65536 [9:02:29<2:15:34,  1.61it/s]
 80%|████████  | 52466/65536 [9:02:30<2:16:15,  1.60it/s]
 80%|████████  | 52467/65536 [9:02:31<2:12:59,  1.64it/s]
 80%|████████  | 52468/65536 [9:02:31<2:12:37,  1.64it/s]
 80%|████████  | 52469/65536 [9:02:32<2:15:22,  1.61it/s]
 80%|████████  | 52470/65536 [9:02:33<2:15:27,  1.61it/s]
 80%|████████  | 52471/65536 [9:02:33<2:11:57,  1.65it/s]
 80%|████████  | 52472/65536 [9:02:34<2:17:22,  1.59it/s]
 80%|████████  | 52473/65536 [9:02:34<2:14:48,  1.62it/s]
 80%|████████  | 52474/65536 [9:02:35<2:14:59,  1.61it/s]
 80%|████████  | 52475/65536 [9:02:36<2:16:01,  1.60it/s]
 80%|████████  | 52476/65536 [9:02:36<2:18:30,  1.57it/s]
 80%|████████  | 52477/65536 [9:02:37<2:18:55,  1.57it/s]
 80%|████████  | 52478/65536 [9:02:38<2:16:07,  1.60it/s]
 80%|████████  | 52479/65536 [9:02:38<2:12:11,  1.65it/s]
 80%|████████  | 52480/65536 [9:02:39<2:12:44,  1.64it/s]
                                                         
{'loss': 1.5853, 'learning_rate': 2.798457205828333e-07, 'epoch': 3239.51}
+
 80%|████████  | 52480/65536 [9:02:39<2:12:44,  1.64it/s]
 80%|████████  | 52481/65536 [9:02:39<2:12:36,  1.64it/s]
 80%|████████  | 52482/65536 [9:02:40<2:16:42,  1.59it/s]
 80%|████████  | 52483/65536 [9:02:41<2:13:11,  1.63it/s]
 80%|████████  | 52484/65536 [9:02:41<2:11:57,  1.65it/s]
 80%|████████  | 52485/65536 [9:02:42<2:11:01,  1.66it/s]
 80%|████████  | 52486/65536 [9:02:42<2:11:15,  1.66it/s]
 80%|████████  | 52487/65536 [9:02:43<2:13:22,  1.63it/s]
 80%|████████  | 52488/65536 [9:02:44<2:14:08,  1.62it/s]
 80%|████████  | 52489/65536 [9:02:44<2:19:56,  1.55it/s]
 80%|████████  | 52490/65536 [9:02:45<2:20:19,  1.55it/s]
 80%|████████  | 52491/65536 [9:02:46<2:22:15,  1.53it/s]
 80%|████████  | 52492/65536 [9:02:46<2:19:43,  1.56it/s]
 80%|██████���█  | 52493/65536 [9:02:47<2:16:19,  1.59it/s]
 80%|████████  | 52494/65536 [9:02:48<2:20:11,  1.55it/s]
 80%|████████  | 52495/65536 [9:02:48<2:20:47,  1.54it/s]
 80%|████████  | 52496/65536 [9:02:49<2:15:44,  1.60it/s]
 80%|████████  | 52497/65536 [9:02:49<2:16:05,  1.60it/s]
 80%|████████  | 52498/65536 [9:02:50<2:14:45,  1.61it/s]
 80%|████████  | 52499/65536 [9:02:51<2:13:18,  1.63it/s]
 80%|████████  | 52500/65536 [9:02:51<2:10:40,  1.66it/s]
                                                         
{'loss': 1.532, 'learning_rate': 2.7957022162360716e-07, 'epoch': 3240.74}
+
 80%|████████  | 52500/65536 [9:02:51<2:10:40,  1.66it/s]
 80%|████████  | 52501/65536 [9:02:52<2:11:37,  1.65it/s]
 80%|████████  | 52502/65536 [9:02:52<2:10:07,  1.67it/s]
 80%|████████  | 52503/65536 [9:02:53<2:10:20,  1.67it/s]
 80%|████████  | 52504/65536 [9:02:54<2:11:00,  1.66it/s]
 80%|████████  | 52505/65536 [9:02:54<2:14:21,  1.62it/s]
 80%|████████  | 52506/65536 [9:02:55<2:13:37,  1.63it/s]
 80%|████████  | 52507/65536 [9:02:55<2:12:13,  1.64it/s]
 80%|████████  | 52508/65536 [9:02:56<2:13:08,  1.63it/s]
 80%|████████  | 52509/65536 [9:02:57<2:16:08,  1.59it/s]
 80%|████████  | 52510/65536 [9:02:57<2:16:17,  1.59it/s]
 80%|████████  | 52511/65536 [9:02:58<2:18:44,  1.56it/s]
 80%|████████  | 52512/65536 [9:02:59<2:18:33,  1.57it/s]
 80%|████████  | 52513/65536 [9:02:59<2:14:57,  1.61it/s]
 80%|████████  | 52514/65536 [9:03:00<2:13:39,  1.62it/s]
 80%|████████  | 52515/65536 [9:03:00<2:13:55,  1.62it/s]
 80%|████████  | 52516/65536 [9:03:01<2:11:47,  1.65it/s]
 80%|████████  | 52517/65536 [9:03:02<2:11:47,  1.65it/s]
 80%|████████  | 52518/65536 [9:03:02<2:11:49,  1.65it/s]
 80%|████████  | 52519/65536 [9:03:03<2:14:32,  1.61it/s]
 80%|████████  | 52520/65536 [9:03:04<2:17:01,  1.58it/s]
                                                         
{'loss': 1.5801, 'learning_rate': 2.79294722664381e-07, 'epoch': 3241.98}
+
 80%|████████  | 52520/65536 [9:03:04<2:17:01,  1.58it/s]
 80%|████████  | 52521/65536 [9:03:04<2:26:17,  1.48it/s]
 80%|████████  | 52522/65536 [9:03:05<2:24:42,  1.50it/s]
 80%|████████  | 52523/65536 [9:03:06<2:24:53,  1.50it/s]
 80%|████████  | 52524/65536 [9:03:06<2:23:10,  1.51it/s]
 80%|████████  | 52525/65536 [9:03:07<2:17:12,  1.58it/s]
 80%|████████  | 52526/65536 [9:03:07<2:13:36,  1.62it/s]
 80%|████████  | 52527/65536 [9:03:08<2:16:34,  1.59it/s]
 80%|████████  | 52528/65536 [9:03:09<2:14:29,  1.61it/s]
 80%|████████  | 52529/65536 [9:03:09<2:15:12,  1.60it/s]
 80%|████████  | 52530/65536 [9:03:10<2:13:51,  1.62it/s]
 80%|████████  | 52531/65536 [9:03:11<2:13:41,  1.62it/s]
 80%|████████  | 52532/65536 [9:03:11<2:13:50,  1.62it/s]
 80%|████████  | 52533/65536 [9:03:12<2:14:11,  1.61it/s]
 80%|████████  | 52534/65536 [9:03:12<2:15:18,  1.60it/s]
 80%|████████  | 52535/65536 [9:03:13<2:13:30,  1.62it/s]
 80%|████████  | 52536/65536 [9:03:14<2:12:45,  1.63it/s]
 80%|████████  | 52537/65536 [9:03:14<2:17:32,  1.58it/s]
 80%|████████  | 52538/65536 [9:03:15<2:12:56,  1.63it/s]
 80%|████████  | 52539/65536 [9:03:16<2:11:00,  1.65it/s]
 80%|████████  | 52540/65536 [9:03:16<2:15:07,  1.60it/s]
                                                         
{'loss': 1.5543, 'learning_rate': 2.790192237051549e-07, 'epoch': 3243.21}
+
 80%|████████  | 52540/65536 [9:03:16<2:15:07,  1.60it/s]
 80%|████████  | 52541/65536 [9:03:17<2:13:22,  1.62it/s]
 80%|████████  | 52542/65536 [9:03:17<2:14:31,  1.61it/s]
 80%|████████  | 52543/65536 [9:03:18<2:15:31,  1.60it/s]
 80%|████████  | 52544/65536 [9:03:19<2:12:06,  1.64it/s]
 80%|████████  | 52545/65536 [9:03:19<2:15:10,  1.60it/s]
 80%|████████  | 52546/65536 [9:03:20<2:17:16,  1.58it/s]
 80%|████████  | 52547/65536 [9:03:21<2:14:22,  1.61it/s]
 80%|████████  | 52548/65536 [9:03:21<2:14:03,  1.61it/s]
 80%|████████  | 52549/65536 [9:03:22<2:14:05,  1.61it/s]
 80%|████████  | 52550/65536 [9:03:22<2:12:53,  1.63it/s]
 80%|████████  | 52551/65536 [9:03:23<2:10:39,  1.66it/s]
 80%|████████  | 52552/65536 [9:03:24<2:16:11,  1.59it/s]
 80%|████████  | 52553/65536 [9:03:24<2:19:24,  1.55it/s]
 80%|████████  | 52554/65536 [9:03:25<2:18:45,  1.56it/s]
 80%|████████  | 52555/65536 [9:03:26<2:15:05,  1.60it/s]
 80%|████████  | 52556/65536 [9:03:26<2:12:27,  1.63it/s]
 80%|████████  | 52557/65536 [9:03:27<2:10:36,  1.66it/s]
 80%|████████  | 52558/65536 [9:03:27<2:10:57,  1.65it/s]
 80%|████████  | 52559/65536 [9:03:28<2:13:51,  1.62it/s]
 80%|████████  | 52560/65536 [9:03:29<2:15:31,  1.60it/s]
                                                         
{'loss': 1.5939, 'learning_rate': 2.787437247459287e-07, 'epoch': 3244.44}
+
 80%|████████  | 52560/65536 [9:03:29<2:15:31,  1.60it/s]
 80%|████████  | 52561/65536 [9:03:29<2:14:13,  1.61it/s]
 80%|████████  | 52562/65536 [9:03:30<2:13:27,  1.62it/s]
 80%|████████  | 52563/65536 [9:03:30<2:11:58,  1.64it/s]
 80%|████████  | 52564/65536 [9:03:31<2:09:01,  1.68it/s]
 80%|████████  | 52565/65536 [9:03:32<2:11:37,  1.64it/s]
 80%|████████  | 52566/65536 [9:03:32<2:14:22,  1.61it/s]
 80%|████████  | 52567/65536 [9:03:33<2:13:31,  1.62it/s]
 80%|████████  | 52568/65536 [9:03:34<2:16:13,  1.59it/s]
 80%|████████  | 52569/65536 [9:03:34<2:16:08,  1.59it/s]
 80%|████████  | 52570/65536 [9:03:35<2:16:43,  1.58it/s]
 80%|████████  | 52571/65536 [9:03:35<2:16:51,  1.58it/s]
 80%|████████  | 52572/65536 [9:03:36<2:17:49,  1.57it/s]
 80%|████████  | 52573/65536 [9:03:37<2:13:33,  1.62it/s]
 80%|████████  | 52574/65536 [9:03:37<2:18:35,  1.56it/s]
 80%|████████  | 52575/65536 [9:03:38<2:17:59,  1.57it/s]
 80%|████████  | 52576/65536 [9:03:39<2:17:11,  1.57it/s]
 80%|████████  | 52577/65536 [9:03:39<2:15:54,  1.59it/s]
 80%|████████  | 52578/65536 [9:03:40<2:11:46,  1.64it/s]
 80%|████████  | 52579/65536 [9:03:40<2:13:46,  1.61it/s]
 80%|████████  | 52580/65536 [9:03:41<2:14:44,  1.60it/s]
                                                         
{'loss': 1.5567, 'learning_rate': 2.7846822578670255e-07, 'epoch': 3245.68}
+
 80%|████████  | 52580/65536 [9:03:41<2:14:44,  1.60it/s]
 80%|████████  | 52581/65536 [9:03:42<2:14:44,  1.60it/s]
 80%|████████  | 52582/65536 [9:03:42<2:15:54,  1.59it/s]
 80%|████████  | 52583/65536 [9:03:43<2:15:05,  1.60it/s]
 80%|████████  | 52584/65536 [9:03:44<2:11:37,  1.64it/s]
 80%|████████  | 52585/65536 [9:03:44<2:11:21,  1.64it/s]
 80%|████████  | 52586/65536 [9:03:45<2:20:08,  1.54it/s]
 80%|████████  | 52587/65536 [9:03:45<2:16:49,  1.58it/s]
 80%|████████  | 52588/65536 [9:03:46<2:13:20,  1.62it/s]
 80%|████████  | 52589/65536 [9:03:47<2:13:36,  1.62it/s]
 80%|████████  | 52590/65536 [9:03:47<2:11:09,  1.64it/s]
 80%|████████  | 52591/65536 [9:03:48<2:14:27,  1.60it/s]
 80%|████████  | 52592/65536 [9:03:49<2:12:11,  1.63it/s]
 80%|████████  | 52593/65536 [9:03:49<2:10:45,  1.65it/s]
 80%|████████  | 52594/65536 [9:03:50<2:17:03,  1.57it/s]
 80%|████████  | 52595/65536 [9:03:50<2:16:28,  1.58it/s]
 80%|████████  | 52596/65536 [9:03:51<2:16:35,  1.58it/s]
 80%|████████  | 52597/65536 [9:03:52<2:13:30,  1.62it/s]
 80%|████████  | 52598/65536 [9:03:52<2:14:38,  1.60it/s]
 80%|████████  | 52599/65536 [9:03:53<2:10:51,  1.65it/s]
 80%|████████  | 52600/65536 [9:03:53<2:12:06,  1.63it/s]
                                                         
{'loss': 1.5631, 'learning_rate': 2.781927268274764e-07, 'epoch': 3246.91}
+
 80%|████████  | 52600/65536 [9:03:53<2:12:06,  1.63it/s]
 80%|████████  | 52601/65536 [9:03:54<2:11:23,  1.64it/s]
 80%|████████  | 52602/65536 [9:03:55<2:16:34,  1.58it/s]
 80%|████████  | 52603/65536 [9:03:55<2:16:06,  1.58it/s]
 80%|████████  | 52604/65536 [9:03:56<2:16:24,  1.58it/s]
 80%|████████  | 52605/65536 [9:03:57<2:16:15,  1.58it/s]
 80%|████████  | 52606/65536 [9:03:57<2:18:44,  1.55it/s]
 80%|████████  | 52607/65536 [9:03:58<2:19:23,  1.55it/s]
 80%|████████  | 52608/65536 [9:03:59<2:15:31,  1.59it/s]
 80%|████████  | 52609/65536 [9:03:59<2:12:37,  1.62it/s]
 80%|████████  | 52610/65536 [9:04:00<2:12:56,  1.62it/s]
 80%|████████  | 52611/65536 [9:04:00<2:13:18,  1.62it/s]
 80%|████████  | 52612/65536 [9:04:01<2:16:03,  1.58it/s]
 80%|████████  | 52613/65536 [9:04:02<2:15:38,  1.59it/s]
 80%|████████  | 52614/65536 [9:04:02<2:14:36,  1.60it/s]
 80%|████████  | 52615/65536 [9:04:03<2:11:22,  1.64it/s]
 80%|████████  | 52616/65536 [9:04:03<2:11:11,  1.64it/s]
 80%|████████  | 52617/65536 [9:04:04<2:10:24,  1.65it/s]
 80%|████████  | 52618/65536 [9:04:05<2:14:30,  1.60it/s]
 80%|████████  | 52619/65536 [9:04:05<2:13:24,  1.61it/s]
 80%|████████  | 52620/65536 [9:04:06<2:11:32,  1.64it/s]
                                                         
{'loss': 1.5841, 'learning_rate': 2.7791722786825026e-07, 'epoch': 3248.15}
+
 80%|████████  | 52620/65536 [9:04:06<2:11:32,  1.64it/s]
 80%|████████  | 52621/65536 [9:04:07<2:10:08,  1.65it/s]
 80%|████████  | 52622/65536 [9:04:07<2:12:42,  1.62it/s]
 80%|████████  | 52623/65536 [9:04:08<2:10:56,  1.64it/s]
 80%|████████  | 52624/65536 [9:04:08<2:10:17,  1.65it/s]
 80%|████████  | 52625/65536 [9:04:09<2:10:10,  1.65it/s]
 80%|████████  | 52626/65536 [9:04:10<2:09:00,  1.67it/s]
 80%|████████  | 52627/65536 [9:04:10<2:07:58,  1.68it/s]
 80%|████████  | 52628/65536 [9:04:11<2:07:53,  1.68it/s]
 80%|████████  | 52629/65536 [9:04:11<2:07:16,  1.69it/s]
 80%|████████  | 52630/65536 [9:04:12<2:14:36,  1.60it/s]
 80%|████████  | 52631/65536 [9:04:13<2:19:37,  1.54it/s]
 80%|████████  | 52632/65536 [9:04:13<2:17:11,  1.57it/s]
 80%|████████  | 52633/65536 [9:04:14<2:14:16,  1.60it/s]
 80%|████████  | 52634/65536 [9:04:15<2:17:56,  1.56it/s]
 80%|████████  | 52635/65536 [9:04:15<2:15:52,  1.58it/s]
 80%|████████  | 52636/65536 [9:04:16<2:18:56,  1.55it/s]
 80%|████████  | 52637/65536 [9:04:17<2:16:18,  1.58it/s]
 80%|████████  | 52638/65536 [9:04:17<2:13:50,  1.61it/s]
 80%|████████  | 52639/65536 [9:04:18<2:14:37,  1.60it/s]
 80%|████████  | 52640/65536 [9:04:18<2:13:41,  1.61it/s]
                                                         
{'loss': 1.5499, 'learning_rate': 2.776417289090241e-07, 'epoch': 3249.38}
+
 80%|████████  | 52640/65536 [9:04:18<2:13:41,  1.61it/s]
 80%|████████  | 52641/65536 [9:04:19<2:14:47,  1.59it/s]
 80%|████████  | 52642/65536 [9:04:20<2:11:36,  1.63it/s]
 80%|████████  | 52643/65536 [9:04:20<2:11:41,  1.63it/s]
 80%|████████  | 52644/65536 [9:04:21<2:09:28,  1.66it/s]
 80%|████████  | 52645/65536 [9:04:21<2:10:56,  1.64it/s]
 80%|████████  | 52646/65536 [9:04:22<2:12:02,  1.63it/s]
 80%|████████  | 52647/65536 [9:04:23<2:12:16,  1.62it/s]
 80%|████████  | 52648/65536 [9:04:23<2:09:04,  1.66it/s]
 80%|████████  | 52649/65536 [9:04:24<2:12:21,  1.62it/s]
 80%|████████  | 52650/65536 [9:04:24<2:11:40,  1.63it/s]
 80%|████████  | 52651/65536 [9:04:25<2:15:09,  1.59it/s]
 80%|████████  | 52652/65536 [9:04:26<2:13:06,  1.61it/s]
 80%|████████  | 52653/65536 [9:04:26<2:12:42,  1.62it/s]
 80%|████████  | 52654/65536 [9:04:27<2:10:37,  1.64it/s]
 80%|████████  | 52655/65536 [9:04:28<2:10:59,  1.64it/s]
 80%|████████  | 52656/65536 [9:04:28<2:11:35,  1.63it/s]
 80%|████████  | 52657/65536 [9:04:29<2:09:45,  1.65it/s]
 80%|████████  | 52658/65536 [9:04:29<2:12:12,  1.62it/s]
 80%|████████  | 52659/65536 [9:04:30<2:13:28,  1.61it/s]
 80%|████████  | 52660/65536 [9:04:31<2:14:30,  1.60it/s]
                                                         
{'loss': 1.5575, 'learning_rate': 2.77366229949798e-07, 'epoch': 3250.62}
+
 80%|████████  | 52660/65536 [9:04:31<2:14:30,  1.60it/s]
 80%|████████  | 52661/65536 [9:04:31<2:13:37,  1.61it/s]
 80%|████████  | 52662/65536 [9:04:32<2:15:28,  1.58it/s]
 80%|████████  | 52663/65536 [9:04:33<2:13:59,  1.60it/s]
 80%|████████  | 52664/65536 [9:04:33<2:13:52,  1.60it/s]
 80%|████████  | 52665/65536 [9:04:34<2:15:14,  1.59it/s]
 80%|████████  | 52666/65536 [9:04:34<2:14:01,  1.60it/s]
 80%|████████  | 52667/65536 [9:04:35<2:15:37,  1.58it/s]
 80%|████████  | 52668/65536 [9:04:36<2:17:46,  1.56it/s]
 80%|████████  | 52669/65536 [9:04:36<2:16:49,  1.57it/s]
 80%|████████  | 52670/65536 [9:04:37<2:13:24,  1.61it/s]
 80%|████████  | 52671/65536 [9:04:38<2:13:24,  1.61it/s]
 80%|████████  | 52672/65536 [9:04:38<2:11:09,  1.63it/s]
 80%|████████  | 52673/65536 [9:04:39<2:08:43,  1.67it/s]
 80%|████████  | 52674/65536 [9:04:39<2:07:16,  1.68it/s]
 80%|████████  | 52675/65536 [9:04:40<2:07:12,  1.69it/s]
 80%|████████  | 52676/65536 [9:04:41<2:11:24,  1.63it/s]
 80%|████████  | 52677/65536 [9:04:41<2:11:37,  1.63it/s]
 80%|████████  | 52678/65536 [9:04:42<2:13:12,  1.61it/s]
 80%|████████  | 52679/65536 [9:04:42<2:13:51,  1.60it/s]
 80%|████████  | 52680/65536 [9:04:43<2:13:41,  1.60it/s]
                                                         
{'loss': 1.5516, 'learning_rate': 2.7709073099057184e-07, 'epoch': 3251.85}
+
 80%|████████  | 52680/65536 [9:04:43<2:13:41,  1.60it/s]
 80%|████████  | 52681/65536 [9:04:44<2:13:10,  1.61it/s]
 80%|████████  | 52682/65536 [9:04:44<2:11:26,  1.63it/s]
 80%|████████  | 52683/65536 [9:04:45<2:15:43,  1.58it/s]
 80%|████████  | 52684/65536 [9:04:46<2:15:36,  1.58it/s]
 80%|████████  | 52685/65536 [9:04:46<2:12:57,  1.61it/s]
 80%|████████  | 52686/65536 [9:04:47<2:16:24,  1.57it/s]
 80%|████████  | 52687/65536 [9:04:47<2:14:49,  1.59it/s]
 80%|████████  | 52688/65536 [9:04:48<2:21:23,  1.51it/s]
 80%|████████  | 52689/65536 [9:04:49<2:19:05,  1.54it/s]
 80%|████████  | 52690/65536 [9:04:49<2:14:44,  1.59it/s]
 80%|████████  | 52691/65536 [9:04:50<2:14:20,  1.59it/s]
 80%|████████  | 52692/65536 [9:04:51<2:13:24,  1.60it/s]
 80%|████████  | 52693/65536 [9:04:51<2:12:22,  1.62it/s]
 80%|████████  | 52694/65536 [9:04:52<2:12:26,  1.62it/s]
 80%|████████  | 52695/65536 [9:04:52<2:11:54,  1.62it/s]
 80%|████████  | 52696/65536 [9:04:53<2:10:35,  1.64it/s]
 80%|████████  | 52697/65536 [9:04:54<2:10:16,  1.64it/s]
 80%|████████  | 52698/65536 [9:04:54<2:09:40,  1.65it/s]
 80%|████████  | 52699/65536 [9:04:55<2:11:29,  1.63it/s]
 80%|████████  | 52700/65536 [9:04:56<2:11:46,  1.62it/s]
                                                         
{'loss': 1.5494, 'learning_rate': 2.768152320313456e-07, 'epoch': 3253.09}
+
 80%|████████  | 52700/65536 [9:04:56<2:11:46,  1.62it/s]
 80%|████████  | 52701/65536 [9:04:56<2:11:18,  1.63it/s]
 80%|████████  | 52702/65536 [9:04:57<2:09:50,  1.65it/s]
 80%|████████  | 52703/65536 [9:04:57<2:10:09,  1.64it/s]
 80%|████████  | 52704/65536 [9:04:58<2:13:32,  1.60it/s]
 80%|████████  | 52705/65536 [9:04:59<2:17:09,  1.56it/s]
 80%|████████  | 52706/65536 [9:04:59<2:13:33,  1.60it/s]
 80%|████████  | 52707/65536 [9:05:00<2:12:53,  1.61it/s]
 80%|████████  | 52708/65536 [9:05:01<2:15:32,  1.58it/s]
 80%|████████  | 52709/65536 [9:05:01<2:17:05,  1.56it/s]
 80%|████████  | 52710/65536 [9:05:02<2:12:34,  1.61it/s]
 80%|████████  | 52711/65536 [9:05:02<2:09:42,  1.65it/s]
 80%|████████  | 52712/65536 [9:05:03<2:08:54,  1.66it/s]
 80%|████████  | 52713/65536 [9:05:04<2:10:04,  1.64it/s]
 80%|████████  | 52714/65536 [9:05:04<2:13:31,  1.60it/s]
 80%|████████  | 52715/65536 [9:05:05<2:15:18,  1.58it/s]
 80%|████████  | 52716/65536 [9:05:05<2:12:20,  1.61it/s]
 80%|████████  | 52717/65536 [9:05:06<2:13:07,  1.60it/s]
 80%|████████  | 52718/65536 [9:05:07<2:09:35,  1.65it/s]
 80%|████████  | 52719/65536 [9:05:07<2:11:31,  1.62it/s]
 80%|████████  | 52720/65536 [9:05:08<2:12:39,  1.61it/s]
                                                         
{'loss': 1.5336, 'learning_rate': 2.7653973307211945e-07, 'epoch': 3254.32}
+
 80%|████████  | 52720/65536 [9:05:08<2:12:39,  1.61it/s]
 80%|████████  | 52721/65536 [9:05:09<2:10:34,  1.64it/s]
 80%|████████  | 52722/65536 [9:05:09<2:11:02,  1.63it/s]
 80%|████████  | 52723/65536 [9:05:10<2:11:16,  1.63it/s]
 80%|████████  | 52724/65536 [9:05:10<2:14:32,  1.59it/s]
 80%|████████  | 52725/65536 [9:05:11<2:14:34,  1.59it/s]
 80%|████████  | 52726/65536 [9:05:12<2:11:54,  1.62it/s]
 80%|████████  | 52727/65536 [9:05:12<2:11:10,  1.63it/s]
 80%|████████  | 52728/65536 [9:05:13<2:10:51,  1.63it/s]
 80%|████████  | 52729/65536 [9:05:13<2:10:35,  1.63it/s]
 80%|████████  | 52730/65536 [9:05:14<2:11:35,  1.62it/s]
 80%|████████  | 52731/65536 [9:05:15<2:09:28,  1.65it/s]
 80%|████████  | 52732/65536 [9:05:15<2:19:03,  1.53it/s]
 80%|████████  | 52733/65536 [9:05:16<2:18:54,  1.54it/s]
 80%|████████  | 52734/65536 [9:05:17<2:16:45,  1.56it/s]
 80%|████████  | 52735/65536 [9:05:17<2:14:10,  1.59it/s]
 80%|██���█████  | 52736/65536 [9:05:18<2:14:29,  1.59it/s]
 80%|████████  | 52737/65536 [9:05:19<2:12:38,  1.61it/s]
 80%|████████  | 52738/65536 [9:05:19<2:12:18,  1.61it/s]
 80%|████████  | 52739/65536 [9:05:20<2:10:48,  1.63it/s]
 80%|████████  | 52740/65536 [9:05:20<2:09:23,  1.65it/s]
                                                         
{'loss': 1.5654, 'learning_rate': 2.762642341128933e-07, 'epoch': 3255.56}
+
 80%|████████  | 52740/65536 [9:05:20<2:09:23,  1.65it/s]
 80%|████████  | 52741/65536 [9:05:21<2:10:43,  1.63it/s]
 80%|████████  | 52742/65536 [9:05:22<2:12:29,  1.61it/s]
 80%|████████  | 52743/65536 [9:05:22<2:10:03,  1.64it/s]
 80%|████████  | 52744/65536 [9:05:23<2:08:20,  1.66it/s]
 80%|████████  | 52745/65536 [9:05:23<2:10:47,  1.63it/s]
 80%|████████  | 52746/65536 [9:05:24<2:08:48,  1.65it/s]
 80%|████████  | 52747/65536 [9:05:25<2:06:07,  1.69it/s]
 80%|████████  | 52748/65536 [9:05:25<2:12:09,  1.61it/s]
 80%|████████  | 52749/65536 [9:05:26<2:14:09,  1.59it/s]
 80%|████████  | 52750/65536 [9:05:27<2:12:31,  1.61it/s]
 80%|████████  | 52751/65536 [9:05:27<2:18:09,  1.54it/s]
 80%|████████  | 52752/65536 [9:05:28<2:15:46,  1.57it/s]
 80%|████████  | 52753/65536 [9:05:28<2:13:49,  1.59it/s]
 80%|████████  | 52754/65536 [9:05:29<2:16:37,  1.56it/s]
 80%|████████  | 52755/65536 [9:05:30<2:13:25,  1.60it/s]
 80%|████████  | 52756/65536 [9:05:30<2:12:36,  1.61it/s]
 81%|████████  | 52757/65536 [9:05:31<2:10:50,  1.63it/s]
 81%|████████  | 52758/65536 [9:05:32<2:12:17,  1.61it/s]
 81%|████████  | 52759/65536 [9:05:32<2:08:43,  1.65it/s]
 81%|████████  | 52760/65536 [9:05:33<2:10:05,  1.64it/s]
                                                         
{'loss': 1.5785, 'learning_rate': 2.759887351536672e-07, 'epoch': 3256.79}
+
 81%|████████  | 52760/65536 [9:05:33<2:10:05,  1.64it/s]
 81%|████████  | 52761/65536 [9:05:33<2:09:56,  1.64it/s]
 81%|████████  | 52762/65536 [9:05:34<2:08:48,  1.65it/s]
 81%|████████  | 52763/65536 [9:05:35<2:08:49,  1.65it/s]
 81%|████████  | 52764/65536 [9:05:35<2:14:11,  1.59it/s]
 81%|████████  | 52765/65536 [9:05:36<2:12:46,  1.60it/s]
 81%|████████  | 52766/65536 [9:05:36<2:10:11,  1.63it/s]
 81%|████████  | 52767/65536 [9:05:37<2:12:48,  1.60it/s]
 81%|████████  | 52768/65536 [9:05:38<2:09:17,  1.65it/s]
 81%|████████  | 52769/65536 [9:05:38<2:08:17,  1.66it/s]
 81%|████████  | 52770/65536 [9:05:39<2:11:31,  1.62it/s]
 81%|████████  | 52771/65536 [9:05:40<2:14:20,  1.58it/s]
 81%|████████  | 52772/65536 [9:05:40<2:13:53,  1.59it/s]
 81%|████████  | 52773/65536 [9:05:41<2:16:33,  1.56it/s]
 81%|████████  | 52774/65536 [9:05:42<2:16:00,  1.56it/s]
 81%|████████  | 52775/65536 [9:05:42<2:14:06,  1.59it/s]
 81%|████████  | 52776/65536 [9:05:43<2:10:20,  1.63it/s]
 81%|████████  | 52777/65536 [9:05:43<2:09:59,  1.64it/s]
 81%|████████  | 52778/65536 [9:05:44<2:13:54,  1.59it/s]
 81%|████████  | 52779/65536 [9:05:45<2:10:27,  1.63it/s]
 81%|████████  | 52780/65536 [9:05:45<2:12:41,  1.60it/s]
                                                         
{'loss': 1.5694, 'learning_rate': 2.757132361944411e-07, 'epoch': 3258.02}
+
 81%|████████  | 52780/65536 [9:05:45<2:12:41,  1.60it/s]
 81%|████████  | 52781/65536 [9:05:46<2:17:03,  1.55it/s]
 81%|████████  | 52782/65536 [9:05:47<2:20:10,  1.52it/s]
 81%|████████  | 52783/65536 [9:05:47<2:17:39,  1.54it/s]
 81%|████████  | 52784/65536 [9:05:48<2:14:36,  1.58it/s]
 81%|████████  | 52785/65536 [9:05:48<2:10:15,  1.63it/s]
 81%|████████  | 52786/65536 [9:05:49<2:09:27,  1.64it/s]
 81%|████████  | 52787/65536 [9:05:50<2:07:08,  1.67it/s]
 81%|████████  | 52788/65536 [9:05:50<2:06:53,  1.67it/s]
 81%|████████  | 52789/65536 [9:05:51<2:06:55,  1.67it/s]
 81%|████████  | 52790/65536 [9:05:51<2:09:52,  1.64it/s]
 81%|████████  | 52791/65536 [9:05:52<2:09:36,  1.64it/s]
 81%|████████  | 52792/65536 [9:05:53<2:08:53,  1.65it/s]
 81%|████████  | 52793/65536 [9:05:53<2:06:30,  1.68it/s]
 81%|████████  | 52794/65536 [9:05:54<2:08:43,  1.65it/s]
 81%|████████  | 52795/65536 [9:05:54<2:11:32,  1.61it/s]
 81%|████████  | 52796/65536 [9:05:55<2:13:15,  1.59it/s]
 81%|████████  | 52797/65536 [9:05:56<2:11:51,  1.61it/s]
 81%|████████  | 52798/65536 [9:05:56<2:12:35,  1.60it/s]
 81%|████████  | 52799/65536 [9:05:57<2:10:17,  1.63it/s]
 81%|████████  | 52800/65536 [9:05:58<2:14:52,  1.57it/s]
                                                         
{'loss': 1.5759, 'learning_rate': 2.7543773723521494e-07, 'epoch': 3259.26}
+
 81%|████████  | 52800/65536 [9:05:58<2:14:52,  1.57it/s]
 81%|████████  | 52801/65536 [9:05:58<2:14:54,  1.57it/s]
 81%|████████  | 52802/65536 [9:05:59<2:12:12,  1.61it/s]
 81%|████████  | 52803/65536 [9:05:59<2:12:56,  1.60it/s]
 81%|████████  | 52804/65536 [9:06:00<2:13:03,  1.59it/s]
 81%|████████  | 52805/65536 [9:06:01<2:12:30,  1.60it/s]
 81%|████████  | 52806/65536 [9:06:01<2:13:13,  1.59it/s]
 81%|████████  | 52807/65536 [9:06:02<2:10:35,  1.62it/s]
 81%|████████  | 52808/65536 [9:06:02<2:06:30,  1.68it/s]
 81%|████████  | 52809/65536 [9:06:03<2:08:07,  1.66it/s]
 81%|████████  | 52810/65536 [9:06:04<2:10:03,  1.63it/s]
 81%|████████  | 52811/65536 [9:06:04<2:10:23,  1.63it/s]
 81%|████████  | 52812/65536 [9:06:05<2:10:25,  1.63it/s]
 81%|████████  | 52813/65536 [9:06:06<2:16:57,  1.55it/s]
 81%|████████  | 52814/65536 [9:06:06<2:18:27,  1.53it/s]
 81%|████████  | 52815/65536 [9:06:07<2:14:58,  1.57it/s]
 81%|████████  | 52816/65536 [9:06:08<2:12:54,  1.60it/s]
 81%|████████  | 52817/65536 [9:06:08<2:10:38,  1.62it/s]
 81%|████████  | 52818/65536 [9:06:09<2:09:05,  1.64it/s]
 81%|████████  | 52819/65536 [9:06:09<2:09:10,  1.64it/s]
 81%|████████  | 52820/65536 [9:06:10<2:10:09,  1.63it/s]
                                                         
{'loss': 1.5569, 'learning_rate': 2.751622382759888e-07, 'epoch': 3260.49}
+
 81%|████████  | 52820/65536 [9:06:10<2:10:09,  1.63it/s]
 81%|████████  | 52821/65536 [9:06:11<2:11:48,  1.61it/s]
 81%|████████  | 52822/65536 [9:06:11<2:08:17,  1.65it/s]
 81%|████████  | 52823/65536 [9:06:12<2:06:54,  1.67it/s]
 81%|████████  | 52824/65536 [9:06:13<2:16:03,  1.56it/s]
 81%|████████  | 52825/65536 [9:06:13<2:13:50,  1.58it/s]
 81%|████████  | 52826/65536 [9:06:14<2:12:14,  1.60it/s]
 81%|████████  | 52827/65536 [9:06:14<2:14:20,  1.58it/s]
 81%|████████  | 52828/65536 [9:06:15<2:13:02,  1.59it/s]
 81%|████████  | 52829/65536 [9:06:16<2:16:00,  1.56it/s]
 81%|████████  | 52830/65536 [9:06:16<2:13:10,  1.59it/s]
 81%|████████  | 52831/65536 [9:06:17<2:11:32,  1.61it/s]
 81%|████████  | 52832/65536 [9:06:17<2:09:07,  1.64it/s]
 81%|████████  | 52833/65536 [9:06:18<2:08:26,  1.65it/s]
 81%|████████  | 52834/65536 [9:06:19<2:09:13,  1.64it/s]
 81%|████████  | 52835/65536 [9:06:19<2:12:07,  1.60it/s]
 81%|████████  | 52836/65536 [9:06:20<2:09:30,  1.63it/s]
 81%|████████  | 52837/65536 [9:06:21<2:10:27,  1.62it/s]
 81%|████████  | 52838/65536 [9:06:21<2:12:03,  1.60it/s]
 81%|████████  | 52839/65536 [9:06:22<2:10:27,  1.62it/s]
 81%|████████  | 52840/65536 [9:06:22<2:13:27,  1.59it/s]
                                                         
{'loss': 1.5951, 'learning_rate': 2.7488673931676255e-07, 'epoch': 3261.73}
+
 81%|████████  | 52840/65536 [9:06:22<2:13:27,  1.59it/s]
 81%|████████  | 52841/65536 [9:06:23<2:13:55,  1.58it/s]
 81%|████████  | 52842/65536 [9:06:24<2:10:36,  1.62it/s]
 81%|████████  | 52843/65536 [9:06:24<2:08:56,  1.64it/s]
 81%|████████  | 52844/65536 [9:06:25<2:08:01,  1.65it/s]
 81%|████████  | 52845/65536 [9:06:26<2:12:14,  1.60it/s]
 81%|████████  | 52846/65536 [9:06:26<2:12:33,  1.60it/s]
 81%|████████  | 52847/65536 [9:06:27<2:10:34,  1.62it/s]
 81%|████████  | 52848/65536 [9:06:27<2:08:30,  1.65it/s]
 81%|████████  | 52849/65536 [9:06:28<2:08:17,  1.65it/s]
 81%|████████  | 52850/65536 [9:06:29<2:09:34,  1.63it/s]
 81%|████████  | 52851/65536 [9:06:29<2:08:00,  1.65it/s]
 81%|████████  | 52852/65536 [9:06:30<2:11:52,  1.60it/s]
 81%|████████  | 52853/65536 [9:06:30<2:09:13,  1.64it/s]
 81%|████████  | 52854/65536 [9:06:31<2:09:43,  1.63it/s]
 81%|████████  | 52855/65536 [9:06:32<2:08:47,  1.64it/s]
 81%|████████  | 52856/65536 [9:06:32<2:08:32,  1.64it/s]
 81%|████████  | 52857/65536 [9:06:33<2:10:41,  1.62it/s]
 81%|████████  | 52858/65536 [9:06:34<2:12:19,  1.60it/s]
 81%|████████  | 52859/65536 [9:06:34<2:13:07,  1.59it/s]
 81%|████████  | 52860/65536 [9:06:35<2:09:28,  1.63it/s]
                                                         
{'loss': 1.5815, 'learning_rate': 2.746112403575364e-07, 'epoch': 3262.96}
+
 81%|████████  | 52860/65536 [9:06:35<2:09:28,  1.63it/s]
 81%|████████  | 52861/65536 [9:06:35<2:16:28,  1.55it/s]
 81%|████████  | 52862/65536 [9:06:36<2:18:02,  1.53it/s]
 81%|████████  | 52863/65536 [9:06:37<2:17:41,  1.53it/s]
 81%|████████  | 52864/65536 [9:06:37<2:16:03,  1.55it/s]
 81%|████████  | 52865/65536 [9:06:38<2:15:36,  1.56it/s]
 81%|████████  | 52866/65536 [9:06:39<2:11:46,  1.60it/s]
 81%|████████  | 52867/65536 [9:06:39<2:13:26,  1.58it/s]
 81%|████████  | 52868/65536 [9:06:40<2:11:20,  1.61it/s]
 81%|████████  | 52869/65536 [9:06:40<2:07:31,  1.66it/s]
 81%|████████  | 52870/65536 [9:06:41<2:09:18,  1.63it/s]
 81%|████████  | 52871/65536 [9:06:42<2:09:18,  1.63it/s]
 81%|████████  | 52872/65536 [9:06:42<2:08:19,  1.64it/s]
 81%|████████  | 52873/65536 [9:06:43<2:07:53,  1.65it/s]
 81%|████████  | 52874/65536 [9:06:43<2:07:33,  1.65it/s]
 81%|████████  | 52875/65536 [9:06:44<2:09:04,  1.63it/s]
 81%|████████  | 52876/65536 [9:06:45<2:07:53,  1.65it/s]
 81%|████████  | 52877/65536 [9:06:45<2:09:23,  1.63it/s]
 81%|████████  | 52878/65536 [9:06:46<2:08:17,  1.64it/s]
 81%|████████  | 52879/65536 [9:06:47<2:09:33,  1.63it/s]
 81%|████████  | 52880/65536 [9:06:47<2:08:40,  1.64it/s]
                                                         
{'loss': 1.5931, 'learning_rate': 2.7433574139831027e-07, 'epoch': 3264.2}
+
 81%|████████  | 52880/65536 [9:06:47<2:08:40,  1.64it/s]
 81%|████████  | 52881/65536 [9:06:48<2:07:38,  1.65it/s]
 81%|████████  | 52882/65536 [9:06:48<2:08:27,  1.64it/s]
 81%|████████  | 52883/65536 [9:06:49<2:07:13,  1.66it/s]
 81%|████████  | 52884/65536 [9:06:50<2:08:29,  1.64it/s]
 81%|████████  | 52885/65536 [9:06:50<2:13:35,  1.58it/s]
 81%|████████  | 52886/65536 [9:06:51<2:11:40,  1.60it/s]
 81%|████████  | 52887/65536 [9:06:51<2:11:53,  1.60it/s]
 81%|████████  | 52888/65536 [9:06:52<2:13:23,  1.58it/s]
 81%|████████  | 52889/65536 [9:06:53<2:12:27,  1.59it/s]
 81%|████████  | 52890/65536 [9:06:53<2:14:53,  1.56it/s]
 81%|████████  | 52891/65536 [9:06:54<2:09:40,  1.63it/s]
 81%|████████  | 52892/65536 [9:06:55<2:08:37,  1.64it/s]
 81%|████████  | 52893/65536 [9:06:55<2:08:24,  1.64it/s]
 81%|████████  | 52894/65536 [9:06:56<2:13:24,  1.58it/s]
 81%|████████  | 52895/65536 [9:06:56<2:10:42,  1.61it/s]
 81%|████████  | 52896/65536 [9:06:57<2:13:02,  1.58it/s]
 81%|████████  | 52897/65536 [9:06:58<2:10:45,  1.61it/s]
 81%|████████  | 52898/65536 [9:06:58<2:07:39,  1.65it/s]
 81%|████████  | 52899/65536 [9:06:59<2:08:35,  1.64it/s]
 81%|████████  | 52900/65536 [9:07:00<2:08:09,  1.64it/s]
                                                         
{'loss': 1.5551, 'learning_rate': 2.7406024243908413e-07, 'epoch': 3265.43}
+
 81%|████████  | 52900/65536 [9:07:00<2:08:09,  1.64it/s]
 81%|████████  | 52901/65536 [9:07:00<2:07:17,  1.65it/s]
 81%|████████  | 52902/65536 [9:07:01<2:06:24,  1.67it/s]
 81%|████████  | 52903/65536 [9:07:01<2:06:42,  1.66it/s]
 81%|████████  | 52904/65536 [9:07:02<2:09:01,  1.63it/s]
 81%|████████  | 52905/65536 [9:07:03<2:07:19,  1.65it/s]
 81%|████████  | 52906/65536 [9:07:03<2:08:21,  1.64it/s]
 81%|████████  | 52907/65536 [9:07:04<2:08:19,  1.64it/s]
 81%|████████  | 52908/65536 [9:07:04<2:10:54,  1.61it/s]
 81%|████████  | 52909/65536 [9:07:05<2:15:54,  1.55it/s]
 81%|████████  | 52910/65536 [9:07:06<2:16:24,  1.54it/s]
 81%|████████  | 52911/65536 [9:07:06<2:16:03,  1.55it/s]
 81%|████████  | 52912/65536 [9:07:07<2:15:10,  1.56it/s]
 81%|████████  | 52913/65536 [9:07:08<2:16:01,  1.55it/s]
 81%|████████  | 52914/65536 [9:07:08<2:18:39,  1.52it/s]
 81%|████████  | 52915/65536 [9:07:09<2:13:38,  1.57it/s]
 81%|████████  | 52916/65536 [9:07:10<2:12:22,  1.59it/s]
 81%|████████  | 52917/65536 [9:07:10<2:10:45,  1.61it/s]
 81%|████████  | 52918/65536 [9:07:11<2:11:13,  1.60it/s]
 81%|████████  | 52919/65536 [9:07:11<2:11:10,  1.60it/s]
 81%|████████  | 52920/65536 [9:07:12<2:14:37,  1.56it/s]
                                                         
{'loss': 1.5237, 'learning_rate': 2.73784743479858e-07, 'epoch': 3266.67}
+
 81%|████████  | 52920/65536 [9:07:12<2:14:37,  1.56it/s]
 81%|████████  | 52921/65536 [9:07:13<2:11:05,  1.60it/s]
 81%|████████  | 52922/65536 [9:07:13<2:11:20,  1.60it/s]
 81%|████████  | 52923/65536 [9:07:14<2:12:51,  1.58it/s]
 81%|████████  | 52924/65536 [9:07:15<2:09:09,  1.63it/s]
 81%|████████  | 52925/65536 [9:07:15<2:09:07,  1.63it/s]
 81%|████████  | 52926/65536 [9:07:16<2:12:44,  1.58it/s]
 81%|████████  | 52927/65536 [9:07:16<2:12:08,  1.59it/s]
 81%|████████  | 52928/65536 [9:07:17<2:11:32,  1.60it/s]
 81%|████████  | 52929/65536 [9:07:18<2:09:56,  1.62it/s]
 81%|████████  | 52930/65536 [9:07:18<2:08:30,  1.64it/s]
 81%|████████  | 52931/65536 [9:07:19<2:07:24,  1.65it/s]
 81%|████████  | 52932/65536 [9:07:20<2:08:42,  1.63it/s]
 81%|████████  | 52933/65536 [9:07:20<2:09:04,  1.63it/s]
 81%|████████  | 52934/65536 [9:07:21<2:06:36,  1.66it/s]
 81%|████████  | 52935/65536 [9:07:21<2:04:52,  1.68it/s]
 81%|████████  | 52936/65536 [9:07:22<2:10:25,  1.61it/s]
 81%|████████  | 52937/65536 [9:07:23<2:08:14,  1.64it/s]
 81%|████████  | 52938/65536 [9:07:23<2:06:07,  1.66it/s]
 81%|████████  | 52939/65536 [9:07:24<2:09:31,  1.62it/s]
 81%|████████  | 52940/65536 [9:07:24<2:13:43,  1.57it/s]
                                                         
{'loss': 1.5692, 'learning_rate': 2.7350924452063185e-07, 'epoch': 3267.9}
+
 81%|████████  | 52940/65536 [9:07:24<2:13:43,  1.57it/s]
 81%|████████  | 52941/65536 [9:07:25<2:11:53,  1.59it/s]
 81%|████████  | 52942/65536 [9:07:26<2:15:23,  1.55it/s]
 81%|████████  | 52943/65536 [9:07:26<2:14:14,  1.56it/s]
 81%|████████  | 52944/65536 [9:07:27<2:12:40,  1.58it/s]
 81%|████████  | 52945/65536 [9:07:28<2:11:59,  1.59it/s]
 81%|████████  | 52946/65536 [9:07:28<2:10:50,  1.60it/s]
 81%|████████  | 52947/65536 [9:07:29<2:07:42,  1.64it/s]
 81%|████████  | 52948/65536 [9:07:29<2:10:39,  1.61it/s]
 81%|████████  | 52949/65536 [9:07:30<2:08:14,  1.64it/s]
 81%|████████  | 52950/65536 [9:07:31<2:07:38,  1.64it/s]
 81%|████████  | 52951/65536 [9:07:31<2:06:32,  1.66it/s]
 81%|████████  | 52952/65536 [9:07:32<2:04:53,  1.68it/s]
 81%|████████  | 52953/65536 [9:07:32<2:05:44,  1.67it/s]
 81%|████████  | 52954/65536 [9:07:33<2:06:02,  1.66it/s]
 81%|████████  | 52955/65536 [9:07:34<2:06:28,  1.66it/s]
 81%|████████  | 52956/65536 [9:07:34<2:08:32,  1.63it/s]
 81%|████████  | 52957/65536 [9:07:35<2:12:05,  1.59it/s]
 81%|████████  | 52958/65536 [9:07:36<2:16:35,  1.53it/s]
 81%|████████  | 52959/65536 [9:07:36<2:11:32,  1.59it/s]
 81%|████████  | 52960/65536 [9:07:37<2:09:07,  1.62it/s]
                                                         
{'loss': 1.5748, 'learning_rate': 2.732337455614057e-07, 'epoch': 3269.14}
+
 81%|████████  | 52960/65536 [9:07:37<2:09:07,  1.62it/s]
 81%|████████  | 52961/65536 [9:07:37<2:08:58,  1.62it/s]
 81%|████████  | 52962/65536 [9:07:38<2:09:20,  1.62it/s]
 81%|████████  | 52963/65536 [9:07:39<2:09:30,  1.62it/s]
 81%|████████  | 52964/65536 [9:07:39<2:09:04,  1.62it/s]
 81%|████████  | 52965/65536 [9:07:40<2:09:31,  1.62it/s]
 81%|████████  | 52966/65536 [9:07:40<2:08:24,  1.63it/s]
 81%|████████  | 52967/65536 [9:07:41<2:08:47,  1.63it/s]
 81%|████████  | 52968/65536 [9:07:42<2:08:32,  1.63it/s]
 81%|████████  | 52969/65536 [9:07:42<2:10:57,  1.60it/s]
 81%|████████  | 52970/65536 [9:07:43<2:10:41,  1.60it/s]
 81%|████████  | 52971/65536 [9:07:44<2:12:21,  1.58it/s]
 81%|████████  | 52972/65536 [9:07:44<2:11:55,  1.59it/s]
 81%|████████  | 52973/65536 [9:07:45<2:16:38,  1.53it/s]
 81%|████████  | 52974/65536 [9:07:46<2:19:20,  1.50it/s]
 81%|████████  | 52975/65536 [9:07:46<2:21:49,  1.48it/s]
 81%|████████  | 52976/65536 [9:07:47<2:16:28,  1.53it/s]
 81%|████████  | 52977/65536 [9:07:48<2:13:52,  1.56it/s]
 81%|████████  | 52978/65536 [9:07:48<2:13:33,  1.57it/s]
 81%|████████  | 52979/65536 [9:07:49<2:13:59,  1.56it/s]
 81%|████████  | 52980/65536 [9:07:49<2:11:35,  1.59it/s]
                                                         
{'loss': 1.5244, 'learning_rate': 2.7295824660217946e-07, 'epoch': 3270.37}
+
 81%|████████  | 52980/65536 [9:07:49<2:11:35,  1.59it/s]
 81%|████████  | 52981/65536 [9:07:50<2:09:59,  1.61it/s]
 81%|████████  | 52982/65536 [9:07:51<2:11:53,  1.59it/s]
 81%|████████  | 52983/65536 [9:07:51<2:10:02,  1.61it/s]
 81%|████████  | 52984/65536 [9:07:52<2:12:18,  1.58it/s]
 81%|████████  | 52985/65536 [9:07:53<2:09:08,  1.62it/s]
 81%|████████  | 52986/65536 [9:07:53<2:07:12,  1.64it/s]
 81%|████████  | 52987/65536 [9:07:54<2:10:10,  1.61it/s]
 81%|████████  | 52988/65536 [9:07:54<2:08:51,  1.62it/s]
 81%|████████  | 52989/65536 [9:07:55<2:06:43,  1.65it/s]
 81%|████████  | 52990/65536 [9:07:56<2:06:16,  1.66it/s]
 81%|████████  | 52991/65536 [9:07:56<2:11:26,  1.59it/s]
 81%|████████  | 52992/65536 [9:07:57<2:11:18,  1.59it/s]
 81%|████████  | 52993/65536 [9:07:58<2:11:06,  1.59it/s]
 81%|████████  | 52994/65536 [9:07:58<2:11:54,  1.58it/s]
 81%|████████  | 52995/65536 [9:07:59<2:10:08,  1.61it/s]
 81%|████████  | 52996/65536 [9:07:59<2:08:40,  1.62it/s]
 81%|████████  | 52997/65536 [9:08:00<2:09:22,  1.62it/s]
 81%|████████  | 52998/65536 [9:08:01<2:08:20,  1.63it/s]
 81%|████████  | 52999/65536 [9:08:01<2:09:14,  1.62it/s]
 81%|████████  | 53000/65536 [9:08:02<2:09:01,  1.62it/s]
                                                         
{'loss': 1.5968, 'learning_rate': 2.726827476429533e-07, 'epoch': 3271.6}
+
 81%|████████  | 53000/65536 [9:08:02<2:09:01,  1.62it/s]
 81%|████████  | 53001/65536 [9:08:02<2:10:54,  1.60it/s]
 81%|████████  | 53002/65536 [9:08:03<2:07:30,  1.64it/s]
 81%|████████  | 53003/65536 [9:08:04<2:06:01,  1.66it/s]
 81%|████████  | 53004/65536 [9:08:04<2:06:00,  1.66it/s]
 81%|████████  | 53005/65536 [9:08:05<2:07:33,  1.64it/s]
 81%|████████  | 53006/65536 [9:08:05<2:07:43,  1.64it/s]
 81%|████████  | 53007/65536 [9:08:06<2:11:27,  1.59it/s]
 81%|████████  | 53008/65536 [9:08:07<2:10:58,  1.59it/s]
 81%|████████  | 53009/65536 [9:08:07<2:10:54,  1.59it/s]
 81%|████████  | 53010/65536 [9:08:08<2:11:39,  1.59it/s]
 81%|████████  | 53011/65536 [9:08:09<2:10:05,  1.60it/s]
 81%|████████  | 53012/65536 [9:08:09<2:11:18,  1.59it/s]
 81%|████████  | 53013/65536 [9:08:10<2:07:47,  1.63it/s]
 81%|████████  | 53014/65536 [9:08:10<2:06:34,  1.65it/s]
 81%|████████  | 53015/65536 [9:08:11<2:10:10,  1.60it/s]
 81%|████████  | 53016/65536 [9:08:12<2:11:09,  1.59it/s]
 81%|████████  | 53017/65536 [9:08:12<2:12:54,  1.57it/s]
 81%|████████  | 53018/65536 [9:08:13<2:08:35,  1.62it/s]
 81%|████████  | 53019/65536 [9:08:14<2:07:43,  1.63it/s]
 81%|████████  | 53020/65536 [9:08:14<2:04:41,  1.67it/s]
                                                         
{'loss': 1.5532, 'learning_rate': 2.724072486837272e-07, 'epoch': 3272.84}
+
 81%|████████  | 53020/65536 [9:08:14<2:04:41,  1.67it/s]
 81%|████████  | 53021/65536 [9:08:15<2:06:22,  1.65it/s]
 81%|████████  | 53022/65536 [9:08:15<2:06:27,  1.65it/s]
 81%|████████  | 53023/65536 [9:08:16<2:13:41,  1.56it/s]
 81%|████████  | 53024/65536 [9:08:17<2:12:07,  1.58it/s]
 81%|████████  | 53025/65536 [9:08:17<2:10:06,  1.60it/s]
 81%|████████  | 53026/65536 [9:08:18<2:15:57,  1.53it/s]
 81%|████████  | 53027/65536 [9:08:19<2:13:32,  1.56it/s]
 81%|████████  | 53028/65536 [9:08:19<2:09:21,  1.61it/s]
 81%|████████  | 53029/65536 [9:08:20<2:09:06,  1.61it/s]
 81%|████████  | 53030/65536 [9:08:20<2:07:30,  1.63it/s]
 81%|████████  | 53031/65536 [9:08:21<2:09:19,  1.61it/s]
 81%|████████  | 53032/65536 [9:08:22<2:06:46,  1.64it/s]
 81%|████████  | 53033/65536 [9:08:22<2:06:03,  1.65it/s]
 81%|████████  | 53034/65536 [9:08:23<2:05:30,  1.66it/s]
 81%|████████  | 53035/65536 [9:08:23<2:05:12,  1.66it/s]
 81%|████████  | 53036/65536 [9:08:24<2:08:55,  1.62it/s]
 81%|████████  | 53037/65536 [9:08:25<2:09:06,  1.61it/s]
 81%|████████  | 53038/65536 [9:08:25<2:08:18,  1.62it/s]
 81%|████████  | 53039/65536 [9:08:26<2:14:17,  1.55it/s]
 81%|████████  | 53040/65536 [9:08:27<2:12:24,  1.57it/s]
                                                         
{'loss': 1.6095, 'learning_rate': 2.7213174972450104e-07, 'epoch': 3274.07}
+
 81%|████████  | 53040/65536 [9:08:27<2:12:24,  1.57it/s]
 81%|████████  | 53041/65536 [9:08:27<2:08:24,  1.62it/s]
 81%|████████  | 53042/65536 [9:08:28<2:06:40,  1.64it/s]
 81%|████████  | 53043/65536 [9:08:28<2:08:05,  1.63it/s]
 81%|████████  | 53044/65536 [9:08:29<2:08:29,  1.62it/s]
 81%|████████  | 53045/65536 [9:08:30<2:09:59,  1.60it/s]
 81%|████████  | 53046/65536 [9:08:30<2:10:53,  1.59it/s]
 81%|████████  | 53047/65536 [9:08:31<2:08:18,  1.62it/s]
 81%|████████  | 53048/65536 [9:08:32<2:08:13,  1.62it/s]
 81%|████████  | 53049/65536 [9:08:32<2:12:18,  1.57it/s]
 81%|████████  | 53050/65536 [9:08:33<2:10:08,  1.60it/s]
 81%|████████  | 53051/65536 [9:08:33<2:06:35,  1.64it/s]
 81%|████████  | 53052/65536 [9:08:34<2:07:19,  1.63it/s]
 81%|████████  | 53053/65536 [9:08:35<2:06:17,  1.65it/s]
 81%|████████  | 53054/65536 [9:08:35<2:04:07,  1.68it/s]
 81%|████████  | 53055/65536 [9:08:36<2:08:39,  1.62it/s]
 81%|████████  | 53056/65536 [9:08:37<2:11:28,  1.58it/s]
 81%|████████  | 53057/65536 [9:08:37<2:10:53,  1.59it/s]
 81%|████████  | 53058/65536 [9:08:38<2:09:13,  1.61it/s]
 81%|████████  | 53059/65536 [9:08:38<2:08:39,  1.62it/s]
 81%|████████  | 53060/65536 [9:08:39<2:09:28,  1.61it/s]
                                                         
{'loss': 1.5268, 'learning_rate': 2.718562507652749e-07, 'epoch': 3275.31}
+
 81%|████████  | 53060/65536 [9:08:39<2:09:28,  1.61it/s]
 81%|████████  | 53061/65536 [9:08:40<2:08:28,  1.62it/s]
 81%|████████  | 53062/65536 [9:08:40<2:06:40,  1.64it/s]
 81%|████████  | 53063/65536 [9:08:41<2:05:47,  1.65it/s]
 81%|████████  | 53064/65536 [9:08:41<2:05:17,  1.66it/s]
 81%|████████  | 53065/65536 [9:08:42<2:03:38,  1.68it/s]
 81%|████████  | 53066/65536 [9:08:43<2:06:56,  1.64it/s]
 81%|████████  | 53067/65536 [9:08:43<2:13:37,  1.56it/s]
 81%|████████  | 53068/65536 [9:08:44<2:12:54,  1.56it/s]
 81%|████████  | 53069/65536 [9:08:45<2:11:43,  1.58it/s]
 81%|████████  | 53070/65536 [9:08:45<2:09:00,  1.61it/s]
 81%|████████  | 53071/65536 [9:08:46<2:10:00,  1.60it/s]
 81%|████████  | 53072/65536 [9:08:46<2:12:30,  1.57it/s]
 81%|████████  | 53073/65536 [9:08:47<2:08:57,  1.61it/s]
 81%|████████  | 53074/65536 [9:08:48<2:07:28,  1.63it/s]
 81%|████████  | 53075/65536 [9:08:48<2:06:05,  1.65it/s]
 81%|████████  | 53076/65536 [9:08:49<2:04:03,  1.67it/s]
 81%|████████  | 53077/65536 [9:08:49<2:08:05,  1.62it/s]
 81%|████████  | 53078/65536 [9:08:50<2:07:33,  1.63it/s]
 81%|████████  | 53079/65536 [9:08:51<2:09:09,  1.61it/s]
 81%|████████  | 53080/65536 [9:08:51<2:08:40,  1.61it/s]
                                                         
{'loss': 1.5713, 'learning_rate': 2.7158075180604875e-07, 'epoch': 3276.54}
+
 81%|████████  | 53080/65536 [9:08:51<2:08:40,  1.61it/s]
 81%|████████  | 53081/65536 [9:08:52<2:07:51,  1.62it/s]
 81%|████████  | 53082/65536 [9:08:53<2:07:07,  1.63it/s]
 81%|████████  | 53083/65536 [9:08:53<2:04:32,  1.67it/s]
 81%|████████  | 53084/65536 [9:08:54<2:07:38,  1.63it/s]
 81%|████████  | 53085/65536 [9:08:54<2:07:45,  1.62it/s]
 81%|████████  | 53086/65536 [9:08:55<2:05:26,  1.65it/s]
 81%|████████  | 53087/65536 [9:08:56<2:09:53,  1.60it/s]
 81%|████████  | 53088/65536 [9:08:56<2:14:23,  1.54it/s]
 81%|████████  | 53089/65536 [9:08:57<2:12:07,  1.57it/s]
 81%|████████  | 53090/65536 [9:08:58<2:10:10,  1.59it/s]
 81%|████████  | 53091/65536 [9:08:58<2:09:22,  1.60it/s]
 81%|████████  | 53092/65536 [9:08:59<2:04:55,  1.66it/s]
 81%|████████  | 53093/65536 [9:08:59<2:06:59,  1.63it/s]
 81%|████████  | 53094/65536 [9:09:00<2:05:26,  1.65it/s]
 81%|████████  | 53095/65536 [9:09:01<2:03:55,  1.67it/s]
 81%|████████  | 53096/65536 [9:09:01<2:05:57,  1.65it/s]
 81%|████████  | 53097/65536 [9:09:02<2:08:23,  1.61it/s]
 81%|████████  | 53098/65536 [9:09:02<2:08:11,  1.62it/s]
 81%|████████  | 53099/65536 [9:09:03<2:09:11,  1.60it/s]
 81%|████████  | 53100/65536 [9:09:04<2:08:03,  1.62it/s]
                                                         
{'loss': 1.5784, 'learning_rate': 2.713052528468225e-07, 'epoch': 3277.78}
+
 81%|████████  | 53100/65536 [9:09:04<2:08:03,  1.62it/s]
 81%|████████  | 53101/65536 [9:09:04<2:09:59,  1.59it/s]
 81%|████████  | 53102/65536 [9:09:05<2:11:05,  1.58it/s]
 81%|████████  | 53103/65536 [9:09:06<2:09:51,  1.60it/s]
 81%|████████  | 53104/65536 [9:09:06<2:12:30,  1.56it/s]
 81%|████████  | 53105/65536 [9:09:07<2:10:12,  1.59it/s]
 81%|████████  | 53106/65536 [9:09:07<2:07:46,  1.62it/s]
 81%|████████  | 53107/65536 [9:09:08<2:08:10,  1.62it/s]
 81%|████████  | 53108/65536 [9:09:09<2:07:18,  1.63it/s]
 81%|████████  | 53109/65536 [9:09:09<2:07:57,  1.62it/s]
 81%|████████  | 53110/65536 [9:09:10<2:07:23,  1.63it/s]
 81%|████████  | 53111/65536 [9:09:11<2:07:52,  1.62it/s]
 81%|████████  | 53112/65536 [9:09:11<2:06:23,  1.64it/s]
 81%|████████  | 53113/65536 [9:09:12<2:06:01,  1.64it/s]
 81%|████████  | 53114/65536 [9:09:12<2:07:36,  1.62it/s]
 81%|████████  | 53115/65536 [9:09:13<2:09:25,  1.60it/s]
 81%|████████  | 53116/65536 [9:09:14<2:07:34,  1.62it/s]
 81%|████████  | 53117/65536 [9:09:14<2:06:05,  1.64it/s]
 81%|████████  | 53118/65536 [9:09:15<2:07:44,  1.62it/s]
 81%|████████  | 53119/65536 [9:09:15<2:09:59,  1.59it/s]
 81%|████████  | 53120/65536 [9:09:16<2:12:42,  1.56it/s]
                                                         
{'loss': 1.5586, 'learning_rate': 2.7102975388759637e-07, 'epoch': 3279.01}
+
 81%|████████  | 53120/65536 [9:09:16<2:12:42,  1.56it/s]
 81%|████████  | 53121/65536 [9:09:17<2:10:21,  1.59it/s]
 81%|████████  | 53122/65536 [9:09:17<2:09:26,  1.60it/s]
 81%|████████  | 53123/65536 [9:09:18<2:07:24,  1.62it/s]
 81%|████████  | 53124/65536 [9:09:19<2:06:06,  1.64it/s]
 81%|████████  | 53125/65536 [9:09:19<2:04:14,  1.66it/s]
 81%|████████  | 53126/65536 [9:09:20<2:07:41,  1.62it/s]
 81%|████████  | 53127/65536 [9:09:20<2:04:56,  1.66it/s]
 81%|████████  | 53128/65536 [9:09:21<2:06:30,  1.63it/s]
 81%|████████  | 53129/65536 [9:09:22<2:07:31,  1.62it/s]
 81%|████████  | 53130/65536 [9:09:22<2:05:30,  1.65it/s]
 81%|████████  | 53131/65536 [9:09:23<2:04:44,  1.66it/s]
 81%|████████  | 53132/65536 [9:09:23<2:05:46,  1.64it/s]
 81%|████████  | 53133/65536 [9:09:24<2:08:06,  1.61it/s]
 81%|████████  | 53134/65536 [9:09:25<2:07:57,  1.62it/s]
 81%|████████  | 53135/65536 [9:09:25<2:07:26,  1.62it/s]
 81%|████████  | 53136/65536 [9:09:26<2:13:46,  1.54it/s]
 81%|████████  | 53137/65536 [9:09:27<2:14:16,  1.54it/s]
 81%|████████  | 53138/65536 [9:09:27<2:10:38,  1.58it/s]
 81%|████████  | 53139/65536 [9:09:28<2:12:04,  1.56it/s]
 81%|████████  | 53140/65536 [9:09:29<2:10:48,  1.58it/s]
                                                         
{'loss': 1.5526, 'learning_rate': 2.707542549283702e-07, 'epoch': 3280.25}
+
 81%|████████  | 53140/65536 [9:09:29<2:10:48,  1.58it/s]
 81%|████████  | 53141/65536 [9:09:29<2:09:04,  1.60it/s]
 81%|████████  | 53142/65536 [9:09:30<2:09:34,  1.59it/s]
 81%|████████  | 53143/65536 [9:09:30<2:09:31,  1.59it/s]
 81%|████████  | 53144/65536 [9:09:31<2:09:29,  1.60it/s]
 81%|████████  | 53145/65536 [9:09:32<2:07:38,  1.62it/s]
 81%|████████  | 53146/65536 [9:09:32<2:08:00,  1.61it/s]
 81%|████████  | 53147/65536 [9:09:33<2:08:21,  1.61it/s]
 81%|████████  | 53148/65536 [9:09:34<2:08:11,  1.61it/s]
 81%|████████  | 53149/65536 [9:09:34<2:11:02,  1.58it/s]
 81%|████████  | 53150/65536 [9:09:35<2:08:57,  1.60it/s]
 81%|████████  | 53151/65536 [9:09:35<2:12:08,  1.56it/s]
 81%|████████  | 53152/65536 [9:09:36<2:09:05,  1.60it/s]
 81%|████████  | 53153/65536 [9:09:37<2:12:05,  1.56it/s]
 81%|████████  | 53154/65536 [9:09:37<2:08:59,  1.60it/s]
 81%|████████  | 53155/65536 [9:09:38<2:07:30,  1.62it/s]
 81%|████████  | 53156/65536 [9:09:39<2:06:49,  1.63it/s]
 81%|████████  | 53157/65536 [9:09:39<2:07:05,  1.62it/s]
 81%|████████  | 53158/65536 [9:09:40<2:06:54,  1.63it/s]
 81%|████████  | 53159/65536 [9:09:40<2:07:11,  1.62it/s]
 81%|████████  | 53160/65536 [9:09:41<2:09:21,  1.59it/s]
                                                         
{'loss': 1.5074, 'learning_rate': 2.704787559691441e-07, 'epoch': 3281.48}
+
 81%|████████  | 53160/65536 [9:09:41<2:09:21,  1.59it/s]
 81%|████████  | 53161/65536 [9:09:42<2:06:28,  1.63it/s]
 81%|████████  | 53162/65536 [9:09:42<2:08:42,  1.60it/s]
 81%|████████  | 53163/65536 [9:09:43<2:07:21,  1.62it/s]
 81%|████████  | 53164/65536 [9:09:43<2:08:48,  1.60it/s]
 81%|████████  | 53165/65536 [9:09:44<2:09:37,  1.59it/s]
 81%|████████  | 53166/65536 [9:09:45<2:09:19,  1.59it/s]
 81%|████████  | 53167/65536 [9:09:45<2:09:14,  1.60it/s]
 81%|████████  | 53168/65536 [9:09:46<2:10:42,  1.58it/s]
 81%|████████  | 53169/65536 [9:09:47<2:15:43,  1.52it/s]
 81%|████████  | 53170/65536 [9:09:47<2:14:20,  1.53it/s]
 81%|████████  | 53171/65536 [9:09:48<2:12:53,  1.55it/s]
 81%|████████  | 53172/65536 [9:09:49<2:11:38,  1.57it/s]
 81%|████████  | 53173/65536 [9:09:49<2:09:44,  1.59it/s]
 81%|████████  | 53174/65536 [9:09:50<2:06:48,  1.62it/s]
 81%|████████  | 53175/65536 [9:09:50<2:03:38,  1.67it/s]
 81%|████████  | 53176/65536 [9:09:51<2:02:03,  1.69it/s]
 81%|████████  | 53177/65536 [9:09:52<2:05:01,  1.65it/s]
 81%|████████  | 53178/65536 [9:09:52<2:07:30,  1.62it/s]
 81%|████████  | 53179/65536 [9:09:53<2:04:17,  1.66it/s]
 81%|████████  | 53180/65536 [9:09:53<2:04:10,  1.66it/s]
                                                         
{'loss': 1.6016, 'learning_rate': 2.7020325700991794e-07, 'epoch': 3282.72}
+
 81%|████████  | 53180/65536 [9:09:53<2:04:10,  1.66it/s]
 81%|████████  | 53181/65536 [9:09:54<2:02:08,  1.69it/s]
 81%|████████  | 53182/65536 [9:09:55<2:07:51,  1.61it/s]
 81%|████████  | 53183/65536 [9:09:55<2:05:14,  1.64it/s]
 81%|████████  | 53184/65536 [9:09:56<2:03:35,  1.67it/s]
 81%|████████  | 53185/65536 [9:09:57<2:09:43,  1.59it/s]
 81%|████████  | 53186/65536 [9:09:57<2:11:55,  1.56it/s]
 81%|████████  | 53187/65536 [9:09:58<2:11:20,  1.57it/s]
 81%|████████  | 53188/65536 [9:09:58<2:09:13,  1.59it/s]
 81%|████████  | 53189/65536 [9:09:59<2:07:16,  1.62it/s]
 81%|████████  | 53190/65536 [9:10:00<2:08:04,  1.61it/s]
 81%|████████  | 53191/65536 [9:10:00<2:10:34,  1.58it/s]
 81%|████████  | 53192/65536 [9:10:01<2:07:07,  1.62it/s]
 81%|████████  | 53193/65536 [9:10:01<2:03:56,  1.66it/s]
 81%|████████  | 53194/65536 [9:10:02<2:05:22,  1.64it/s]
 81%|████████  | 53195/65536 [9:10:03<2:06:33,  1.63it/s]
 81%|████████  | 53196/65536 [9:10:03<2:06:28,  1.63it/s]
 81%|████████  | 53197/65536 [9:10:04<2:06:08,  1.63it/s]
 81%|████████  | 53198/65536 [9:10:05<2:05:55,  1.63it/s]
 81%|████████  | 53199/65536 [9:10:05<2:06:26,  1.63it/s]
 81%|████████  | 53200/65536 [9:10:06<2:06:37,  1.62it/s]
                                                         
{'loss': 1.5833, 'learning_rate': 2.699277580506918e-07, 'epoch': 3283.95}
+
 81%|████████  | 53200/65536 [9:10:06<2:06:37,  1.62it/s]
 81%|████████  | 53201/65536 [9:10:06<2:08:42,  1.60it/s]
 81%|████████  | 53202/65536 [9:10:07<2:11:50,  1.56it/s]
 81%|████████  | 53203/65536 [9:10:08<2:11:08,  1.57it/s]
 81%|████████  | 53204/65536 [9:10:08<2:08:47,  1.60it/s]
 81%|████████  | 53205/65536 [9:10:09<2:05:22,  1.64it/s]
 81%|████████  | 53206/65536 [9:10:10<2:08:54,  1.59it/s]
 81%|████████  | 53207/65536 [9:10:10<2:07:26,  1.61it/s]
 81%|████████  | 53208/65536 [9:10:11<2:05:38,  1.64it/s]
 81%|████████  | 53209/65536 [9:10:11<2:10:27,  1.57it/s]
 81%|████████  | 53210/65536 [9:10:12<2:07:57,  1.61it/s]
 81%|████████  | 53211/65536 [9:10:13<2:05:25,  1.64it/s]
 81%|████████  | 53212/65536 [9:10:13<2:06:53,  1.62it/s]
 81%|████████  | 53213/65536 [9:10:14<2:06:51,  1.62it/s]
 81%|████████  | 53214/65536 [9:10:14<2:03:50,  1.66it/s]
 81%|████████  | 53215/65536 [9:10:15<2:02:50,  1.67it/s]
 81%|████████  | 53216/65536 [9:10:16<2:05:03,  1.64it/s]
 81%|████████  | 53217/65536 [9:10:16<2:04:47,  1.65it/s]
 81%|████████  | 53218/65536 [9:10:17<2:09:10,  1.59it/s]
 81%|████████  | 53219/65536 [9:10:18<2:10:06,  1.58it/s]
 81%|████████  | 53220/65536 [9:10:18<2:06:36,  1.62it/s]
                                                         
{'loss': 1.5729, 'learning_rate': 2.6965225909146566e-07, 'epoch': 3285.19}
+
 81%|████████  | 53220/65536 [9:10:18<2:06:36,  1.62it/s]
 81%|████████  | 53221/65536 [9:10:19<2:06:17,  1.63it/s]
 81%|████████  | 53222/65536 [9:10:19<2:07:37,  1.61it/s]
 81%|████████  | 53223/65536 [9:10:20<2:09:57,  1.58it/s]
 81%|████████  | 53224/65536 [9:10:21<2:10:38,  1.57it/s]
 81%|████████  | 53225/65536 [9:10:21<2:08:57,  1.59it/s]
 81%|████████  | 53226/65536 [9:10:22<2:07:08,  1.61it/s]
 81%|████████  | 53227/65536 [9:10:23<2:04:42,  1.64it/s]
 81%|████████  | 53228/65536 [9:10:23<2:03:04,  1.67it/s]
 81%|████████  | 53229/65536 [9:10:24<2:05:21,  1.64it/s]
 81%|████████  | 53230/65536 [9:10:24<2:05:46,  1.63it/s]
 81%|████████  | 53231/65536 [9:10:25<2:05:11,  1.64it/s]
 81%|████████  | 53232/65536 [9:10:26<2:06:14,  1.62it/s]
 81%|████████  | 53233/65536 [9:10:26<2:04:55,  1.64it/s]
 81%|████████  | 53234/65536 [9:10:27<2:10:42,  1.57it/s]
 81%|████████  | 53235/65536 [9:10:28<2:09:34,  1.58it/s]
 81%|████████  | 53236/65536 [9:10:28<2:11:38,  1.56it/s]
 81%|████████  | 53237/65536 [9:10:29<2:07:50,  1.60it/s]
 81%|████████  | 53238/65536 [9:10:29<2:06:59,  1.61it/s]
 81%|████████  | 53239/65536 [9:10:30<2:05:17,  1.64it/s]
 81%|████████  | 53240/65536 [9:10:31<2:02:18,  1.68it/s]
                                                         
{'loss': 1.5773, 'learning_rate': 2.693767601322394e-07, 'epoch': 3286.42}
+
 81%|████████  | 53240/65536 [9:10:31<2:02:18,  1.68it/s]
 81%|████████  | 53241/65536 [9:10:31<2:03:29,  1.66it/s]
 81%|████████  | 53242/65536 [9:10:32<2:04:16,  1.65it/s]
 81%|████████  | 53243/65536 [9:10:32<2:06:35,  1.62it/s]
 81%|████████  | 53244/65536 [9:10:33<2:05:22,  1.63it/s]
 81%|████████  | 53245/65536 [9:10:34<2:05:44,  1.63it/s]
 81%|████████  | 53246/65536 [9:10:34<2:05:51,  1.63it/s]
 81%|████████  | 53247/65536 [9:10:35<2:09:14,  1.58it/s]
 81%|████████▏ | 53248/65536 [9:10:36<2:08:59,  1.59it/s]
 81%|████████▏ | 53249/65536 [9:10:36<2:06:19,  1.62it/s]
 81%|████████▏ | 53250/65536 [9:10:37<2:10:01,  1.57it/s]
 81%|████████▏ | 53251/65536 [9:10:37<2:09:55,  1.58it/s]
 81%|████████▏ | 53252/65536 [9:10:38<2:07:28,  1.61it/s]
 81%|████████▏ | 53253/65536 [9:10:39<2:08:32,  1.59it/s]
 81%|████████▏ | 53254/65536 [9:10:39<2:07:09,  1.61it/s]
 81%|████████▏ | 53255/65536 [9:10:40<2:05:39,  1.63it/s]
 81%|████████▏ | 53256/65536 [9:10:41<2:05:33,  1.63it/s]
 81%|████████▏ | 53257/65536 [9:10:41<2:06:46,  1.61it/s]
 81%|████████▏ | 53258/65536 [9:10:42<2:07:54,  1.60it/s]
 81%|████████▏ | 53259/65536 [9:10:42<2:06:15,  1.62it/s]
 81%|████████▏ | 53260/65536 [9:10:43<2:10:41,  1.57it/s]
                                                         
{'loss': 1.5507, 'learning_rate': 2.6910126117301327e-07, 'epoch': 3287.65}
+
 81%|████████▏ | 53260/65536 [9:10:43<2:10:41,  1.57it/s]
 81%|████████▏ | 53261/65536 [9:10:44<2:08:42,  1.59it/s]
 81%|████████▏ | 53262/65536 [9:10:44<2:06:56,  1.61it/s]
 81%|████████▏ | 53263/65536 [9:10:45<2:07:58,  1.60it/s]
 81%|████████▏ | 53264/65536 [9:10:46<2:06:47,  1.61it/s]
 81%|████████▏ | 53265/65536 [9:10:46<2:04:21,  1.64it/s]
 81%|████████▏ | 53266/65536 [9:10:47<2:10:48,  1.56it/s]
 81%|████████▏ | 53267/65536 [9:10:47<2:08:41,  1.59it/s]
 81%|████████▏ | 53268/65536 [9:10:48<2:09:43,  1.58it/s]
 81%|████████▏ | 53269/65536 [9:10:49<2:08:25,  1.59it/s]
 81%|████████▏ | 53270/65536 [9:10:49<2:11:07,  1.56it/s]
 81%|████████▏ | 53271/65536 [9:10:50<2:10:35,  1.57it/s]
 81%|████████▏ | 53272/65536 [9:10:51<2:07:21,  1.60it/s]
 81%|████████▏ | 53273/65536 [9:10:51<2:08:36,  1.59it/s]
 81%|████████▏ | 53274/65536 [9:10:52<2:07:52,  1.60it/s]
 81%|████████▏ | 53275/65536 [9:10:52<2:07:25,  1.60it/s]
 81%|████████▏ | 53276/65536 [9:10:53<2:07:41,  1.60it/s]
 81%|████████▏ | 53277/65536 [9:10:54<2:06:59,  1.61it/s]
 81%|████████▏ | 53278/65536 [9:10:54<2:04:41,  1.64it/s]
 81%|████████▏ | 53279/65536 [9:10:55<2:10:07,  1.57it/s]
 81%|████████▏ | 53280/65536 [9:10:56<2:08:39,  1.59it/s]
                                                         
{'loss': 1.5789, 'learning_rate': 2.6882576221378713e-07, 'epoch': 3288.89}
+
 81%|████████▏ | 53280/65536 [9:10:56<2:08:39,  1.59it/s]
 81%|████████▏ | 53281/65536 [9:10:56<2:06:45,  1.61it/s]
 81%|████████▏ | 53282/65536 [9:10:57<2:09:28,  1.58it/s]
 81%|████████▏ | 53283/65536 [9:10:57<2:08:47,  1.59it/s]
 81%|████████▏ | 53284/65536 [9:10:58<2:08:55,  1.58it/s]
 81%|████████▏ | 53285/65536 [9:10:59<2:06:42,  1.61it/s]
 81%|████████▏ | 53286/65536 [9:10:59<2:03:47,  1.65it/s]
 81%|████████▏ | 53287/65536 [9:11:00<2:06:07,  1.62it/s]
 81%|████████▏ | 53288/65536 [9:11:01<2:09:49,  1.57it/s]
 81%|████████▏ | 53289/65536 [9:11:01<2:09:30,  1.58it/s]
 81%|████████▏ | 53290/65536 [9:11:02<2:10:37,  1.56it/s]
 81%|████████▏ | 53291/65536 [9:11:02<2:07:43,  1.60it/s]
 81%|████████▏ | 53292/65536 [9:11:03<2:06:18,  1.62it/s]
 81%|████████▏ | 53293/65536 [9:11:04<2:05:32,  1.63it/s]
 81%|████████▏ | 53294/65536 [9:11:04<2:07:50,  1.60it/s]
 81%|████████▏ | 53295/65536 [9:11:05<2:07:33,  1.60it/s]
 81%|████████▏ | 53296/65536 [9:11:06<2:05:49,  1.62it/s]
 81%|████████▏ | 53297/65536 [9:11:06<2:09:04,  1.58it/s]
 81%|████████▏ | 53298/65536 [9:11:07<2:06:44,  1.61it/s]
 81%|████████▏ | 53299/65536 [9:11:07<2:10:14,  1.57it/s]
 81%|████████▏ | 53300/65536 [9:11:08<2:08:43,  1.58it/s]
                                                         
{'loss': 1.569, 'learning_rate': 2.68550263254561e-07, 'epoch': 3290.12}
+
 81%|████████▏ | 53300/65536 [9:11:08<2:08:43,  1.58it/s]
 81%|████████▏ | 53301/65536 [9:11:09<2:09:59,  1.57it/s]
 81%|████████▏ | 53302/65536 [9:11:09<2:07:31,  1.60it/s]
 81%|████████▏ | 53303/65536 [9:11:10<2:05:23,  1.63it/s]
 81%|████████▏ | 53304/65536 [9:11:11<2:04:46,  1.63it/s]
 81%|████████▏ | 53305/65536 [9:11:11<2:03:16,  1.65it/s]
 81%|████████▏ | 53306/65536 [9:11:12<2:04:16,  1.64it/s]
 81%|████████▏ | 53307/65536 [9:11:12<2:06:02,  1.62it/s]
 81%|████████▏ | 53308/65536 [9:11:13<2:04:49,  1.63it/s]
 81%|████████▏ | 53309/65536 [9:11:14<2:02:20,  1.67it/s]
 81%|████████▏ | 53310/65536 [9:11:14<2:03:49,  1.65it/s]
 81%|████████▏ | 53311/65536 [9:11:15<2:08:07,  1.59it/s]
 81%|████████▏ | 53312/65536 [9:11:16<2:09:39,  1.57it/s]
 81%|████████▏ | 53313/65536 [9:11:16<2:06:53,  1.61it/s]
 81%|████████▏ | 53314/65536 [9:11:17<2:08:23,  1.59it/s]
 81%|████████▏ | 53315/65536 [9:11:17<2:11:44,  1.55it/s]
 81%|████████▏ | 53316/65536 [9:11:18<2:12:30,  1.54it/s]
 81%|████████▏ | 53317/65536 [9:11:19<2:11:17,  1.55it/s]
 81%|████████▏ | 53318/65536 [9:11:19<2:10:10,  1.56it/s]
 81%|████████▏ | 53319/65536 [9:11:20<2:08:01,  1.59it/s]
 81%|████████▏ | 53320/65536 [9:11:21<2:11:03,  1.55it/s]
                                                         
{'loss': 1.5787, 'learning_rate': 2.682747642953349e-07, 'epoch': 3291.36}
+
 81%|████████▏ | 53320/65536 [9:11:21<2:11:03,  1.55it/s]
 81%|████████▏ | 53321/65536 [9:11:21<2:08:19,  1.59it/s]
 81%|████████▏ | 53322/65536 [9:11:22<2:05:14,  1.63it/s]
 81%|████████▏ | 53323/65536 [9:11:22<2:04:05,  1.64it/s]
 81%|████████▏ | 53324/65536 [9:11:23<2:06:09,  1.61it/s]
 81%|████████▏ | 53325/65536 [9:11:24<2:06:32,  1.61it/s]
 81%|████████▏ | 53326/65536 [9:11:24<2:03:48,  1.64it/s]
 81%|████████▏ | 53327/65536 [9:11:25<2:05:03,  1.63it/s]
 81%|████████▏ | 53328/65536 [9:11:26<2:04:44,  1.63it/s]
 81%|████████▏ | 53329/65536 [9:11:26<2:05:17,  1.62it/s]
 81%|████████▏ | 53330/65536 [9:11:27<2:06:05,  1.61it/s]
 81%|████████▏ | 53331/65536 [9:11:27<2:09:23,  1.57it/s]
 81%|████████▏ | 53332/65536 [9:11:28<2:10:54,  1.55it/s]
 81%|████████▏ | 53333/65536 [9:11:29<2:05:39,  1.62it/s]
 81%|████████▏ | 53334/65536 [9:11:29<2:05:15,  1.62it/s]
 81%|████████▏ | 53335/65536 [9:11:30<2:08:37,  1.58it/s]
 81%|████████▏ | 53336/65536 [9:11:31<2:06:08,  1.61it/s]
 81%|████████▏ | 53337/65536 [9:11:31<2:06:36,  1.61it/s]
 81%|████████▏ | 53338/65536 [9:11:32<2:10:56,  1.55it/s]
 81%|████████▏ | 53339/65536 [9:11:32<2:07:01,  1.60it/s]
 81%|████████▏ | 53340/65536 [9:11:33<2:06:52,  1.60it/s]
                                                         
{'loss': 1.5788, 'learning_rate': 2.6799926533610876e-07, 'epoch': 3292.59}
+
 81%|████████▏ | 53340/65536 [9:11:33<2:06:52,  1.60it/s]
 81%|████████▏ | 53341/65536 [9:11:34<2:04:19,  1.63it/s]
 81%|████████▏ | 53342/65536 [9:11:34<2:08:22,  1.58it/s]
 81%|████████▏ | 53343/65536 [9:11:35<2:08:23,  1.58it/s]
 81%|████████▏ | 53344/65536 [9:11:36<2:06:52,  1.60it/s]
 81%|████████▏ | 53345/65536 [9:11:36<2:06:39,  1.60it/s]
 81%|████████▏ | 53346/65536 [9:11:37<2:04:57,  1.63it/s]
 81%|████████▏ | 53347/65536 [9:11:37<2:09:04,  1.57it/s]
 81%|████████▏ | 53348/65536 [9:11:38<2:06:03,  1.61it/s]
 81%|████████▏ | 53349/65536 [9:11:39<2:04:50,  1.63it/s]
 81%|████████▏ | 53350/65536 [9:11:39<2:06:41,  1.60it/s]
 81%|████████▏ | 53351/65536 [9:11:40<2:06:41,  1.60it/s]
 81%|████████▏ | 53352/65536 [9:11:41<2:06:52,  1.60it/s]
 81%|████████▏ | 53353/65536 [9:11:41<2:09:03,  1.57it/s]
 81%|████████▏ | 53354/65536 [9:11:42<2:14:33,  1.51it/s]
 81%|████████▏ | 53355/65536 [9:11:43<2:10:03,  1.56it/s]
 81%|████████▏ | 53356/65536 [9:11:43<2:07:43,  1.59it/s]
 81%|████████▏ | 53357/65536 [9:11:44<2:06:35,  1.60it/s]
 81%|████████▏ | 53358/65536 [9:11:44<2:07:07,  1.60it/s]
 81%|████████▏ | 53359/65536 [9:11:45<2:05:41,  1.61it/s]
 81%|████████▏ | 53360/65536 [9:11:46<2:08:37,  1.58it/s]
                                                         
{'loss': 1.5194, 'learning_rate': 2.677237663768826e-07, 'epoch': 3293.83}
+
 81%|████████▏ | 53360/65536 [9:11:46<2:08:37,  1.58it/s]
 81%|████████▏ | 53361/65536 [9:11:46<2:06:08,  1.61it/s]
 81%|████████▏ | 53362/65536 [9:11:47<2:06:48,  1.60it/s]
 81%|████████▏ | 53363/65536 [9:11:48<2:11:52,  1.54it/s]
 81%|████████▏ | 53364/65536 [9:11:48<2:13:18,  1.52it/s]
 81%|████████▏ | 53365/65536 [9:11:49<2:14:13,  1.51it/s]
 81%|████████▏ | 53366/65536 [9:11:50<2:19:00,  1.46it/s]
 81%|████████▏ | 53367/65536 [9:11:50<2:16:43,  1.48it/s]
 81%|████████▏ | 53368/65536 [9:11:51<2:09:49,  1.56it/s]
 81%|████████▏ | 53369/65536 [9:11:52<2:11:26,  1.54it/s]
 81%|████████▏ | 53370/65536 [9:11:52<2:09:32,  1.57it/s]
 81%|████████▏ | 53371/65536 [9:11:53<2:07:11,  1.59it/s]
 81%|████████▏ | 53372/65536 [9:11:53<2:06:16,  1.61it/s]
 81%|████████▏ | 53373/65536 [9:11:54<2:07:39,  1.59it/s]
 81%|████████▏ | 53374/65536 [9:11:55<2:06:23,  1.60it/s]
 81%|████████▏ | 53375/65536 [9:11:55<2:04:40,  1.63it/s]
 81%|████████▏ | 53376/65536 [9:11:56<2:05:13,  1.62it/s]
 81%|████████▏ | 53377/65536 [9:11:56<2:06:34,  1.60it/s]
 81%|████████▏ | 53378/65536 [9:11:57<2:05:25,  1.62it/s]
 81%|████████▏ | 53379/65536 [9:11:58<2:06:39,  1.60it/s]
 81%|████████▏ | 53380/65536 [9:11:58<2:08:28,  1.58it/s]
                                                         
{'loss': 1.5846, 'learning_rate': 2.6744826741765637e-07, 'epoch': 3295.06}
+
 81%|████████▏ | 53380/65536 [9:11:58<2:08:28,  1.58it/s]
 81%|████████▏ | 53381/65536 [9:11:59<2:04:29,  1.63it/s]
 81%|████████▏ | 53382/65536 [9:12:00<2:03:15,  1.64it/s]
 81%|████████▏ | 53383/65536 [9:12:00<2:02:56,  1.65it/s]
 81%|████████▏ | 53384/65536 [9:12:01<2:02:34,  1.65it/s]
 81%|████████▏ | 53385/65536 [9:12:01<2:01:37,  1.66it/s]
 81%|████████▏ | 53386/65536 [9:12:02<2:02:47,  1.65it/s]
 81%|████████▏ | 53387/65536 [9:12:03<2:02:17,  1.66it/s]
 81%|████████▏ | 53388/65536 [9:12:03<2:04:45,  1.62it/s]
 81%|████████▏ | 53389/65536 [9:12:04<2:04:28,  1.63it/s]
 81%|████████▏ | 53390/65536 [9:12:04<2:07:41,  1.59it/s]
 81%|████████▏ | 53391/65536 [9:12:05<2:08:55,  1.57it/s]
 81%|████████▏ | 53392/65536 [9:12:06<2:07:52,  1.58it/s]
 81%|████████▏ | 53393/65536 [9:12:06<2:11:21,  1.54it/s]
 81%|████████▏ | 53394/65536 [9:12:07<2:12:00,  1.53it/s]
 81%|████████▏ | 53395/65536 [9:12:08<2:07:49,  1.58it/s]
 81%|████████▏ | 53396/65536 [9:12:08<2:11:21,  1.54it/s]
 81%|████████▏ | 53397/65536 [9:12:09<2:12:11,  1.53it/s]
 81%|████████▏ | 53398/65536 [9:12:10<2:13:14,  1.52it/s]
 81%|████████▏ | 53399/65536 [9:12:10<2:08:57,  1.57it/s]
 81%|████████▏ | 53400/65536 [9:12:11<2:06:56,  1.59it/s]
                                                         
{'loss': 1.5296, 'learning_rate': 2.6717276845843023e-07, 'epoch': 3296.3}
+
 81%|████████▏ | 53400/65536 [9:12:11<2:06:56,  1.59it/s]
 81%|████████▏ | 53401/65536 [9:12:12<2:06:58,  1.59it/s]
 81%|████████▏ | 53402/65536 [9:12:12<2:05:41,  1.61it/s]
 81%|████████▏ | 53403/65536 [9:12:13<2:05:11,  1.62it/s]
 81%|████████▏ | 53404/65536 [9:12:13<2:05:39,  1.61it/s]
 81%|████████▏ | 53405/65536 [9:12:14<2:04:18,  1.63it/s]
 81%|████████▏ | 53406/65536 [9:12:15<2:07:10,  1.59it/s]
 81%|████████▏ | 53407/65536 [9:12:15<2:05:31,  1.61it/s]
 81%|████████▏ | 53408/65536 [9:12:16<2:03:34,  1.64it/s]
 81%|████████▏ | 53409/65536 [9:12:16<2:00:30,  1.68it/s]
 81%|████████▏ | 53410/65536 [9:12:17<1:59:18,  1.69it/s]
 81%|████████▏ | 53411/65536 [9:12:18<2:00:19,  1.68it/s]
 82%|████████▏ | 53412/65536 [9:12:18<2:08:33,  1.57it/s]
 82%|████████▏ | 53413/65536 [9:12:19<2:05:11,  1.61it/s]
 82%|████████▏ | 53414/65536 [9:12:19<2:03:24,  1.64it/s]
 82%|████████▏ | 53415/65536 [9:12:20<2:03:56,  1.63it/s]
 82%|████████▏ | 53416/65536 [9:12:21<2:04:59,  1.62it/s]
 82%|████████▏ | 53417/65536 [9:12:21<2:07:25,  1.59it/s]
 82%|████████▏ | 53418/65536 [9:12:22<2:05:46,  1.61it/s]
 82%|████████▏ | 53419/65536 [9:12:23<2:05:30,  1.61it/s]
 82%|████████▏ | 53420/65536 [9:12:23<2:04:25,  1.62it/s]
                                                         
{'loss': 1.5572, 'learning_rate': 2.668972694992041e-07, 'epoch': 3297.53}
+
 82%|████████▏ | 53420/65536 [9:12:23<2:04:25,  1.62it/s]
 82%|████████▏ | 53421/65536 [9:12:24<2:08:04,  1.58it/s]
 82%|████████▏ | 53422/65536 [9:12:24<2:04:55,  1.62it/s]
 82%|████████▏ | 53423/65536 [9:12:25<2:08:08,  1.58it/s]
 82%|████████▏ | 53424/65536 [9:12:26<2:06:26,  1.60it/s]
 82%|████████▏ | 53425/65536 [9:12:26<2:03:33,  1.63it/s]
 82%|████████▏ | 53426/65536 [9:12:27<2:02:23,  1.65it/s]
 82%|████████▏ | 53427/65536 [9:12:28<2:03:02,  1.64it/s]
 82%|████████▏ | 53428/65536 [9:12:28<2:06:31,  1.59it/s]
 82%|████████▏ | 53429/65536 [9:12:29<2:04:03,  1.63it/s]
 82%|████████▏ | 53430/65536 [9:12:29<2:03:32,  1.63it/s]
 82%|████████▏ | 53431/65536 [9:12:30<2:03:51,  1.63it/s]
 82%|████████▏ | 53432/65536 [9:12:31<2:03:28,  1.63it/s]
 82%|████████▏ | 53433/65536 [9:12:31<2:02:05,  1.65it/s]
 82%|████████▏ | 53434/65536 [9:12:32<2:01:06,  1.67it/s]
 82%|████████▏ | 53435/65536 [9:12:32<2:01:43,  1.66it/s]
 82%|████████▏ | 53436/65536 [9:12:33<2:03:52,  1.63it/s]
 82%|████████▏ | 53437/65536 [9:12:34<2:04:21,  1.62it/s]
 82%|████████▏ | 53438/65536 [9:12:34<2:04:25,  1.62it/s]
 82%|████████▏ | 53439/65536 [9:12:35<2:04:01,  1.63it/s]
 82%|████████▏ | 53440/65536 [9:12:36<2:07:01,  1.59it/s]
                                                         
{'loss': 1.5682, 'learning_rate': 2.66621770539978e-07, 'epoch': 3298.77}
+
 82%|████████▏ | 53440/65536 [9:12:36<2:07:01,  1.59it/s]
 82%|████████▏ | 53441/65536 [9:12:36<2:07:54,  1.58it/s]
 82%|████████▏ | 53442/65536 [9:12:37<2:05:25,  1.61it/s]
 82%|████████▏ | 53443/65536 [9:12:37<2:06:13,  1.60it/s]
 82%|████████▏ | 53444/65536 [9:12:38<2:09:02,  1.56it/s]
 82%|████████▏ | 53445/65536 [9:12:39<2:07:25,  1.58it/s]
 82%|████████▏ | 53446/65536 [9:12:39<2:07:02,  1.59it/s]
 82%|████████▏ | 53447/65536 [9:12:40<2:11:13,  1.54it/s]
 82%|████████▏ | 53448/65536 [9:12:41<2:07:17,  1.58it/s]
 82%|████████▏ | 53449/65536 [9:12:41<2:05:33,  1.60it/s]
 82%|████████▏ | 53450/65536 [9:12:42<2:03:40,  1.63it/s]
 82%|████████▏ | 53451/65536 [9:12:43<2:07:41,  1.58it/s]
 82%|████████▏ | 53452/65536 [9:12:43<2:05:39,  1.60it/s]
 82%|████████▏ | 53453/65536 [9:12:44<2:06:14,  1.60it/s]
 82%|████████▏ | 53454/65536 [9:12:44<2:02:18,  1.65it/s]
 82%|████████▏ | 53455/65536 [9:12:45<2:01:34,  1.66it/s]
 82%|████████▏ | 53456/65536 [9:12:46<2:02:10,  1.65it/s]
 82%|████████▏ | 53457/65536 [9:12:46<2:01:48,  1.65it/s]
 82%|████████▏ | 53458/65536 [9:12:47<2:00:57,  1.66it/s]
 82%|████████▏ | 53459/65536 [9:12:47<2:01:36,  1.66it/s]
 82%|████████▏ | 53460/65536 [9:12:48<2:01:08,  1.66it/s]
                                                         
{'loss': 1.5689, 'learning_rate': 2.663462715807518e-07, 'epoch': 3300.0}
+
 82%|████████▏ | 53460/65536 [9:12:48<2:01:08,  1.66it/s]
 82%|████████▏ | 53461/65536 [9:12:49<2:05:32,  1.60it/s]
 82%|████████▏ | 53462/65536 [9:12:49<2:07:20,  1.58it/s]
 82%|████████▏ | 53463/65536 [9:12:50<2:03:34,  1.63it/s]
 82%|████████▏ | 53464/65536 [9:12:50<2:05:07,  1.61it/s]
 82%|████████▏ | 53465/65536 [9:12:51<2:05:15,  1.61it/s]
 82%|████████▏ | 53466/65536 [9:12:52<2:04:50,  1.61it/s]
 82%|████████▏ | 53467/65536 [9:12:52<2:06:29,  1.59it/s]
 82%|████████▏ | 53468/65536 [9:12:53<2:04:48,  1.61it/s]
 82%|████████▏ | 53469/65536 [9:12:54<2:03:25,  1.63it/s]
 82%|████████▏ | 53470/65536 [9:12:54<2:04:51,  1.61it/s]
 82%|████████▏ | 53471/65536 [9:12:55<2:04:08,  1.62it/s]
 82%|████████▏ | 53472/65536 [9:12:55<2:01:05,  1.66it/s]
 82%|████████▏ | 53473/65536 [9:12:56<2:02:54,  1.64it/s]
 82%|████████▏ | 53474/65536 [9:12:57<2:04:26,  1.62it/s]
 82%|████████▏ | 53475/65536 [9:12:57<2:06:07,  1.59it/s]
 82%|████████▏ | 53476/65536 [9:12:58<2:04:44,  1.61it/s]
 82%|████████▏ | 53477/65536 [9:12:59<2:10:35,  1.54it/s]
 82%|████████▏ | 53478/65536 [9:12:59<2:10:48,  1.54it/s]
 82%|████████▏ | 53479/65536 [9:13:00<2:09:29,  1.55it/s]
 82%|████████▏ | 53480/65536 [9:13:01<2:07:36,  1.57it/s]
                                                         
{'loss': 1.5466, 'learning_rate': 2.6607077262152567e-07, 'epoch': 3301.23}
+
 82%|████████▏ | 53480/65536 [9:13:01<2:07:36,  1.57it/s]
 82%|████████▏ | 53481/65536 [9:13:01<2:08:25,  1.56it/s]
 82%|████████▏ | 53482/65536 [9:13:02<2:04:23,  1.62it/s]
 82%|████████▏ | 53483/65536 [9:13:02<2:03:00,  1.63it/s]
 82%|████████▏ | 53484/65536 [9:13:03<2:05:30,  1.60it/s]
 82%|████████▏ | 53485/65536 [9:13:04<2:06:18,  1.59it/s]
 82%|████████▏ | 53486/65536 [9:13:04<2:02:44,  1.64it/s]
 82%|████████▏ | 53487/65536 [9:13:05<2:03:33,  1.63it/s]
 82%|████████▏ | 53488/65536 [9:13:05<2:04:24,  1.61it/s]
 82%|████████▏ | 53489/65536 [9:13:06<2:08:27,  1.56it/s]
 82%|████████▏ | 53490/65536 [9:13:07<2:07:06,  1.58it/s]
 82%|████████▏ | 53491/65536 [9:13:07<2:05:34,  1.60it/s]
 82%|████████▏ | 53492/65536 [9:13:08<2:02:15,  1.64it/s]
 82%|████████▏ | 53493/65536 [9:13:09<2:07:35,  1.57it/s]
 82%|████████▏ | 53494/65536 [9:13:09<2:05:05,  1.60it/s]
 82%|████████▏ | 53495/65536 [9:13:10<2:03:49,  1.62it/s]
 82%|████████▏ | 53496/65536 [9:13:10<2:07:51,  1.57it/s]
 82%|████████▏ | 53497/65536 [9:13:11<2:07:01,  1.58it/s]
 82%|████████▏ | 53498/65536 [9:13:12<2:03:19,  1.63it/s]
 82%|████████▏ | 53499/65536 [9:13:12<2:02:04,  1.64it/s]
 82%|████████▏ | 53500/65536 [9:13:13<2:00:44,  1.66it/s]
                                                         
{'loss': 1.5556, 'learning_rate': 2.657952736622995e-07, 'epoch': 3302.47}
+
 82%|████████▏ | 53500/65536 [9:13:13<2:00:44,  1.66it/s]
 82%|████████▏ | 53501/65536 [9:13:14<2:03:00,  1.63it/s]
 82%|████████▏ | 53502/65536 [9:13:14<2:01:50,  1.65it/s]
 82%|████████▏ | 53503/65536 [9:13:15<2:02:12,  1.64it/s]
 82%|████████▏ | 53504/65536 [9:13:15<1:59:18,  1.68it/s]
 82%|████████▏ | 53505/65536 [9:13:16<2:00:21,  1.67it/s]
 82%|████████▏ | 53506/65536 [9:13:16<2:00:11,  1.67it/s]
 82%|████████▏ | 53507/65536 [9:13:17<2:03:42,  1.62it/s]
 82%|████████▏ | 53508/65536 [9:13:18<2:04:43,  1.61it/s]
 82%|████████▏ | 53509/65536 [9:13:19<2:10:55,  1.53it/s]
 82%|████████▏ | 53510/65536 [9:13:19<2:07:48,  1.57it/s]
 82%|████████▏ | 53511/65536 [9:13:20<2:08:19,  1.56it/s]
 82%|████████▏ | 53512/65536 [9:13:20<2:04:29,  1.61it/s]
 82%|████████▏ | 53513/65536 [9:13:21<2:06:22,  1.59it/s]
 82%|████████▏ | 53514/65536 [9:13:22<2:04:25,  1.61it/s]
 82%|████████▏ | 53515/65536 [9:13:22<2:04:50,  1.60it/s]
 82%|████████▏ | 53516/65536 [9:13:23<2:05:06,  1.60it/s]
 82%|████████▏ | 53517/65536 [9:13:23<2:03:05,  1.63it/s]
 82%|████████▏ | 53518/65536 [9:13:24<2:04:45,  1.61it/s]
 82%|████████▏ | 53519/65536 [9:13:25<2:05:03,  1.60it/s]
 82%|████████▏ | 53520/65536 [9:13:25<2:06:05,  1.59it/s]
                                                         
{'loss': 1.5757, 'learning_rate': 2.6551977470307333e-07, 'epoch': 3303.7}
+
 82%|████████▏ | 53520/65536 [9:13:25<2:06:05,  1.59it/s]
 82%|████████▏ | 53521/65536 [9:13:26<2:05:32,  1.60it/s]
 82%|████████▏ | 53522/65536 [9:13:27<2:06:35,  1.58it/s]
 82%|████████▏ | 53523/65536 [9:13:27<2:06:37,  1.58it/s]
 82%|████████▏ | 53524/65536 [9:13:28<2:04:55,  1.60it/s]
 82%|████████▏ | 53525/65536 [9:13:28<2:05:53,  1.59it/s]
 82%|████████▏ | 53526/65536 [9:13:29<2:04:32,  1.61it/s]
 82%|████████▏ | 53527/65536 [9:13:30<2:03:12,  1.62it/s]
 82%|████████▏ | 53528/65536 [9:13:30<2:03:52,  1.62it/s]
 82%|████████▏ | 53529/65536 [9:13:31<2:04:36,  1.61it/s]
 82%|████████▏ | 53530/65536 [9:13:32<2:04:20,  1.61it/s]
 82%|████████▏ | 53531/65536 [9:13:32<2:04:16,  1.61it/s]
 82%|████████▏ | 53532/65536 [9:13:33<2:03:58,  1.61it/s]
 82%|████████▏ | 53533/65536 [9:13:33<2:03:51,  1.62it/s]
 82%|████████▏ | 53534/65536 [9:13:34<2:02:47,  1.63it/s]
 82%|████████▏ | 53535/65536 [9:13:35<2:04:57,  1.60it/s]
 82%|████████▏ | 53536/65536 [9:13:35<2:06:56,  1.58it/s]
 82%|████████▏ | 53537/65536 [9:13:36<2:07:29,  1.57it/s]
 82%|████████▏ | 53538/65536 [9:13:37<2:08:41,  1.55it/s]
 82%|████████▏ | 53539/65536 [9:13:37<2:05:45,  1.59it/s]
 82%|████████▏ | 53540/65536 [9:13:38<2:02:34,  1.63it/s]
                                                         
{'loss': 1.538, 'learning_rate': 2.652442757438472e-07, 'epoch': 3304.94}
+
 82%|████████▏ | 53540/65536 [9:13:38<2:02:34,  1.63it/s]
 82%|████████▏ | 53541/65536 [9:13:38<1:59:53,  1.67it/s]
 82%|████████▏ | 53542/65536 [9:13:39<2:04:24,  1.61it/s]
 82%|████████▏ | 53543/65536 [9:13:40<2:04:50,  1.60it/s]
 82%|████████▏ | 53544/65536 [9:13:40<2:03:42,  1.62it/s]
 82%|████████▏ | 53545/65536 [9:13:41<2:03:48,  1.61it/s]
 82%|████████▏ | 53546/65536 [9:13:42<2:04:15,  1.61it/s]
 82%|████████▏ | 53547/65536 [9:13:42<2:03:37,  1.62it/s]
 82%|████████▏ | 53548/65536 [9:13:43<2:02:55,  1.63it/s]
 82%|████████▏ | 53549/65536 [9:13:43<2:02:15,  1.63it/s]
 82%|████████▏ | 53550/65536 [9:13:44<2:04:36,  1.60it/s]
 82%|████████▏ | 53551/65536 [9:13:45<2:01:05,  1.65it/s]
 82%|████████▏ | 53552/65536 [9:13:45<2:01:21,  1.65it/s]
 82%|████████▏ | 53553/65536 [9:13:46<2:02:39,  1.63it/s]
 82%|████████▏ | 53554/65536 [9:13:46<2:02:34,  1.63it/s]
 82%|████████▏ | 53555/65536 [9:13:47<2:04:20,  1.61it/s]
 82%|████████▏ | 53556/65536 [9:13:48<2:05:12,  1.59it/s]
 82%|████████▏ | 53557/65536 [9:13:48<2:06:21,  1.58it/s]
 82%|████████▏ | 53558/65536 [9:13:49<2:09:43,  1.54it/s]
 82%|████████▏ | 53559/65536 [9:13:50<2:07:58,  1.56it/s]
 82%|████████▏ | 53560/65536 [9:13:50<2:05:46,  1.59it/s]
                                                         
{'loss': 1.521, 'learning_rate': 2.6496877678462105e-07, 'epoch': 3306.17}
+
 82%|████████▏ | 53560/65536 [9:13:50<2:05:46,  1.59it/s]
 82%|████████▏ | 53561/65536 [9:13:51<2:06:32,  1.58it/s]
 82%|████████▏ | 53562/65536 [9:13:52<2:08:48,  1.55it/s]
 82%|████████▏ | 53563/65536 [9:13:52<2:03:08,  1.62it/s]
 82%|████████▏ | 53564/65536 [9:13:53<2:03:41,  1.61it/s]
 82%|████████▏ | 53565/65536 [9:13:53<2:00:46,  1.65it/s]
 82%|████████▏ | 53566/65536 [9:13:54<2:03:01,  1.62it/s]
 82%|████████▏ | 53567/65536 [9:13:55<2:00:56,  1.65it/s]
 82%|████████▏ | 53568/65536 [9:13:55<1:59:13,  1.67it/s]
 82%|████████▏ | 53569/65536 [9:13:56<1:58:09,  1.69it/s]
 82%|████████▏ | 53570/65536 [9:13:56<2:01:17,  1.64it/s]
 82%|████████▏ | 53571/65536 [9:13:57<2:04:12,  1.61it/s]
 82%|████████▏ | 53572/65536 [9:13:58<2:04:05,  1.61it/s]
 82%|████████▏ | 53573/65536 [9:13:58<2:03:38,  1.61it/s]
 82%|████████▏ | 53574/65536 [9:13:59<2:05:19,  1.59it/s]
 82%|████████▏ | 53575/65536 [9:14:00<2:05:53,  1.58it/s]
 82%|████████▏ | 53576/65536 [9:14:00<2:06:32,  1.58it/s]
 82%|████████▏ | 53577/65536 [9:14:01<2:04:12,  1.60it/s]
 82%|████████▏ | 53578/65536 [9:14:01<2:01:53,  1.64it/s]
 82%|████████▏ | 53579/65536 [9:14:02<2:02:56,  1.62it/s]
 82%|████████▏ | 53580/65536 [9:14:03<2:01:14,  1.64it/s]
                                                         
{'loss': 1.5797, 'learning_rate': 2.646932778253949e-07, 'epoch': 3307.41}
+
 82%|████████▏ | 53580/65536 [9:14:03<2:01:14,  1.64it/s]
 82%|████████▏ | 53581/65536 [9:14:03<1:58:49,  1.68it/s]
 82%|████████▏ | 53582/65536 [9:14:04<1:59:03,  1.67it/s]
 82%|████████▏ | 53583/65536 [9:14:04<2:03:03,  1.62it/s]
 82%|████████▏ | 53584/65536 [9:14:05<2:04:45,  1.60it/s]
 82%|████████▏ | 53585/65536 [9:14:06<2:03:47,  1.61it/s]
 82%|████████▏ | 53586/65536 [9:14:06<2:04:00,  1.61it/s]
 82%|████████▏ | 53587/65536 [9:14:07<2:02:14,  1.63it/s]
 82%|████████▏ | 53588/65536 [9:14:07<2:00:39,  1.65it/s]
 82%|████████▏ | 53589/65536 [9:14:08<2:05:29,  1.59it/s]
 82%|████████▏ | 53590/65536 [9:14:09<2:07:58,  1.56it/s]
 82%|████████▏ | 53591/65536 [9:14:10<2:09:26,  1.54it/s]
 82%|████████▏ | 53592/65536 [9:14:10<2:08:29,  1.55it/s]
 82%|████████▏ | 53593/65536 [9:14:11<2:03:38,  1.61it/s]
 82%|████████▏ | 53594/65536 [9:14:11<2:01:32,  1.64it/s]
 82%|████████▏ | 53595/65536 [9:14:12<2:04:06,  1.60it/s]
 82%|████████▏ | 53596/65536 [9:14:13<2:03:01,  1.62it/s]
 82%|████████▏ | 53597/65536 [9:14:13<2:04:05,  1.60it/s]
 82%|████████▏ | 53598/65536 [9:14:14<2:00:34,  1.65it/s]
 82%|████████▏ | 53599/65536 [9:14:14<2:02:15,  1.63it/s]
 82%|████████▏ | 53600/65536 [9:14:15<2:01:40,  1.64it/s]
                                                         
{'loss': 1.5624, 'learning_rate': 2.6441777886616877e-07, 'epoch': 3308.64}
+
 82%|████████▏ | 53600/65536 [9:14:15<2:01:40,  1.64it/s]
 82%|████████▏ | 53601/65536 [9:14:16<1:58:17,  1.68it/s]
 82%|████████▏ | 53602/65536 [9:14:16<2:02:29,  1.62it/s]
 82%|████████▏ | 53603/65536 [9:14:17<1:59:57,  1.66it/s]
 82%|████████▏ | 53604/65536 [9:14:17<2:00:56,  1.64it/s]
 82%|████████▏ | 53605/65536 [9:14:18<2:02:52,  1.62it/s]
 82%|████████▏ | 53606/65536 [9:14:19<2:06:06,  1.58it/s]
 82%|████████▏ | 53607/65536 [9:14:19<2:03:41,  1.61it/s]
 82%|████████▏ | 53608/65536 [9:14:20<2:02:48,  1.62it/s]
 82%|████████▏ | 53609/65536 [9:14:21<2:02:08,  1.63it/s]
 82%|████████▏ | 53610/65536 [9:14:21<2:00:56,  1.64it/s]
 82%|████████▏ | 53611/65536 [9:14:22<2:01:09,  1.64it/s]
 82%|████████▏ | 53612/65536 [9:14:22<2:03:04,  1.61it/s]
 82%|████████▏ | 53613/65536 [9:14:23<2:02:13,  1.63it/s]
 82%|████████▏ | 53614/65536 [9:14:24<2:00:47,  1.64it/s]
 82%|████████▏ | 53615/65536 [9:14:24<1:58:17,  1.68it/s]
 82%|████████▏ | 53616/65536 [9:14:25<1:59:14,  1.67it/s]
 82%|████████▏ | 53617/65536 [9:14:25<1:58:57,  1.67it/s]
 82%|████████▏ | 53618/65536 [9:14:26<2:03:35,  1.61it/s]
 82%|████████▏ | 53619/65536 [9:14:27<2:02:53,  1.62it/s]
 82%|████████▏ | 53620/65536 [9:14:27<2:05:41,  1.58it/s]
                                                         
{'loss': 1.5675, 'learning_rate': 2.641422799069426e-07, 'epoch': 3309.88}
+
 82%|████████▏ | 53620/65536 [9:14:27<2:05:41,  1.58it/s]
 82%|████████▏ | 53621/65536 [9:14:28<2:08:00,  1.55it/s]
 82%|████████▏ | 53622/65536 [9:14:29<2:05:43,  1.58it/s]
 82%|████████▏ | 53623/65536 [9:14:29<2:09:35,  1.53it/s]
 82%|████████▏ | 53624/65536 [9:14:30<2:04:51,  1.59it/s]
 82%|████████▏ | 53625/65536 [9:14:30<2:05:02,  1.59it/s]
 82%|████████▏ | 53626/65536 [9:14:31<2:06:23,  1.57it/s]
 82%|████████▏ | 53627/65536 [9:14:32<2:05:49,  1.58it/s]
 82%|████████▏ | 53628/65536 [9:14:32<2:04:31,  1.59it/s]
 82%|████████▏ | 53629/65536 [9:14:33<2:03:34,  1.61it/s]
 82%|████████▏ | 53630/65536 [9:14:34<2:01:55,  1.63it/s]
 82%|████████▏ | 53631/65536 [9:14:34<2:03:09,  1.61it/s]
 82%|████████▏ | 53632/65536 [9:14:35<2:04:08,  1.60it/s]
 82%|████████▏ | 53633/65536 [9:14:35<2:04:41,  1.59it/s]
 82%|████████▏ | 53634/65536 [9:14:36<2:03:49,  1.60it/s]
 82%|████████▏ | 53635/65536 [9:14:37<2:01:47,  1.63it/s]
 82%|████████▏ | 53636/65536 [9:14:37<2:00:17,  1.65it/s]
 82%|████████▏ | 53637/65536 [9:14:38<2:00:05,  1.65it/s]
 82%|████████▏ | 53638/65536 [9:14:38<1:59:09,  1.66it/s]
 82%|████████▏ | 53639/65536 [9:14:39<2:04:46,  1.59it/s]
 82%|████████▏ | 53640/65536 [9:14:40<2:01:45,  1.63it/s]
                                                         
{'loss': 1.5882, 'learning_rate': 2.638667809477164e-07, 'epoch': 3311.11}
+
 82%|████████▏ | 53640/65536 [9:14:40<2:01:45,  1.63it/s]
 82%|████████▏ | 53641/65536 [9:14:40<2:01:27,  1.63it/s]
 82%|████████▏ | 53642/65536 [9:14:41<2:03:12,  1.61it/s]
 82%|████████▏ | 53643/65536 [9:14:42<2:02:31,  1.62it/s]
 82%|████████▏ | 53644/65536 [9:14:42<2:01:39,  1.63it/s]
 82%|████████▏ | 53645/65536 [9:14:43<2:00:37,  1.64it/s]
 82%|████████▏ | 53646/65536 [9:14:43<2:02:16,  1.62it/s]
 82%|████████▏ | 53647/65536 [9:14:44<1:59:42,  1.66it/s]
 82%|████████▏ | 53648/65536 [9:14:45<2:00:48,  1.64it/s]
 82%|████████▏ | 53649/65536 [9:14:45<1:59:41,  1.66it/s]
 82%|████████▏ | 53650/65536 [9:14:46<2:01:09,  1.64it/s]
 82%|████████▏ | 53651/65536 [9:14:46<2:00:09,  1.65it/s]
 82%|████████▏ | 53652/65536 [9:14:47<1:59:43,  1.65it/s]
 82%|████████▏ | 53653/65536 [9:14:48<2:03:11,  1.61it/s]
 82%|████████▏ | 53654/65536 [9:14:48<2:02:51,  1.61it/s]
 82%|████████▏ | 53655/65536 [9:14:49<2:07:27,  1.55it/s]
 82%|████████▏ | 53656/65536 [9:14:50<2:04:27,  1.59it/s]
 82%|████████▏ | 53657/65536 [9:14:50<2:02:52,  1.61it/s]
 82%|████████▏ | 53658/65536 [9:14:51<2:02:24,  1.62it/s]
 82%|████████▏ | 53659/65536 [9:14:51<2:00:13,  1.65it/s]
 82%|████████▏ | 53660/65536 [9:14:52<2:02:55,  1.61it/s]
                                                         
{'loss': 1.5721, 'learning_rate': 2.6359128198849024e-07, 'epoch': 3312.35}
+
 82%|████████▏ | 53660/65536 [9:14:52<2:02:55,  1.61it/s]
 82%|████████▏ | 53661/65536 [9:14:53<2:00:12,  1.65it/s]
 82%|████████▏ | 53662/65536 [9:14:53<1:59:35,  1.65it/s]
 82%|████████▏ | 53663/65536 [9:14:54<2:04:28,  1.59it/s]
 82%|████████▏ | 53664/65536 [9:14:55<2:02:53,  1.61it/s]
 82%|████████▏ | 53665/65536 [9:14:55<2:02:37,  1.61it/s]
 82%|████████▏ | 53666/65536 [9:14:56<2:04:45,  1.59it/s]
 82%|████████▏ | 53667/65536 [9:14:56<2:06:04,  1.57it/s]
 82%|████████▏ | 53668/65536 [9:14:57<2:02:27,  1.62it/s]
 82%|████████▏ | 53669/65536 [9:14:58<2:02:48,  1.61it/s]
 82%|████████▏ | 53670/65536 [9:14:58<2:03:16,  1.60it/s]
 82%|████████▏ | 53671/65536 [9:14:59<2:08:26,  1.54it/s]
 82%|████████▏ | 53672/65536 [9:15:00<2:08:21,  1.54it/s]
 82%|████████▏ | 53673/65536 [9:15:00<2:05:40,  1.57it/s]
 82%|████████▏ | 53674/65536 [9:15:01<2:04:22,  1.59it/s]
 82%|████████▏ | 53675/65536 [9:15:02<2:06:26,  1.56it/s]
 82%|████████▏ | 53676/65536 [9:15:02<2:10:55,  1.51it/s]
 82%|████████▏ | 53677/65536 [9:15:03<2:07:23,  1.55it/s]
 82%|████████▏ | 53678/65536 [9:15:03<2:05:38,  1.57it/s]
 82%|████████▏ | 53679/65536 [9:15:04<2:06:50,  1.56it/s]
 82%|████████▏ | 53680/65536 [9:15:05<2:04:40,  1.58it/s]
                                                         
{'loss': 1.5667, 'learning_rate': 2.633157830292641e-07, 'epoch': 3313.58}
+
 82%|████████▏ | 53680/65536 [9:15:05<2:04:40,  1.58it/s]
 82%|████████▏ | 53681/65536 [9:15:05<2:04:29,  1.59it/s]
 82%|████████▏ | 53682/65536 [9:15:06<2:07:17,  1.55it/s]
 82%|████████▏ | 53683/65536 [9:15:07<2:08:31,  1.54it/s]
 82%|████████▏ | 53684/65536 [9:15:07<2:05:35,  1.57it/s]
 82%|████████▏ | 53685/65536 [9:15:08<2:00:53,  1.63it/s]
 82%|████████▏ | 53686/65536 [9:15:08<2:00:24,  1.64it/s]
 82%|████████▏ | 53687/65536 [9:15:09<2:03:23,  1.60it/s]
 82%|████████▏ | 53688/65536 [9:15:10<2:02:54,  1.61it/s]
 82%|████████▏ | 53689/65536 [9:15:10<2:04:22,  1.59it/s]
 82%|████████▏ | 53690/65536 [9:15:11<2:03:34,  1.60it/s]
 82%|████████▏ | 53691/65536 [9:15:12<2:03:23,  1.60it/s]
 82%|████████▏ | 53692/65536 [9:15:12<2:03:19,  1.60it/s]
 82%|████████▏ | 53693/65536 [9:15:13<2:01:32,  1.62it/s]
 82%|████████▏ | 53694/65536 [9:15:13<2:02:44,  1.61it/s]
 82%|████████▏ | 53695/65536 [9:15:14<2:01:48,  1.62it/s]
 82%|████████▏ | 53696/65536 [9:15:15<1:59:55,  1.65it/s]
 82%|████████▏ | 53697/65536 [9:15:15<1:59:12,  1.66it/s]
 82%|████████▏ | 53698/65536 [9:15:16<1:59:17,  1.65it/s]
 82%|████████▏ | 53699/65536 [9:15:16<1:59:43,  1.65it/s]
 82%|████████▏ | 53700/65536 [9:15:17<2:00:06,  1.64it/s]
                                                         
{'loss': 1.5926, 'learning_rate': 2.6304028407003795e-07, 'epoch': 3314.81}
+
 82%|████████▏ | 53700/65536 [9:15:17<2:00:06,  1.64it/s]
 82%|████████▏ | 53701/65536 [9:15:18<2:00:42,  1.63it/s]
 82%|████████▏ | 53702/65536 [9:15:18<2:02:00,  1.62it/s]
 82%|████████▏ | 53703/65536 [9:15:19<2:04:02,  1.59it/s]
 82%|████████▏ | 53704/65536 [9:15:20<2:10:55,  1.51it/s]
 82%|████████▏ | 53705/65536 [9:15:20<2:11:13,  1.50it/s]
 82%|████████▏ | 53706/65536 [9:15:21<2:10:27,  1.51it/s]
 82%|████████▏ | 53707/65536 [9:15:22<2:07:08,  1.55it/s]
 82%|████████▏ | 53708/65536 [9:15:22<2:06:29,  1.56it/s]
 82%|████████▏ | 53709/65536 [9:15:23<2:04:27,  1.58it/s]
 82%|████████▏ | 53710/65536 [9:15:24<2:03:47,  1.59it/s]
 82%|████████▏ | 53711/65536 [9:15:24<2:02:06,  1.61it/s]
 82%|████████▏ | 53712/65536 [9:15:25<2:00:02,  1.64it/s]
 82%|████████▏ | 53713/65536 [9:15:25<1:58:54,  1.66it/s]
 82%|████████▏ | 53714/65536 [9:15:26<1:59:51,  1.64it/s]
 82%|████████▏ | 53715/65536 [9:15:27<1:59:57,  1.64it/s]
 82%|████████▏ | 53716/65536 [9:15:27<2:03:01,  1.60it/s]
 82%|████████▏ | 53717/65536 [9:15:28<2:03:56,  1.59it/s]
 82%|████████▏ | 53718/65536 [9:15:28<2:03:16,  1.60it/s]
 82%|████████▏ | 53719/65536 [9:15:29<2:02:08,  1.61it/s]
 82%|████████▏ | 53720/65536 [9:15:30<2:05:37,  1.57it/s]
                                                         
{'loss': 1.5634, 'learning_rate': 2.627647851108118e-07, 'epoch': 3316.05}
+
 82%|████████▏ | 53720/65536 [9:15:30<2:05:37,  1.57it/s]
 82%|████████▏ | 53721/65536 [9:15:30<2:04:10,  1.59it/s]
 82%|████████▏ | 53722/65536 [9:15:31<2:05:47,  1.57it/s]
 82%|████████▏ | 53723/65536 [9:15:32<2:04:58,  1.58it/s]
 82%|████████▏ | 53724/65536 [9:15:32<2:02:12,  1.61it/s]
 82%|████████▏ | 53725/65536 [9:15:33<2:00:00,  1.64it/s]
 82%|████████▏ | 53726/65536 [9:15:33<2:00:21,  1.64it/s]
 82%|████████▏ | 53727/65536 [9:15:34<2:01:38,  1.62it/s]
 82%|████████▏ | 53728/65536 [9:15:35<1:59:28,  1.65it/s]
 82%|████████▏ | 53729/65536 [9:15:35<2:01:40,  1.62it/s]
 82%|████████▏ | 53730/65536 [9:15:36<2:02:43,  1.60it/s]
 82%|████████▏ | 53731/65536 [9:15:37<2:03:40,  1.59it/s]
 82%|████████▏ | 53732/65536 [9:15:37<2:02:42,  1.60it/s]
 82%|████████▏ | 53733/65536 [9:15:38<2:02:46,  1.60it/s]
 82%|████████▏ | 53734/65536 [9:15:38<2:02:57,  1.60it/s]
 82%|████████▏ | 53735/65536 [9:15:39<2:03:55,  1.59it/s]
 82%|████████▏ | 53736/65536 [9:15:40<2:11:10,  1.50it/s]
 82%|████████▏ | 53737/65536 [9:15:40<2:06:42,  1.55it/s]
 82%|████████▏ | 53738/65536 [9:15:41<2:08:07,  1.53it/s]
 82%|████████▏ | 53739/65536 [9:15:42<2:04:03,  1.58it/s]
 82%|████████▏ | 53740/65536 [9:15:42<2:06:40,  1.55it/s]
                                                         
{'loss': 1.5571, 'learning_rate': 2.6248928615158567e-07, 'epoch': 3317.28}
+
 82%|████████▏ | 53740/65536 [9:15:42<2:06:40,  1.55it/s]
 82%|████████▏ | 53741/65536 [9:15:43<2:05:25,  1.57it/s]
 82%|████████▏ | 53742/65536 [9:15:44<2:01:51,  1.61it/s]
 82%|████████▏ | 53743/65536 [9:15:44<2:01:37,  1.62it/s]
 82%|████████▏ | 53744/65536 [9:15:45<2:00:21,  1.63it/s]
 82%|████████▏ | 53745/65536 [9:15:45<2:01:15,  1.62it/s]
 82%|████████▏ | 53746/65536 [9:15:46<2:02:09,  1.61it/s]
 82%|████████▏ | 53747/65536 [9:15:47<2:00:14,  1.63it/s]
 82%|████████▏ | 53748/65536 [9:15:47<1:57:37,  1.67it/s]
 82%|████████▏ | 53749/65536 [9:15:48<2:00:24,  1.63it/s]
 82%|████████▏ | 53750/65536 [9:15:48<2:00:37,  1.63it/s]
 82%|████████▏ | 53751/65536 [9:15:49<2:01:05,  1.62it/s]
 82%|████████▏ | 53752/65536 [9:15:50<2:04:14,  1.58it/s]
 82%|████████▏ | 53753/65536 [9:15:50<2:07:09,  1.54it/s]
 82%|████████▏ | 53754/65536 [9:15:51<2:05:58,  1.56it/s]
 82%|████████▏ | 53755/65536 [9:15:52<2:07:29,  1.54it/s]
 82%|████████▏ | 53756/65536 [9:15:52<2:04:19,  1.58it/s]
 82%|████████▏ | 53757/65536 [9:15:53<2:02:11,  1.61it/s]
 82%|████████▏ | 53758/65536 [9:15:54<2:03:13,  1.59it/s]
 82%|████████▏ | 53759/65536 [9:15:54<2:01:43,  1.61it/s]
 82%|████████▏ | 53760/65536 [9:15:55<1:59:52,  1.64it/s]
                                                         
{'loss': 1.5422, 'learning_rate': 2.6221378719235953e-07, 'epoch': 3318.52}
+
 82%|████████▏ | 53760/65536 [9:15:55<1:59:52,  1.64it/s]
 82%|████████▏ | 53761/65536 [9:15:55<1:59:36,  1.64it/s]
 82%|████████▏ | 53762/65536 [9:15:56<2:01:26,  1.62it/s]
 82%|████████▏ | 53763/65536 [9:15:57<2:02:12,  1.61it/s]
 82%|████████▏ | 53764/65536 [9:15:57<1:59:35,  1.64it/s]
 82%|████████▏ | 53765/65536 [9:15:58<2:00:25,  1.63it/s]
 82%|████████▏ | 53766/65536 [9:15:58<2:01:56,  1.61it/s]
 82%|████████▏ | 53767/65536 [9:15:59<2:01:26,  1.62it/s]
 82%|████████▏ | 53768/65536 [9:16:00<2:03:23,  1.59it/s]
 82%|████████▏ | 53769/65536 [9:16:00<2:01:04,  1.62it/s]
 82%|████████▏ | 53770/65536 [9:16:01<2:00:11,  1.63it/s]
 82%|████████▏ | 53771/65536 [9:16:02<2:02:46,  1.60it/s]
 82%|████████▏ | 53772/65536 [9:16:02<2:03:45,  1.58it/s]
 82%|████████▏ | 53773/65536 [9:16:03<2:01:20,  1.62it/s]
 82%|████████▏ | 53774/65536 [9:16:03<2:01:49,  1.61it/s]
 82%|████████▏ | 53775/65536 [9:16:04<1:58:54,  1.65it/s]
 82%|████████▏ | 53776/65536 [9:16:05<2:00:35,  1.63it/s]
 82%|████████▏ | 53777/65536 [9:16:05<2:00:20,  1.63it/s]
 82%|████████▏ | 53778/65536 [9:16:06<2:01:36,  1.61it/s]
 82%|████████▏ | 53779/65536 [9:16:07<2:05:23,  1.56it/s]
 82%|████████▏ | 53780/65536 [9:16:07<2:04:40,  1.57it/s]
                                                         
{'loss': 1.5624, 'learning_rate': 2.619382882331333e-07, 'epoch': 3319.75}
+
 82%|████████▏ | 53780/65536 [9:16:07<2:04:40,  1.57it/s]
 82%|████████▏ | 53781/65536 [9:16:08<2:03:17,  1.59it/s]
 82%|████████▏ | 53782/65536 [9:16:08<2:02:48,  1.60it/s]
 82%|████████▏ | 53783/65536 [9:16:09<2:00:15,  1.63it/s]
 82%|████████▏ | 53784/65536 [9:16:10<1:59:29,  1.64it/s]
 82%|████████▏ | 53785/65536 [9:16:10<2:03:31,  1.59it/s]
 82%|████████▏ | 53786/65536 [9:16:11<2:04:30,  1.57it/s]
 82%|████████▏ | 53787/65536 [9:16:12<2:03:21,  1.59it/s]
 82%|████████▏ | 53788/65536 [9:16:12<2:00:27,  1.63it/s]
 82%|████████▏ | 53789/65536 [9:16:13<2:02:28,  1.60it/s]
 82%|████████▏ | 53790/65536 [9:16:13<2:01:48,  1.61it/s]
 82%|████████▏ | 53791/65536 [9:16:14<2:04:02,  1.58it/s]
 82%|████████▏ | 53792/65536 [9:16:15<2:06:10,  1.55it/s]
 82%|████████▏ | 53793/65536 [9:16:15<2:01:59,  1.60it/s]
 82%|████████▏ | 53794/65536 [9:16:16<2:04:44,  1.57it/s]
 82%|████████▏ | 53795/65536 [9:16:17<2:03:13,  1.59it/s]
 82%|████████▏ | 53796/65536 [9:16:17<2:00:02,  1.63it/s]
 82%|████████▏ | 53797/65536 [9:16:18<1:59:17,  1.64it/s]
 82%|████████▏ | 53798/65536 [9:16:18<1:58:00,  1.66it/s]
 82%|████████▏ | 53799/65536 [9:16:19<1:56:32,  1.68it/s]
 82%|████████▏ | 53800/65536 [9:16:20<1:59:52,  1.63it/s]
                                                         
{'loss': 1.545, 'learning_rate': 2.616627892739072e-07, 'epoch': 3320.99}
+
 82%|████████▏ | 53800/65536 [9:16:20<1:59:52,  1.63it/s]
 82%|████████▏ | 53801/65536 [9:16:20<2:05:18,  1.56it/s]
 82%|████████▏ | 53802/65536 [9:16:21<2:02:41,  1.59it/s]
 82%|████████▏ | 53803/65536 [9:16:22<2:02:00,  1.60it/s]
 82%|████████▏ | 53804/65536 [9:16:22<2:03:21,  1.58it/s]
 82%|████████▏ | 53805/65536 [9:16:23<2:02:15,  1.60it/s]
 82%|████████▏ | 53806/65536 [9:16:23<1:59:51,  1.63it/s]
 82%|████████▏ | 53807/65536 [9:16:24<1:58:03,  1.66it/s]
 82%|████████▏ | 53808/65536 [9:16:25<1:57:44,  1.66it/s]
 82%|████████▏ | 53809/65536 [9:16:25<1:58:22,  1.65it/s]
 82%|████████▏ | 53810/65536 [9:16:26<2:00:23,  1.62it/s]
 82%|████████▏ | 53811/65536 [9:16:26<1:59:43,  1.63it/s]
 82%|████████▏ | 53812/65536 [9:16:27<2:01:02,  1.61it/s]
 82%|████████▏ | 53813/65536 [9:16:28<2:01:57,  1.60it/s]
 82%|████████▏ | 53814/65536 [9:16:28<2:02:40,  1.59it/s]
 82%|████████▏ | 53815/65536 [9:16:29<2:08:09,  1.52it/s]
 82%|████████▏ | 53816/65536 [9:16:30<2:10:43,  1.49it/s]
 82%|████████▏ | 53817/65536 [9:16:30<2:14:28,  1.45it/s]
 82%|████████▏ | 53818/65536 [9:16:31<2:09:06,  1.51it/s]
 82%|████████▏ | 53819/65536 [9:16:32<2:05:02,  1.56it/s]
 82%|████████▏ | 53820/65536 [9:16:32<2:05:55,  1.55it/s]
                                                         
{'loss': 1.5573, 'learning_rate': 2.61387290314681e-07, 'epoch': 3322.22}
+
 82%|████████▏ | 53820/65536 [9:16:32<2:05:55,  1.55it/s]
 82%|████████▏ | 53821/65536 [9:16:33<2:05:34,  1.55it/s]
 82%|████████▏ | 53822/65536 [9:16:34<2:02:51,  1.59it/s]
 82%|████████▏ | 53823/65536 [9:16:34<2:04:37,  1.57it/s]
 82%|████████▏ | 53824/65536 [9:16:35<2:03:38,  1.58it/s]
 82%|████████▏ | 53825/65536 [9:16:35<2:05:31,  1.55it/s]
 82%|████████▏ | 53826/65536 [9:16:36<2:03:53,  1.58it/s]
 82%|████████▏ | 53827/65536 [9:16:37<2:04:37,  1.57it/s]
 82%|████████▏ | 53828/65536 [9:16:37<2:05:58,  1.55it/s]
 82%|████████▏ | 53829/65536 [9:16:38<2:04:04,  1.57it/s]
 82%|████████▏ | 53830/65536 [9:16:39<2:02:10,  1.60it/s]
 82%|████████▏ | 53831/65536 [9:16:39<2:01:38,  1.60it/s]
 82%|████████▏ | 53832/65536 [9:16:40<1:59:25,  1.63it/s]
 82%|████████▏ | 53833/65536 [9:16:40<2:01:47,  1.60it/s]
 82%|████████▏ | 53834/65536 [9:16:41<2:04:56,  1.56it/s]
 82%|████████▏ | 53835/65536 [9:16:42<2:04:38,  1.56it/s]
 82%|████████▏ | 53836/65536 [9:16:42<2:02:16,  1.59it/s]
 82%|████████▏ | 53837/65536 [9:16:43<2:01:36,  1.60it/s]
 82%|████████▏ | 53838/65536 [9:16:44<1:59:38,  1.63it/s]
 82%|████████▏ | 53839/65536 [9:16:44<1:57:00,  1.67it/s]
 82%|████████▏ | 53840/65536 [9:16:45<1:58:06,  1.65it/s]
                                                         
{'loss': 1.5567, 'learning_rate': 2.6111179135545486e-07, 'epoch': 3323.46}
+
 82%|████████▏ | 53840/65536 [9:16:45<1:58:06,  1.65it/s]
 82%|████████▏ | 53841/65536 [9:16:45<1:58:26,  1.65it/s]
 82%|████████▏ | 53842/65536 [9:16:46<1:59:28,  1.63it/s]
 82%|████████▏ | 53843/65536 [9:16:47<1:57:33,  1.66it/s]
 82%|████████▏ | 53844/65536 [9:16:47<2:00:31,  1.62it/s]
 82%|████████▏ | 53845/65536 [9:16:48<2:00:19,  1.62it/s]
 82%|████████▏ | 53846/65536 [9:16:48<1:57:30,  1.66it/s]
 82%|████████▏ | 53847/65536 [9:16:49<2:01:01,  1.61it/s]
 82%|████████▏ | 53848/65536 [9:16:50<2:02:21,  1.59it/s]
 82%|████████▏ | 53849/65536 [9:16:50<2:06:49,  1.54it/s]
 82%|████████▏ | 53850/65536 [9:16:51<2:06:40,  1.54it/s]
 82%|████████▏ | 53851/65536 [9:16:52<2:04:56,  1.56it/s]
 82%|████████▏ | 53852/65536 [9:16:52<2:01:43,  1.60it/s]
 82%|████████▏ | 53853/65536 [9:16:53<2:05:39,  1.55it/s]
 82%|████████▏ | 53854/65536 [9:16:54<2:08:38,  1.51it/s]
 82%|████████▏ | 53855/65536 [9:16:54<2:07:03,  1.53it/s]
 82%|████████▏ | 53856/65536 [9:16:55<2:04:58,  1.56it/s]
 82%|████████▏ | 53857/65536 [9:16:56<2:00:57,  1.61it/s]
 82%|████████▏ | 53858/65536 [9:16:56<1:57:46,  1.65it/s]
 82%|████████▏ | 53859/65536 [9:16:57<1:57:02,  1.66it/s]
 82%|████████▏ | 53860/65536 [9:16:57<1:57:07,  1.66it/s]
                                                         
{'loss': 1.5713, 'learning_rate': 2.608362923962287e-07, 'epoch': 3324.69}
+
 82%|████████▏ | 53860/65536 [9:16:57<1:57:07,  1.66it/s]
 82%|████████▏ | 53861/65536 [9:16:58<1:56:11,  1.67it/s]
 82%|████████▏ | 53862/65536 [9:16:59<1:58:05,  1.65it/s]
 82%|████████▏ | 53863/65536 [9:16:59<1:54:48,  1.69it/s]
 82%|████████▏ | 53864/65536 [9:17:00<1:55:45,  1.68it/s]
 82%|████████▏ | 53865/65536 [9:17:00<1:54:45,  1.70it/s]
 82%|████████▏ | 53866/65536 [9:17:01<2:02:49,  1.58it/s]
 82%|████████▏ | 53867/65536 [9:17:02<2:03:07,  1.58it/s]
 82%|████████▏ | 53868/65536 [9:17:02<2:02:18,  1.59it/s]
 82%|████████▏ | 53869/65536 [9:17:03<2:01:20,  1.60it/s]
 82%|████████▏ | 53870/65536 [9:17:03<1:59:54,  1.62it/s]
 82%|████████▏ | 53871/65536 [9:17:04<2:01:27,  1.60it/s]
 82%|████████▏ | 53872/65536 [9:17:05<2:00:49,  1.61it/s]
 82%|████████▏ | 53873/65536 [9:17:05<1:59:54,  1.62it/s]
 82%|████████▏ | 53874/65536 [9:17:06<1:59:07,  1.63it/s]
 82%|████████▏ | 53875/65536 [9:17:06<1:57:54,  1.65it/s]
 82%|████████▏ | 53876/65536 [9:17:07<1:59:00,  1.63it/s]
 82%|████████▏ | 53877/65536 [9:17:08<2:01:27,  1.60it/s]
 82%|████████▏ | 53878/65536 [9:17:08<2:00:43,  1.61it/s]
 82%|████████▏ | 53879/65536 [9:17:09<2:01:40,  1.60it/s]
 82%|████████▏ | 53880/65536 [9:17:10<2:01:05,  1.60it/s]
                                                         
{'loss': 1.5671, 'learning_rate': 2.605607934370026e-07, 'epoch': 3325.93}
+
 82%|████████▏ | 53880/65536 [9:17:10<2:01:05,  1.60it/s]
 82%|████████▏ | 53881/65536 [9:17:10<2:01:06,  1.60it/s]
 82%|████████▏ | 53882/65536 [9:17:11<2:03:02,  1.58it/s]
 82%|████████▏ | 53883/65536 [9:17:12<2:01:16,  1.60it/s]
 82%|████████▏ | 53884/65536 [9:17:12<1:58:27,  1.64it/s]
 82%|████████▏ | 53885/65536 [9:17:13<1:57:58,  1.65it/s]
 82%|████████▏ | 53886/65536 [9:17:13<1:56:58,  1.66it/s]
 82%|████████▏ | 53887/65536 [9:17:14<1:56:32,  1.67it/s]
 82%|████████▏ | 53888/65536 [9:17:14<1:54:53,  1.69it/s]
 82%|████████▏ | 53889/65536 [9:17:15<1:56:57,  1.66it/s]
 82%|████████▏ | 53890/65536 [9:17:16<2:00:34,  1.61it/s]
 82%|████████▏ | 53891/65536 [9:17:16<2:01:28,  1.60it/s]
 82%|████████▏ | 53892/65536 [9:17:17<2:03:01,  1.58it/s]
 82%|████████▏ | 53893/65536 [9:17:18<2:05:29,  1.55it/s]
 82%|████████▏ | 53894/65536 [9:17:18<2:04:11,  1.56it/s]
 82%|████████▏ | 53895/65536 [9:17:19<2:02:34,  1.58it/s]
 82%|████████▏ | 53896/65536 [9:17:20<2:02:56,  1.58it/s]
 82%|████████▏ | 53897/65536 [9:17:20<2:00:39,  1.61it/s]
 82%|████████▏ | 53898/65536 [9:17:21<2:03:35,  1.57it/s]
 82%|████████▏ | 53899/65536 [9:17:22<2:03:23,  1.57it/s]
 82%|████████▏ | 53900/65536 [9:17:22<2:04:00,  1.56it/s]
                                                         
{'loss': 1.5582, 'learning_rate': 2.6028529447777644e-07, 'epoch': 3327.16}
+
 82%|████████▏ | 53900/65536 [9:17:22<2:04:00,  1.56it/s]
 82%|████████▏ | 53901/65536 [9:17:23<2:01:29,  1.60it/s]
 82%|████████▏ | 53902/65536 [9:17:23<2:04:38,  1.56it/s]
 82%|████████▏ | 53903/65536 [9:17:24<2:02:03,  1.59it/s]
 82%|████████▏ | 53904/65536 [9:17:25<1:58:18,  1.64it/s]
 82%|████████▏ | 53905/65536 [9:17:25<1:59:00,  1.63it/s]
 82%|████████▏ | 53906/65536 [9:17:26<1:59:01,  1.63it/s]
 82%|████████▏ | 53907/65536 [9:17:26<2:01:04,  1.60it/s]
 82%|████████▏ | 53908/65536 [9:17:27<1:59:02,  1.63it/s]
 82%|████████▏ | 53909/65536 [9:17:28<1:58:20,  1.64it/s]
 82%|████████▏ | 53910/65536 [9:17:28<2:01:33,  1.59it/s]
 82%|████████▏ | 53911/65536 [9:17:29<1:59:31,  1.62it/s]
 82%|████████▏ | 53912/65536 [9:17:30<1:57:20,  1.65it/s]
 82%|████████▏ | 53913/65536 [9:17:30<1:56:41,  1.66it/s]
 82%|████████▏ | 53914/65536 [9:17:31<1:59:04,  1.63it/s]
 82%|████████▏ | 53915/65536 [9:17:31<2:03:57,  1.56it/s]
 82%|████████▏ | 53916/65536 [9:17:32<2:01:08,  1.60it/s]
 82%|████████▏ | 53917/65536 [9:17:33<2:04:38,  1.55it/s]
 82%|████████▏ | 53918/65536 [9:17:33<2:03:36,  1.57it/s]
 82%|████████▏ | 53919/65536 [9:17:34<2:02:33,  1.58it/s]
 82%|████████▏ | 53920/65536 [9:17:35<1:59:52,  1.61it/s]
                                                         
{'loss': 1.5544, 'learning_rate': 2.600097955185502e-07, 'epoch': 3328.4}
+
 82%|████████▏ | 53920/65536 [9:17:35<1:59:52,  1.61it/s]
 82%|████████▏ | 53921/65536 [9:17:35<1:59:44,  1.62it/s]
 82%|████████▏ | 53922/65536 [9:17:36<1:59:14,  1.62it/s]
 82%|████████▏ | 53923/65536 [9:17:36<1:58:37,  1.63it/s]
 82%|████████▏ | 53924/65536 [9:17:37<1:58:28,  1.63it/s]
 82%|████████▏ | 53925/65536 [9:17:38<1:57:04,  1.65it/s]
 82%|████████▏ | 53926/65536 [9:17:38<1:56:50,  1.66it/s]
 82%|████████▏ | 53927/65536 [9:17:39<1:57:27,  1.65it/s]
 82%|████████▏ | 53928/65536 [9:17:39<2:01:23,  1.59it/s]
 82%|████████▏ | 53929/65536 [9:17:40<2:01:10,  1.60it/s]
 82%|████████▏ | 53930/65536 [9:17:41<2:04:27,  1.55it/s]
 82%|████████▏ | 53931/65536 [9:17:41<2:04:32,  1.55it/s]
 82%|████████▏ | 53932/65536 [9:17:42<2:05:42,  1.54it/s]
 82%|████████▏ | 53933/65536 [9:17:43<2:03:03,  1.57it/s]
 82%|████████▏ | 53934/65536 [9:17:43<1:59:29,  1.62it/s]
 82%|████████▏ | 53935/65536 [9:17:44<1:58:48,  1.63it/s]
 82%|████████▏ | 53936/65536 [9:17:45<2:01:09,  1.60it/s]
 82%|████████▏ | 53937/65536 [9:17:45<2:03:59,  1.56it/s]
 82%|████████▏ | 53938/65536 [9:17:46<2:02:21,  1.58it/s]
 82%|████████▏ | 53939/65536 [9:17:46<2:01:52,  1.59it/s]
 82%|████████▏ | 53940/65536 [9:17:47<2:00:15,  1.61it/s]
                                                         
{'loss': 1.5697, 'learning_rate': 2.5973429655932405e-07, 'epoch': 3329.63}
+
 82%|████████▏ | 53940/65536 [9:17:47<2:00:15,  1.61it/s]
 82%|████████▏ | 53941/65536 [9:17:48<1:58:02,  1.64it/s]
 82%|████████▏ | 53942/65536 [9:17:48<2:00:10,  1.61it/s]
 82%|████████▏ | 53943/65536 [9:17:49<1:59:06,  1.62it/s]
 82%|████████▏ | 53944/65536 [9:17:50<1:58:52,  1.63it/s]
 82%|████████▏ | 53945/65536 [9:17:50<1:58:55,  1.62it/s]
 82%|████████▏ | 53946/65536 [9:17:51<1:58:18,  1.63it/s]
 82%|████████▏ | 53947/65536 [9:17:51<2:01:50,  1.59it/s]
 82%|████████▏ | 53948/65536 [9:17:52<2:00:17,  1.61it/s]
 82%|████████▏ | 53949/65536 [9:17:53<2:03:42,  1.56it/s]
 82%|████████▏ | 53950/65536 [9:17:53<2:01:04,  1.59it/s]
 82%|████████▏ | 53951/65536 [9:17:54<2:01:34,  1.59it/s]
 82%|████████▏ | 53952/65536 [9:17:54<1:58:51,  1.62it/s]
 82%|████████▏ | 53953/65536 [9:17:55<1:57:45,  1.64it/s]
 82%|████████▏ | 53954/65536 [9:17:56<2:00:18,  1.60it/s]
 82%|████████▏ | 53955/65536 [9:17:56<2:01:30,  1.59it/s]
 82%|████████▏ | 53956/65536 [9:17:57<1:59:29,  1.62it/s]
 82%|████████▏ | 53957/65536 [9:17:58<1:59:04,  1.62it/s]
 82%|████████▏ | 53958/65536 [9:17:58<1:57:47,  1.64it/s]
 82%|████████▏ | 53959/65536 [9:17:59<1:57:32,  1.64it/s]
 82%|████████▏ | 53960/65536 [9:17:59<1:58:39,  1.63it/s]
                                                         
{'loss': 1.5314, 'learning_rate': 2.594587976000979e-07, 'epoch': 3330.86}
+
 82%|████████▏ | 53960/65536 [9:17:59<1:58:39,  1.63it/s]
 82%|████████▏ | 53961/65536 [9:18:00<1:57:02,  1.65it/s]
 82%|████████▏ | 53962/65536 [9:18:01<1:59:57,  1.61it/s]
 82%|████████▏ | 53963/65536 [9:18:01<2:04:09,  1.55it/s]
 82%|████████▏ | 53964/65536 [9:18:02<2:01:56,  1.58it/s]
 82%|████████▏ | 53965/65536 [9:18:03<2:01:21,  1.59it/s]
 82%|████████▏ | 53966/65536 [9:18:03<1:58:07,  1.63it/s]
 82%|████████▏ | 53967/65536 [9:18:04<1:59:12,  1.62it/s]
 82%|████████▏ | 53968/65536 [9:18:04<2:01:04,  1.59it/s]
 82%|████████▏ | 53969/65536 [9:18:05<2:01:33,  1.59it/s]
 82%|████████▏ | 53970/65536 [9:18:06<2:01:57,  1.58it/s]
 82%|████████▏ | 53971/65536 [9:18:06<2:01:43,  1.58it/s]
 82%|████████▏ | 53972/65536 [9:18:07<2:01:32,  1.59it/s]
 82%|████████▏ | 53973/65536 [9:18:08<1:59:24,  1.61it/s]
 82%|████████▏ | 53974/65536 [9:18:08<2:00:04,  1.60it/s]
 82%|████████▏ | 53975/65536 [9:18:09<1:56:16,  1.66it/s]
 82%|████████▏ | 53976/65536 [9:18:09<1:54:00,  1.69it/s]
 82%|████████▏ | 53977/65536 [9:18:10<1:55:55,  1.66it/s]
 82%|████████▏ | 53978/65536 [9:18:11<1:55:51,  1.66it/s]
 82%|████████▏ | 53979/65536 [9:18:11<2:04:08,  1.55it/s]
 82%|████████▏ | 53980/65536 [9:18:12<2:02:49,  1.57it/s]
                                                         
{'loss': 1.5656, 'learning_rate': 2.591832986408717e-07, 'epoch': 3332.1}
+
 82%|████████▏ | 53980/65536 [9:18:12<2:02:49,  1.57it/s]
 82%|████████▏ | 53981/65536 [9:18:13<2:03:54,  1.55it/s]
 82%|████████▏ | 53982/65536 [9:18:13<2:04:25,  1.55it/s]
 82%|████████▏ | 53983/65536 [9:18:14<2:01:16,  1.59it/s]
 82%|████████▏ | 53984/65536 [9:18:14<1:59:47,  1.61it/s]
 82%|████████▏ | 53985/65536 [9:18:15<2:00:36,  1.60it/s]
 82%|████████▏ | 53986/65536 [9:18:16<2:01:38,  1.58it/s]
 82%|████████▏ | 53987/65536 [9:18:16<1:58:12,  1.63it/s]
 82%|████████▏ | 53988/65536 [9:18:17<1:57:44,  1.63it/s]
 82%|████████▏ | 53989/65536 [9:18:18<1:57:54,  1.63it/s]
 82%|████████▏ | 53990/65536 [9:18:18<1:58:05,  1.63it/s]
 82%|████████▏ | 53991/65536 [9:18:19<1:58:30,  1.62it/s]
 82%|████████▏ | 53992/65536 [9:18:19<1:58:15,  1.63it/s]
 82%|████████▏ | 53993/65536 [9:18:20<1:58:21,  1.63it/s]
 82%|████████▏ | 53994/65536 [9:18:21<1:56:40,  1.65it/s]
 82%|████████▏ | 53995/65536 [9:18:21<1:59:38,  1.61it/s]
 82%|████████▏ | 53996/65536 [9:18:22<1:59:27,  1.61it/s]
 82%|████████▏ | 53997/65536 [9:18:22<1:58:43,  1.62it/s]
 82%|████████▏ | 53998/65536 [9:18:23<2:00:38,  1.59it/s]
 82%|████████▏ | 53999/65536 [9:18:24<1:59:04,  1.61it/s]
 82%|████████▏ | 54000/65536 [9:18:24<1:59:01,  1.62it/s]
                                                         
{'loss': 1.576, 'learning_rate': 2.589077996816456e-07, 'epoch': 3333.33}
+
 82%|████████▏ | 54000/65536 [9:18:24<1:59:01,  1.62it/s]
 82%|████████▏ | 54001/65536 [9:18:25<2:01:13,  1.59it/s]
 82%|████████▏ | 54002/65536 [9:18:26<2:02:16,  1.57it/s]
 82%|████████▏ | 54003/65536 [9:18:26<2:00:48,  1.59it/s]
 82%|████████▏ | 54004/65536 [9:18:27<2:03:14,  1.56it/s]
 82%|████████▏ | 54005/65536 [9:18:27<2:00:02,  1.60it/s]
 82%|████████▏ | 54006/65536 [9:18:28<1:56:58,  1.64it/s]
 82%|████████▏ | 54007/65536 [9:18:29<1:56:42,  1.65it/s]
 82%|████████▏ | 54008/65536 [9:18:29<1:55:41,  1.66it/s]
 82%|████████▏ | 54009/65536 [9:18:30<1:56:15,  1.65it/s]
 82%|████████▏ | 54010/65536 [9:18:30<1:54:09,  1.68it/s]
 82%|████████▏ | 54011/65536 [9:18:31<1:58:57,  1.61it/s]
 82%|████████▏ | 54012/65536 [9:18:32<1:59:13,  1.61it/s]
 82%|████████▏ | 54013/65536 [9:18:32<1:56:31,  1.65it/s]
 82%|████████▏ | 54014/65536 [9:18:33<1:57:02,  1.64it/s]
 82%|████████▏ | 54015/65536 [9:18:34<1:56:50,  1.64it/s]
 82%|████████▏ | 54016/65536 [9:18:34<1:58:02,  1.63it/s]
 82%|████████▏ | 54017/65536 [9:18:35<1:58:07,  1.63it/s]
 82%|████████▏ | 54018/65536 [9:18:35<1:59:00,  1.61it/s]
 82%|████████▏ | 54019/65536 [9:18:36<1:57:07,  1.64it/s]
 82%|████████▏ | 54020/65536 [9:18:37<1:58:13,  1.62it/s]
                                                         
{'loss': 1.5541, 'learning_rate': 2.586323007224195e-07, 'epoch': 3334.57}
+
 82%|████████▏ | 54020/65536 [9:18:37<1:58:13,  1.62it/s]
 82%|████████▏ | 54021/65536 [9:18:37<1:57:43,  1.63it/s]
 82%|████████▏ | 54022/65536 [9:18:38<1:56:01,  1.65it/s]
 82%|████████▏ | 54023/65536 [9:18:38<1:57:57,  1.63it/s]
 82%|████████▏ | 54024/65536 [9:18:39<2:00:00,  1.60it/s]
 82%|████████▏ | 54025/65536 [9:18:40<2:00:08,  1.60it/s]
 82%|████████▏ | 54026/65536 [9:18:40<1:57:35,  1.63it/s]
 82%|████████▏ | 54027/65536 [9:18:41<1:58:11,  1.62it/s]
 82%|████████▏ | 54028/65536 [9:18:42<1:59:59,  1.60it/s]
 82%|████████▏ | 54029/65536 [9:18:42<1:59:16,  1.61it/s]
 82%|████████▏ | 54030/65536 [9:18:43<1:55:16,  1.66it/s]
 82%|████████▏ | 54031/65536 [9:18:43<1:53:48,  1.68it/s]
 82%|████████▏ | 54032/65536 [9:18:44<1:55:09,  1.67it/s]
 82%|████████▏ | 54033/65536 [9:18:45<1:59:02,  1.61it/s]
 82%|████████▏ | 54034/65536 [9:18:45<1:57:23,  1.63it/s]
 82%|████████▏ | 54035/65536 [9:18:46<2:00:27,  1.59it/s]
 82%|████████▏ | 54036/65536 [9:18:46<2:00:02,  1.60it/s]
 82%|████████▏ | 54037/65536 [9:18:47<1:59:30,  1.60it/s]
 82%|████████▏ | 54038/65536 [9:18:48<2:00:07,  1.60it/s]
 82%|████████▏ | 54039/65536 [9:18:48<1:58:59,  1.61it/s]
 82%|████████▏ | 54040/65536 [9:18:49<1:59:06,  1.61it/s]
                                                         
{'loss': 1.6052, 'learning_rate': 2.5835680176319334e-07, 'epoch': 3335.8}
+
 82%|████████▏ | 54040/65536 [9:18:49<1:59:06,  1.61it/s]
 82%|████████▏ | 54041/65536 [9:18:50<1:57:43,  1.63it/s]
 82%|████████▏ | 54042/65536 [9:18:50<1:58:19,  1.62it/s]
 82%|████████▏ | 54043/65536 [9:18:51<1:58:46,  1.61it/s]
 82%|████████▏ | 54044/65536 [9:18:52<2:03:36,  1.55it/s]
 82%|████████▏ | 54045/65536 [9:18:52<2:02:31,  1.56it/s]
 82%|████████▏ | 54046/65536 [9:18:53<2:01:15,  1.58it/s]
 82%|████████▏ | 54047/65536 [9:18:53<2:01:59,  1.57it/s]
 82%|████████▏ | 54048/65536 [9:18:54<2:01:16,  1.58it/s]
 82%|████████▏ | 54049/65536 [9:18:55<1:59:57,  1.60it/s]
 82%|████████▏ | 54050/65536 [9:18:55<1:57:04,  1.64it/s]
 82%|████████▏ | 54051/65536 [9:18:56<2:00:15,  1.59it/s]
 82%|████████▏ | 54052/65536 [9:18:56<1:58:17,  1.62it/s]
 82%|████████▏ | 54053/65536 [9:18:57<1:57:46,  1.62it/s]
 82%|████████▏ | 54054/65536 [9:18:58<1:55:32,  1.66it/s]
 82%|████████▏ | 54055/65536 [9:18:58<1:56:35,  1.64it/s]
 82%|████████▏ | 54056/65536 [9:18:59<1:57:14,  1.63it/s]
 82%|████████▏ | 54057/65536 [9:19:00<1:58:29,  1.61it/s]
 82%|████████▏ | 54058/65536 [9:19:00<1:59:06,  1.61it/s]
 82%|████████▏ | 54059/65536 [9:19:01<1:58:23,  1.62it/s]
 82%|████████▏ | 54060/65536 [9:19:01<2:02:00,  1.57it/s]
                                                         
{'loss': 1.5452, 'learning_rate': 2.5808130280396715e-07, 'epoch': 3337.04}
+
 82%|████████▏ | 54060/65536 [9:19:01<2:02:00,  1.57it/s]
 82%|████████▏ | 54061/65536 [9:19:02<1:59:14,  1.60it/s]
 82%|████████▏ | 54062/65536 [9:19:03<1:58:17,  1.62it/s]
 82%|████████▏ | 54063/65536 [9:19:03<2:02:21,  1.56it/s]
 82%|████████▏ | 54064/65536 [9:19:04<1:58:39,  1.61it/s]
 82%|████████▏ | 54065/65536 [9:19:05<1:57:11,  1.63it/s]
 82%|████████▏ | 54066/65536 [9:19:05<1:56:06,  1.65it/s]
 82%|████████▏ | 54067/65536 [9:19:06<1:59:44,  1.60it/s]
 83%|████████▎ | 54068/65536 [9:19:06<1:58:19,  1.62it/s]
 83%|████████▎ | 54069/65536 [9:19:07<1:59:11,  1.60it/s]
 83%|████████▎ | 54070/65536 [9:19:08<1:59:55,  1.59it/s]
 83%|████████▎ | 54071/65536 [9:19:08<1:59:25,  1.60it/s]
 83%|████████▎ | 54072/65536 [9:19:09<2:02:18,  1.56it/s]
 83%|████████▎ | 54073/65536 [9:19:10<1:59:11,  1.60it/s]
 83%|████████▎ | 54074/65536 [9:19:10<1:58:27,  1.61it/s]
 83%|████████▎ | 54075/65536 [9:19:11<1:56:27,  1.64it/s]
 83%|████████▎ | 54076/65536 [9:19:11<1:58:34,  1.61it/s]
 83%|████████▎ | 54077/65536 [9:19:12<1:58:37,  1.61it/s]
 83%|████████▎ | 54078/65536 [9:19:13<1:57:49,  1.62it/s]
 83%|████████▎ | 54079/65536 [9:19:13<1:56:08,  1.64it/s]
 83%|████████▎ | 54080/65536 [9:19:14<1:55:17,  1.66it/s]
                                                         
{'loss': 1.5654, 'learning_rate': 2.57805803844741e-07, 'epoch': 3338.27}
+
 83%|████████▎ | 54080/65536 [9:19:14<1:55:17,  1.66it/s]
 83%|████████▎ | 54081/65536 [9:19:14<1:56:37,  1.64it/s]
 83%|████████▎ | 54082/65536 [9:19:15<2:00:29,  1.58it/s]
 83%|████████▎ | 54083/65536 [9:19:16<2:00:54,  1.58it/s]
 83%|████████▎ | 54084/65536 [9:19:16<2:00:10,  1.59it/s]
 83%|████████▎ | 54085/65536 [9:19:17<2:00:28,  1.58it/s]
 83%|████████▎ | 54086/65536 [9:19:18<1:57:36,  1.62it/s]
 83%|████████▎ | 54087/65536 [9:19:18<1:56:35,  1.64it/s]
 83%|████████▎ | 54088/65536 [9:19:19<1:56:01,  1.64it/s]
 83%|████████▎ | 54089/65536 [9:19:19<1:59:13,  1.60it/s]
 83%|████████▎ | 54090/65536 [9:19:20<2:00:01,  1.59it/s]
 83%|████████▎ | 54091/65536 [9:19:21<2:00:42,  1.58it/s]
 83%|████████▎ | 54092/65536 [9:19:21<2:03:04,  1.55it/s]
 83%|████████▎ | 54093/65536 [9:19:22<2:01:29,  1.57it/s]
 83%|████████▎ | 54094/65536 [9:19:23<2:02:19,  1.56it/s]
 83%|████████▎ | 54095/65536 [9:19:23<2:00:33,  1.58it/s]
 83%|████████▎ | 54096/65536 [9:19:24<1:59:05,  1.60it/s]
 83%|████████▎ | 54097/65536 [9:19:25<1:59:35,  1.59it/s]
 83%|████████▎ | 54098/65536 [9:19:25<1:56:47,  1.63it/s]
 83%|████████▎ | 54099/65536 [9:19:26<1:59:08,  1.60it/s]
 83%|████████▎ | 54100/65536 [9:19:26<1:57:08,  1.63it/s]
                                                         
{'loss': 1.52, 'learning_rate': 2.5753030488551487e-07, 'epoch': 3339.51}
+
 83%|████████▎ | 54100/65536 [9:19:26<1:57:08,  1.63it/s]
 83%|████████▎ | 54101/65536 [9:19:27<2:00:09,  1.59it/s]
 83%|████████▎ | 54102/65536 [9:19:28<1:58:08,  1.61it/s]
 83%|████████▎ | 54103/65536 [9:19:28<1:54:58,  1.66it/s]
 83%|████████▎ | 54104/65536 [9:19:29<1:55:42,  1.65it/s]
 83%|████████▎ | 54105/65536 [9:19:29<1:54:25,  1.67it/s]
 83%|████████▎ | 54106/65536 [9:19:30<1:55:29,  1.65it/s]
 83%|████████▎ | 54107/65536 [9:19:31<1:56:09,  1.64it/s]
 83%|████████▎ | 54108/65536 [9:19:31<1:59:42,  1.59it/s]
 83%|████████▎ | 54109/65536 [9:19:32<2:03:12,  1.55it/s]
 83%|████████▎ | 54110/65536 [9:19:33<2:00:23,  1.58it/s]
 83%|████████▎ | 54111/65536 [9:19:33<1:59:01,  1.60it/s]
 83%|████████▎ | 54112/65536 [9:19:34<1:59:34,  1.59it/s]
 83%|████████▎ | 54113/65536 [9:19:34<1:59:58,  1.59it/s]
 83%|████████▎ | 54114/65536 [9:19:35<1:58:04,  1.61it/s]
 83%|████████▎ | 54115/65536 [9:19:36<1:57:48,  1.62it/s]
 83%|████████▎ | 54116/65536 [9:19:36<1:57:32,  1.62it/s]
 83%|████████▎ | 54117/65536 [9:19:37<2:01:15,  1.57it/s]
 83%|████████▎ | 54118/65536 [9:19:38<2:02:53,  1.55it/s]
 83%|████████▎ | 54119/65536 [9:19:38<2:00:15,  1.58it/s]
 83%|████████▎ | 54120/65536 [9:19:39<1:57:43,  1.62it/s]
                                                         
{'loss': 1.5224, 'learning_rate': 2.572548059262887e-07, 'epoch': 3340.74}
+
 83%|████████▎ | 54120/65536 [9:19:39<1:57:43,  1.62it/s]
 83%|████████▎ | 54121/65536 [9:19:39<1:56:34,  1.63it/s]
 83%|████████▎ | 54122/65536 [9:19:40<1:54:11,  1.67it/s]
 83%|████████▎ | 54123/65536 [9:19:41<1:56:20,  1.64it/s]
 83%|████████▎ | 54124/65536 [9:19:41<1:58:40,  1.60it/s]
 83%|████████▎ | 54125/65536 [9:19:42<2:03:04,  1.55it/s]
 83%|████████▎ | 54126/65536 [9:19:43<1:59:37,  1.59it/s]
 83%|████████▎ | 54127/65536 [9:19:43<1:59:03,  1.60it/s]
 83%|████████▎ | 54128/65536 [9:19:44<1:59:01,  1.60it/s]
 83%|████████▎ | 54129/65536 [9:19:44<1:59:14,  1.59it/s]
 83%|████████▎ | 54130/65536 [9:19:45<1:56:59,  1.62it/s]
 83%|████████▎ | 54131/65536 [9:19:46<1:57:27,  1.62it/s]
 83%|████████▎ | 54132/65536 [9:19:46<1:58:20,  1.61it/s]
 83%|████████▎ | 54133/65536 [9:19:47<1:55:17,  1.65it/s]
 83%|████████▎ | 54134/65536 [9:19:47<1:55:29,  1.65it/s]
 83%|████████▎ | 54135/65536 [9:19:48<1:54:47,  1.66it/s]
 83%|████████▎ | 54136/65536 [9:19:49<1:59:08,  1.59it/s]
 83%|████████▎ | 54137/65536 [9:19:49<1:57:27,  1.62it/s]
 83%|████████▎ | 54138/65536 [9:19:50<1:58:11,  1.61it/s]
 83%|████████▎ | 54139/65536 [9:19:51<1:57:22,  1.62it/s]
 83%|████████▎ | 54140/65536 [9:19:51<1:56:53,  1.62it/s]
                                                         
{'loss': 1.5441, 'learning_rate': 2.569793069670626e-07, 'epoch': 3341.98}
+
 83%|████████▎ | 54140/65536 [9:19:51<1:56:53,  1.62it/s]
 83%|████████▎ | 54141/65536 [9:19:52<2:02:06,  1.56it/s]
 83%|████████▎ | 54142/65536 [9:19:52<1:58:55,  1.60it/s]
 83%|████████▎ | 54143/65536 [9:19:53<1:59:55,  1.58it/s]
 83%|████████▎ | 54144/65536 [9:19:54<1:57:26,  1.62it/s]
 83%|████████▎ | 54145/65536 [9:19:54<1:57:03,  1.62it/s]
 83%|████████▎ | 54146/65536 [9:19:55<1:58:06,  1.61it/s]
 83%|████████▎ | 54147/65536 [9:19:56<1:59:12,  1.59it/s]
 83%|████████▎ | 54148/65536 [9:19:56<1:57:42,  1.61it/s]
 83%|████████▎ | 54149/65536 [9:19:57<1:55:26,  1.64it/s]
 83%|████████▎ | 54150/65536 [9:19:57<1:56:21,  1.63it/s]
 83%|████████▎ | 54151/65536 [9:19:58<1:55:23,  1.64it/s]
 83%|████████▎ | 54152/65536 [9:19:59<1:55:14,  1.65it/s]
 83%|████████▎ | 54153/65536 [9:19:59<1:55:05,  1.65it/s]
 83%|████████▎ | 54154/65536 [9:20:00<1:58:18,  1.60it/s]
 83%|████████▎ | 54155/65536 [9:20:01<1:58:15,  1.60it/s]
 83%|████████▎ | 54156/65536 [9:20:01<2:00:15,  1.58it/s]
 83%|████████▎ | 54157/65536 [9:20:02<2:01:35,  1.56it/s]
 83%|████████▎ | 54158/65536 [9:20:02<2:00:43,  1.57it/s]
 83%|████████▎ | 54159/65536 [9:20:03<2:01:28,  1.56it/s]
 83%|████████▎ | 54160/65536 [9:20:04<2:02:15,  1.55it/s]
                                                         
{'loss': 1.5758, 'learning_rate': 2.5670380800783644e-07, 'epoch': 3343.21}
+
 83%|████████▎ | 54160/65536 [9:20:04<2:02:15,  1.55it/s]
 83%|████████▎ | 54161/65536 [9:20:04<2:02:44,  1.54it/s]
 83%|████████▎ | 54162/65536 [9:20:05<1:59:05,  1.59it/s]
 83%|████████▎ | 54163/65536 [9:20:06<1:59:20,  1.59it/s]
 83%|████████▎ | 54164/65536 [9:20:06<1:57:40,  1.61it/s]
 83%|████████▎ | 54165/65536 [9:20:07<1:55:52,  1.64it/s]
 83%|████████▎ | 54166/65536 [9:20:07<1:55:00,  1.65it/s]
 83%|█��██████▎ | 54167/65536 [9:20:08<1:55:36,  1.64it/s]
 83%|████████▎ | 54168/65536 [9:20:09<1:56:39,  1.62it/s]
 83%|████████▎ | 54169/65536 [9:20:09<1:59:37,  1.58it/s]
 83%|████████▎ | 54170/65536 [9:20:10<1:59:01,  1.59it/s]
 83%|████████▎ | 54171/65536 [9:20:11<1:56:59,  1.62it/s]
 83%|████████▎ | 54172/65536 [9:20:11<1:57:00,  1.62it/s]
 83%|████████▎ | 54173/65536 [9:20:12<1:59:36,  1.58it/s]
 83%|████████▎ | 54174/65536 [9:20:12<1:58:00,  1.60it/s]
 83%|████████▎ | 54175/65536 [9:20:13<1:58:14,  1.60it/s]
 83%|████████▎ | 54176/65536 [9:20:14<1:58:11,  1.60it/s]
 83%|████████▎ | 54177/65536 [9:20:14<1:58:52,  1.59it/s]
 83%|████████▎ | 54178/65536 [9:20:15<1:58:59,  1.59it/s]
 83%|████████▎ | 54179/65536 [9:20:16<1:57:33,  1.61it/s]
 83%|████████▎ | 54180/65536 [9:20:16<1:56:42,  1.62it/s]
                                                         
{'loss': 1.5885, 'learning_rate': 2.564283090486102e-07, 'epoch': 3344.44}
+
 83%|████████▎ | 54180/65536 [9:20:16<1:56:42,  1.62it/s]
 83%|████████▎ | 54181/65536 [9:20:17<1:56:53,  1.62it/s]
 83%|████████▎ | 54182/65536 [9:20:17<1:59:04,  1.59it/s]
 83%|████████▎ | 54183/65536 [9:20:18<1:58:46,  1.59it/s]
 83%|████████▎ | 54184/65536 [9:20:19<1:56:27,  1.62it/s]
 83%|████████▎ | 54185/65536 [9:20:19<1:54:42,  1.65it/s]
 83%|████████▎ | 54186/65536 [9:20:20<1:54:00,  1.66it/s]
 83%|████████▎ | 54187/65536 [9:20:20<1:53:29,  1.67it/s]
 83%|████████▎ | 54188/65536 [9:20:21<1:57:16,  1.61it/s]
 83%|████████▎ | 54189/65536 [9:20:22<1:57:28,  1.61it/s]
 83%|████████▎ | 54190/65536 [9:20:22<1:59:42,  1.58it/s]
 83%|████████▎ | 54191/65536 [9:20:23<1:57:25,  1.61it/s]
 83%|████████▎ | 54192/65536 [9:20:24<1:58:27,  1.60it/s]
 83%|████████▎ | 54193/65536 [9:20:24<1:57:47,  1.60it/s]
 83%|████████▎ | 54194/65536 [9:20:25<1:56:03,  1.63it/s]
 83%|████████▎ | 54195/65536 [9:20:25<1:55:38,  1.63it/s]
 83%|████████▎ | 54196/65536 [9:20:26<1:57:36,  1.61it/s]
 83%|████████▎ | 54197/65536 [9:20:27<1:59:02,  1.59it/s]
 83%|████████▎ | 54198/65536 [9:20:27<1:59:13,  1.59it/s]
 83%|████████▎ | 54199/65536 [9:20:28<1:57:50,  1.60it/s]
 83%|████████▎ | 54200/65536 [9:20:29<1:57:03,  1.61it/s]
                                                         
{'loss': 1.5522, 'learning_rate': 2.5615281008938405e-07, 'epoch': 3345.68}
+
 83%|████████▎ | 54200/65536 [9:20:29<1:57:03,  1.61it/s]
 83%|████████▎ | 54201/65536 [9:20:29<2:00:50,  1.56it/s]
 83%|████████▎ | 54202/65536 [9:20:30<1:59:06,  1.59it/s]
 83%|████████▎ | 54203/65536 [9:20:30<1:57:41,  1.61it/s]
 83%|████████▎ | 54204/65536 [9:20:31<1:55:22,  1.64it/s]
 83%|████████▎ | 54205/65536 [9:20:32<1:55:23,  1.64it/s]
 83%|████████▎ | 54206/65536 [9:20:32<1:59:30,  1.58it/s]
 83%|████████▎ | 54207/65536 [9:20:33<1:56:55,  1.61it/s]
 83%|████████▎ | 54208/65536 [9:20:33<1:54:30,  1.65it/s]
 83%|████████▎ | 54209/65536 [9:20:34<1:53:55,  1.66it/s]
 83%|████████▎ | 54210/65536 [9:20:35<1:52:18,  1.68it/s]
 83%|████████▎ | 54211/65536 [9:20:35<1:54:55,  1.64it/s]
 83%|████████▎ | 54212/65536 [9:20:36<1:58:18,  1.60it/s]
 83%|████████▎ | 54213/65536 [9:20:37<1:57:39,  1.60it/s]
 83%|████████▎ | 54214/65536 [9:20:37<1:55:36,  1.63it/s]
 83%|████████▎ | 54215/65536 [9:20:38<1:57:57,  1.60it/s]
 83%|████████▎ | 54216/65536 [9:20:38<1:57:15,  1.61it/s]
 83%|████████▎ | 54217/65536 [9:20:39<1:56:46,  1.62it/s]
 83%|████████▎ | 54218/65536 [9:20:40<1:59:28,  1.58it/s]
 83%|████████▎ | 54219/65536 [9:20:40<1:57:47,  1.60it/s]
 83%|████████▎ | 54220/65536 [9:20:41<2:02:13,  1.54it/s]
                                                         
{'loss': 1.5588, 'learning_rate': 2.558773111301579e-07, 'epoch': 3346.91}
+
 83%|████████▎ | 54220/65536 [9:20:41<2:02:13,  1.54it/s]
 83%|████████▎ | 54221/65536 [9:20:42<2:01:25,  1.55it/s]
 83%|████████▎ | 54222/65536 [9:20:42<2:01:20,  1.55it/s]
 83%|████████▎ | 54223/65536 [9:20:43<1:57:19,  1.61it/s]
 83%|████████▎ | 54224/65536 [9:20:44<1:58:28,  1.59it/s]
 83%|████████▎ | 54225/65536 [9:20:44<1:58:09,  1.60it/s]
 83%|██���█████▎ | 54226/65536 [9:20:45<1:56:21,  1.62it/s]
 83%|████████▎ | 54227/65536 [9:20:45<1:57:27,  1.60it/s]
 83%|████████▎ | 54228/65536 [9:20:46<1:57:36,  1.60it/s]
 83%|████████▎ | 54229/65536 [9:20:47<1:56:36,  1.62it/s]
 83%|████████▎ | 54230/65536 [9:20:47<1:54:26,  1.65it/s]
 83%|████████▎ | 54231/65536 [9:20:48<1:57:22,  1.61it/s]
 83%|████████▎ | 54232/65536 [9:20:48<1:55:33,  1.63it/s]
 83%|████████▎ | 54233/65536 [9:20:49<1:57:07,  1.61it/s]
 83%|████████▎ | 54234/65536 [9:20:50<2:01:44,  1.55it/s]
 83%|████████▎ | 54235/65536 [9:20:50<1:56:58,  1.61it/s]
 83%|████████▎ | 54236/65536 [9:20:51<1:56:16,  1.62it/s]
 83%|████████▎ | 54237/65536 [9:20:52<1:54:09,  1.65it/s]
 83%|████████▎ | 54238/65536 [9:20:52<2:00:27,  1.56it/s]
 83%|████████▎ | 54239/65536 [9:20:53<1:57:59,  1.60it/s]
 83%|████████▎ | 54240/65536 [9:20:53<1:55:23,  1.63it/s]
                                                         
{'loss': 1.6006, 'learning_rate': 2.5560181217093177e-07, 'epoch': 3348.15}
+
 83%|████████▎ | 54240/65536 [9:20:53<1:55:23,  1.63it/s]
 83%|████████▎ | 54241/65536 [9:20:54<1:55:26,  1.63it/s]
 83%|████████▎ | 54242/65536 [9:20:55<1:54:25,  1.64it/s]
 83%|████████▎ | 54243/65536 [9:20:55<1:54:03,  1.65it/s]
 83%|████████▎ | 54244/65536 [9:20:56<1:57:27,  1.60it/s]
 83%|████████▎ | 54245/65536 [9:20:57<1:57:26,  1.60it/s]
 83%|████████▎ | 54246/65536 [9:20:57<1:55:54,  1.62it/s]
 83%|████████▎ | 54247/65536 [9:20:58<1:54:39,  1.64it/s]
 83%|████████▎ | 54248/65536 [9:20:58<1:55:32,  1.63it/s]
 83%|████████▎ | 54249/65536 [9:20:59<1:54:04,  1.65it/s]
 83%|████████▎ | 54250/65536 [9:21:00<1:55:54,  1.62it/s]
 83%|████████▎ | 54251/65536 [9:21:00<1:54:19,  1.65it/s]
 83%|████████▎ | 54252/65536 [9:21:01<1:54:53,  1.64it/s]
 83%|████████▎ | 54253/65536 [9:21:01<1:55:57,  1.62it/s]
 83%|████████▎ | 54254/65536 [9:21:02<1:59:38,  1.57it/s]
 83%|████████▎ | 54255/65536 [9:21:03<2:00:45,  1.56it/s]
 83%|████████▎ | 54256/65536 [9:21:03<1:58:51,  1.58it/s]
 83%|████████▎ | 54257/65536 [9:21:04<1:59:50,  1.57it/s]
 83%|████████▎ | 54258/65536 [9:21:05<1:57:04,  1.61it/s]
 83%|████████▎ | 54259/65536 [9:21:05<1:55:20,  1.63it/s]
 83%|████████▎ | 54260/65536 [9:21:06<1:54:16,  1.64it/s]
                                                         
{'loss': 1.5793, 'learning_rate': 2.553263132117057e-07, 'epoch': 3349.38}
+
 83%|████████▎ | 54260/65536 [9:21:06<1:54:16,  1.64it/s]
 83%|████████▎ | 54261/65536 [9:21:06<1:54:31,  1.64it/s]
 83%|████████▎ | 54262/65536 [9:21:07<1:53:36,  1.65it/s]
 83%|████████▎ | 54263/65536 [9:21:08<1:52:23,  1.67it/s]
 83%|████████▎ | 54264/65536 [9:21:08<1:54:57,  1.63it/s]
 83%|████████▎ | 54265/65536 [9:21:09<1:55:20,  1.63it/s]
 83%|████████▎ | 54266/65536 [9:21:09<1:56:51,  1.61it/s]
 83%|████████▎ | 54267/65536 [9:21:10<1:56:31,  1.61it/s]
 83%|████████▎ | 54268/65536 [9:21:11<1:57:10,  1.60it/s]
 83%|████████▎ | 54269/65536 [9:21:11<2:03:20,  1.52it/s]
 83%|████████▎ | 54270/65536 [9:21:12<2:02:17,  1.54it/s]
 83%|████████▎ | 54271/65536 [9:21:13<2:04:50,  1.50it/s]
 83%|████████▎ | 54272/65536 [9:21:13<2:04:15,  1.51it/s]
 83%|████████▎ | 54273/65536 [9:21:14<2:04:22,  1.51it/s]
 83%|████████▎ | 54274/65536 [9:21:15<2:05:34,  1.49it/s]
 83%|████████▎ | 54275/65536 [9:21:15<2:00:33,  1.56it/s]
 83%|████████▎ | 54276/65536 [9:21:16<2:02:59,  1.53it/s]
 83%|████████▎ | 54277/65536 [9:21:17<2:01:23,  1.55it/s]
 83%|████████▎ | 54278/65536 [9:21:17<1:56:35,  1.61it/s]
 83%|████████▎ | 54279/65536 [9:21:18<1:55:44,  1.62it/s]
 83%|████████▎ | 54280/65536 [9:21:18<1:54:32,  1.64it/s]
                                                         
{'loss': 1.5483, 'learning_rate': 2.5505081425247954e-07, 'epoch': 3350.62}
+
 83%|████████▎ | 54280/65536 [9:21:18<1:54:32,  1.64it/s]
 83%|████████▎ | 54281/65536 [9:21:19<1:52:16,  1.67it/s]
 83%|████████▎ | 54282/65536 [9:21:20<1:50:50,  1.69it/s]
 83%|████████▎ | 54283/65536 [9:21:20<1:51:48,  1.68it/s]
 83%|████████▎ | 54284/65536 [9:21:21<1:54:19,  1.64it/s]
 83%|██��█████▎ | 54285/65536 [9:21:21<1:53:18,  1.65it/s]
 83%|████████▎ | 54286/65536 [9:21:22<1:53:33,  1.65it/s]
 83%|████████▎ | 54287/65536 [9:21:23<1:58:00,  1.59it/s]
 83%|████████▎ | 54288/65536 [9:21:23<1:57:20,  1.60it/s]
 83%|████████▎ | 54289/65536 [9:21:24<1:55:24,  1.62it/s]
 83%|████████▎ | 54290/65536 [9:21:25<1:56:09,  1.61it/s]
 83%|████████▎ | 54291/65536 [9:21:25<1:57:07,  1.60it/s]
 83%|████████▎ | 54292/65536 [9:21:26<1:56:45,  1.61it/s]
 83%|████████▎ | 54293/65536 [9:21:26<1:55:48,  1.62it/s]
 83%|████████▎ | 54294/65536 [9:21:27<1:57:49,  1.59it/s]
 83%|████████▎ | 54295/65536 [9:21:28<1:56:08,  1.61it/s]
 83%|████████▎ | 54296/65536 [9:21:28<1:58:16,  1.58it/s]
 83%|████████▎ | 54297/65536 [9:21:29<1:57:43,  1.59it/s]
 83%|████████▎ | 54298/65536 [9:21:30<1:56:12,  1.61it/s]
 83%|████████▎ | 54299/65536 [9:21:30<1:54:52,  1.63it/s]
 83%|████████▎ | 54300/65536 [9:21:31<1:52:12,  1.67it/s]
                                                         
{'loss': 1.5487, 'learning_rate': 2.547753152932534e-07, 'epoch': 3351.85}
+
 83%|████████▎ | 54300/65536 [9:21:31<1:52:12,  1.67it/s]
 83%|████████▎ | 54301/65536 [9:21:31<1:51:12,  1.68it/s]
 83%|████████▎ | 54302/65536 [9:21:32<1:53:10,  1.65it/s]
 83%|████████▎ | 54303/65536 [9:21:33<1:56:54,  1.60it/s]
 83%|████████▎ | 54304/65536 [9:21:33<1:57:24,  1.59it/s]
 83%|████████▎ | 54305/65536 [9:21:34<1:54:06,  1.64it/s]
 83%|████████▎ | 54306/65536 [9:21:34<1:54:15,  1.64it/s]
 83%|████████▎ | 54307/65536 [9:21:35<1:53:11,  1.65it/s]
 83%|████████▎ | 54308/65536 [9:21:36<1:54:06,  1.64it/s]
 83%|████████▎ | 54309/65536 [9:21:36<1:52:58,  1.66it/s]
 83%|████████▎ | 54310/65536 [9:21:37<1:55:52,  1.61it/s]
 83%|████████▎ | 54311/65536 [9:21:37<1:53:32,  1.65it/s]
 83%|████████▎ | 54312/65536 [9:21:38<1:54:11,  1.64it/s]
 83%|████████▎ | 54313/65536 [9:21:39<1:53:02,  1.65it/s]
 83%|████████▎ | 54314/65536 [9:21:39<1:51:53,  1.67it/s]
 83%|████████▎ | 54315/65536 [9:21:40<1:54:38,  1.63it/s]
 83%|████████▎ | 54316/65536 [9:21:41<1:55:07,  1.62it/s]
 83%|████████▎ | 54317/65536 [9:21:41<1:54:36,  1.63it/s]
 83%|████████▎ | 54318/65536 [9:21:42<1:55:38,  1.62it/s]
 83%|████████▎ | 54319/65536 [9:21:43<2:03:42,  1.51it/s]
 83%|████████▎ | 54320/65536 [9:21:43<2:03:52,  1.51it/s]
                                                         
{'loss': 1.576, 'learning_rate': 2.5449981633402715e-07, 'epoch': 3353.09}
+
 83%|████████▎ | 54320/65536 [9:21:43<2:03:52,  1.51it/s]
 83%|████████▎ | 54321/65536 [9:21:44<2:04:34,  1.50it/s]
 83%|████████▎ | 54322/65536 [9:21:44<2:01:19,  1.54it/s]
 83%|████████▎ | 54323/65536 [9:21:45<1:57:39,  1.59it/s]
 83%|████████▎ | 54324/65536 [9:21:46<1:57:14,  1.59it/s]
 83%|████████▎ | 54325/65536 [9:21:46<1:57:01,  1.60it/s]
 83%|████████▎ | 54326/65536 [9:21:47<1:59:54,  1.56it/s]
 83%|████████▎ | 54327/65536 [9:21:48<1:57:40,  1.59it/s]
 83%|████████▎ | 54328/65536 [9:21:48<1:59:07,  1.57it/s]
 83%|████████▎ | 54329/65536 [9:21:49<1:57:45,  1.59it/s]
 83%|████████▎ | 54330/65536 [9:21:49<1:56:58,  1.60it/s]
 83%|████████▎ | 54331/65536 [9:21:50<1:55:16,  1.62it/s]
 83%|████████▎ | 54332/65536 [9:21:51<1:56:02,  1.61it/s]
 83%|████████▎ | 54333/65536 [9:21:51<1:54:24,  1.63it/s]
 83%|████████▎ | 54334/65536 [9:21:52<1:53:54,  1.64it/s]
 83%|████████▎ | 54335/65536 [9:21:53<1:56:57,  1.60it/s]
 83%|████████▎ | 54336/65536 [9:21:53<1:56:29,  1.60it/s]
 83%|████████▎ | 54337/65536 [9:21:54<1:54:05,  1.64it/s]
 83%|████████▎ | 54338/65536 [9:21:54<1:53:56,  1.64it/s]
 83%|████████▎ | 54339/65536 [9:21:55<1:51:14,  1.68it/s]
 83%|████████▎ | 54340/65536 [9:21:56<1:54:02,  1.64it/s]
                                                         
{'loss': 1.5713, 'learning_rate': 2.54224317374801e-07, 'epoch': 3354.32}
+
 83%|████████▎ | 54340/65536 [9:21:56<1:54:02,  1.64it/s]
 83%|████████▎ | 54341/65536 [9:21:56<1:53:41,  1.64it/s]
 83%|████████▎ | 54342/65536 [9:21:57<1:52:02,  1.67it/s]
 83%|████████▎ | 54343/65536 [9:21:57<1:50:21,  1.69it/s]
 83%|████████▎ | 54344/65536 [9:21:58<1:56:36,  1.60it/s]
 83%|████████▎ | 54345/65536 [9:21:59<1:55:55,  1.61it/s]
 83%|████████▎ | 54346/65536 [9:21:59<1:53:54,  1.64it/s]
 83%|████████▎ | 54347/65536 [9:22:00<1:54:00,  1.64it/s]
 83%|████████▎ | 54348/65536 [9:22:00<1:53:50,  1.64it/s]
 83%|████████▎ | 54349/65536 [9:22:01<1:53:28,  1.64it/s]
 83%|████████▎ | 54350/65536 [9:22:02<1:54:34,  1.63it/s]
 83%|████████▎ | 54351/65536 [9:22:02<1:55:36,  1.61it/s]
 83%|████████▎ | 54352/65536 [9:22:03<1:59:22,  1.56it/s]
 83%|████████▎ | 54353/65536 [9:22:04<2:03:01,  1.52it/s]
 83%|████████▎ | 54354/65536 [9:22:04<2:03:32,  1.51it/s]
 83%|████████▎ | 54355/65536 [9:22:05<2:03:44,  1.51it/s]
 83%|████████▎ | 54356/65536 [9:22:06<2:00:09,  1.55it/s]
 83%|████████▎ | 54357/65536 [9:22:06<1:56:54,  1.59it/s]
 83%|████████▎ | 54358/65536 [9:22:07<1:54:11,  1.63it/s]
 83%|████████▎ | 54359/65536 [9:22:07<1:51:39,  1.67it/s]
 83%|████████▎ | 54360/65536 [9:22:08<1:51:59,  1.66it/s]
                                                         
{'loss': 1.5772, 'learning_rate': 2.5394881841557487e-07, 'epoch': 3355.56}
+
 83%|████████▎ | 54360/65536 [9:22:08<1:51:59,  1.66it/s]
 83%|████████▎ | 54361/65536 [9:22:09<1:51:10,  1.68it/s]
 83%|████████▎ | 54362/65536 [9:22:09<1:53:17,  1.64it/s]
 83%|████████▎ | 54363/65536 [9:22:10<1:51:49,  1.67it/s]
 83%|████████▎ | 54364/65536 [9:22:10<1:50:27,  1.69it/s]
 83%|████████▎ | 54365/65536 [9:22:11<1:51:41,  1.67it/s]
 83%|████████▎ | 54366/65536 [9:22:12<1:51:55,  1.66it/s]
 83%|████████▎ | 54367/65536 [9:22:12<1:53:44,  1.64it/s]
 83%|████████▎ | 54368/65536 [9:22:13<1:55:45,  1.61it/s]
 83%|████████▎ | 54369/65536 [9:22:14<1:57:25,  1.59it/s]
 83%|████████▎ | 54370/65536 [9:22:14<1:59:09,  1.56it/s]
 83%|████████▎ | 54371/65536 [9:22:15<1:55:42,  1.61it/s]
 83%|████████▎ | 54372/65536 [9:22:15<1:56:27,  1.60it/s]
 83%|████████▎ | 54373/65536 [9:22:16<1:55:52,  1.61it/s]
 83%|████████▎ | 54374/65536 [9:22:17<1:57:13,  1.59it/s]
 83%|████████▎ | 54375/65536 [9:22:17<1:55:03,  1.62it/s]
 83%|████████▎ | 54376/65536 [9:22:18<1:55:15,  1.61it/s]
 83%|████████▎ | 54377/65536 [9:22:19<1:57:12,  1.59it/s]
 83%|████████▎ | 54378/65536 [9:22:19<1:55:18,  1.61it/s]
 83%|████████▎ | 54379/65536 [9:22:20<1:54:03,  1.63it/s]
 83%|████████▎ | 54380/65536 [9:22:20<1:54:45,  1.62it/s]
                                                         
{'loss': 1.5755, 'learning_rate': 2.5367331945634873e-07, 'epoch': 3356.79}
+
 83%|████████▎ | 54380/65536 [9:22:20<1:54:45,  1.62it/s]
 83%|████████▎ | 54381/65536 [9:22:21<1:55:17,  1.61it/s]
 83%|████████▎ | 54382/65536 [9:22:22<1:55:14,  1.61it/s]
 83%|████████▎ | 54383/65536 [9:22:22<1:52:45,  1.65it/s]
 83%|████████▎ | 54384/65536 [9:22:23<1:57:05,  1.59it/s]
 83%|████████▎ | 54385/65536 [9:22:23<1:56:09,  1.60it/s]
 83%|████████▎ | 54386/65536 [9:22:24<1:54:47,  1.62it/s]
 83%|████████▎ | 54387/65536 [9:22:25<1:55:06,  1.61it/s]
 83%|████████▎ | 54388/65536 [9:22:25<1:52:21,  1.65it/s]
 83%|████████▎ | 54389/65536 [9:22:26<1:52:59,  1.64it/s]
 83%|████████▎ | 54390/65536 [9:22:27<1:54:05,  1.63it/s]
 83%|████████▎ | 54391/65536 [9:22:27<1:54:53,  1.62it/s]
 83%|████████▎ | 54392/65536 [9:22:28<1:53:12,  1.64it/s]
 83%|████████▎ | 54393/65536 [9:22:28<1:56:29,  1.59it/s]
 83%|████████▎ | 54394/65536 [9:22:29<1:57:20,  1.58it/s]
 83%|████████▎ | 54395/65536 [9:22:30<1:58:07,  1.57it/s]
 83%|████████▎ | 54396/65536 [9:22:30<1:58:35,  1.57it/s]
 83%|████████▎ | 54397/65536 [9:22:31<1:54:43,  1.62it/s]
 83%|████████▎ | 54398/65536 [9:22:31<1:51:47,  1.66it/s]
 83%|████████▎ | 54399/65536 [9:22:32<1:52:55,  1.64it/s]
 83%|████████▎ | 54400/65536 [9:22:33<1:54:43,  1.62it/s]
                                                         
{'loss': 1.59, 'learning_rate': 2.533978204971226e-07, 'epoch': 3358.02}
+
 83%|████████▎ | 54400/65536 [9:22:33<1:54:43,  1.62it/s]
 83%|████████▎ | 54401/65536 [9:22:33<1:55:11,  1.61it/s]
 83%|████████▎ | 54402/65536 [9:22:34<1:54:16,  1.62it/s]
 83%|████████▎ | 54403/65536 [9:22:35<1:51:57,  1.66it/s]
 83%|████████▎ | 54404/65536 [9:22:35<1:56:38,  1.59it/s]
 83%|████████▎ | 54405/65536 [9:22:36<1:57:36,  1.58it/s]
 83%|████████▎ | 54406/65536 [9:22:37<1:58:03,  1.57it/s]
 83%|████████▎ | 54407/65536 [9:22:37<1:55:25,  1.61it/s]
 83%|████████▎ | 54408/65536 [9:22:38<1:55:00,  1.61it/s]
 83%|████████▎ | 54409/65536 [9:22:38<1:57:13,  1.58it/s]
 83%|████████▎ | 54410/65536 [9:22:39<1:54:41,  1.62it/s]
 83%|████████▎ | 54411/65536 [9:22:40<1:52:59,  1.64it/s]
 83%|████████▎ | 54412/65536 [9:22:40<1:52:28,  1.65it/s]
 83%|████████▎ | 54413/65536 [9:22:41<1:53:52,  1.63it/s]
 83%|████████▎ | 54414/65536 [9:22:41<1:55:27,  1.61it/s]
 83%|████████▎ | 54415/65536 [9:22:42<1:54:41,  1.62it/s]
 83%|████████▎ | 54416/65536 [9:22:43<1:57:59,  1.57it/s]
 83%|████████▎ | 54417/65536 [9:22:43<1:56:21,  1.59it/s]
 83%|████████▎ | 54418/65536 [9:22:44<1:55:03,  1.61it/s]
 83%|████████▎ | 54419/65536 [9:22:45<1:57:39,  1.57it/s]
 83%|████████▎ | 54420/65536 [9:22:45<1:56:46,  1.59it/s]
                                                         
{'loss': 1.5586, 'learning_rate': 2.531223215378964e-07, 'epoch': 3359.26}
+
 83%|████████▎ | 54420/65536 [9:22:45<1:56:46,  1.59it/s]
 83%|████████▎ | 54421/65536 [9:22:46<1:57:41,  1.57it/s]
 83%|████████▎ | 54422/65536 [9:22:47<1:58:20,  1.57it/s]
 83%|████████▎ | 54423/65536 [9:22:47<1:56:11,  1.59it/s]
 83%|████████▎ | 54424/65536 [9:22:48<1:57:09,  1.58it/s]
 83%|████████▎ | 54425/65536 [9:22:48<1:54:33,  1.62it/s]
 83%|████████▎ | 54426/65536 [9:22:49<1:52:29,  1.65it/s]
 83%|████████▎ | 54427/65536 [9:22:50<1:54:19,  1.62it/s]
 83%|████████▎ | 54428/65536 [9:22:50<1:52:59,  1.64it/s]
 83%|████████▎ | 54429/65536 [9:22:51<1:52:50,  1.64it/s]
 83%|████████▎ | 54430/65536 [9:22:51<1:53:09,  1.64it/s]
 83%|████████▎ | 54431/65536 [9:22:52<1:55:09,  1.61it/s]
 83%|████████▎ | 54432/65536 [9:22:53<1:55:38,  1.60it/s]
 83%|████████▎ | 54433/65536 [9:22:53<1:55:59,  1.60it/s]
 83%|████████▎ | 54434/65536 [9:22:54<1:57:41,  1.57it/s]
 83%|████████▎ | 54435/65536 [9:22:55<1:55:46,  1.60it/s]
 83%|████████▎ | 54436/65536 [9:22:55<1:57:42,  1.57it/s]
 83%|████████▎ | 54437/65536 [9:22:56<1:58:42,  1.56it/s]
 83%|████████▎ | 54438/65536 [9:22:57<1:58:19,  1.56it/s]
 83%|████████▎ | 54439/65536 [9:22:57<1:55:29,  1.60it/s]
 83%|████████▎ | 54440/65536 [9:22:58<1:53:24,  1.63it/s]
                                                         
{'loss': 1.5461, 'learning_rate': 2.528468225786703e-07, 'epoch': 3360.49}
+
 83%|████████▎ | 54440/65536 [9:22:58<1:53:24,  1.63it/s]
 83%|████████▎ | 54441/65536 [9:22:58<1:52:59,  1.64it/s]
 83%|████████▎ | 54442/65536 [9:22:59<1:52:56,  1.64it/s]
 83%|████████▎ | 54443/65536 [9:23:00<1:52:36,  1.64it/s]
 83%|████████▎ | 54444/65536 [9:23:00<1:53:44,  1.63it/s]
 83%|████████▎ | 54445/65536 [9:23:01<1:54:37,  1.61it/s]
 83%|████████▎ | 54446/65536 [9:23:01<1:55:35,  1.60it/s]
 83%|████████▎ | 54447/65536 [9:23:02<1:56:07,  1.59it/s]
 83%|████████▎ | 54448/65536 [9:23:03<1:54:04,  1.62it/s]
 83%|████████▎ | 54449/65536 [9:23:03<1:56:54,  1.58it/s]
 83%|████████▎ | 54450/65536 [9:23:04<1:53:55,  1.62it/s]
 83%|████████▎ | 54451/65536 [9:23:05<1:57:11,  1.58it/s]
 83%|████████▎ | 54452/65536 [9:23:05<1:55:55,  1.59it/s]
 83%|████████▎ | 54453/65536 [9:23:06<1:53:41,  1.62it/s]
 83%|████████▎ | 54454/65536 [9:23:06<1:56:45,  1.58it/s]
 83%|████████▎ | 54455/65536 [9:23:07<1:57:01,  1.58it/s]
 83%|████████▎ | 54456/65536 [9:23:08<1:54:38,  1.61it/s]
 83%|████████▎ | 54457/65536 [9:23:08<1:53:15,  1.63it/s]
 83%|████████▎ | 54458/65536 [9:23:09<1:52:26,  1.64it/s]
 83%|████████▎ | 54459/65536 [9:23:09<1:53:09,  1.63it/s]
 83%|████████▎ | 54460/65536 [9:23:10<1:52:17,  1.64it/s]
                                                         
{'loss': 1.5718, 'learning_rate': 2.5257132361944406e-07, 'epoch': 3361.73}
+
 83%|████████▎ | 54460/65536 [9:23:10<1:52:17,  1.64it/s]
 83%|████████▎ | 54461/65536 [9:23:11<1:51:58,  1.65it/s]
 83%|█████��██▎ | 54462/65536 [9:23:11<1:56:04,  1.59it/s]
 83%|████████▎ | 54463/65536 [9:23:12<1:54:23,  1.61it/s]
 83%|████████▎ | 54464/65536 [9:23:13<1:53:02,  1.63it/s]
 83%|████████▎ | 54465/65536 [9:23:13<1:55:47,  1.59it/s]
 83%|████████▎ | 54466/65536 [9:23:14<1:53:26,  1.63it/s]
 83%|████████▎ | 54467/65536 [9:23:14<1:55:09,  1.60it/s]
 83%|████████▎ | 54468/65536 [9:23:15<1:53:55,  1.62it/s]
 83%|████████▎ | 54469/65536 [9:23:16<1:57:28,  1.57it/s]
 83%|████████▎ | 54470/65536 [9:23:16<1:57:46,  1.57it/s]
 83%|████████▎ | 54471/65536 [9:23:17<1:59:05,  1.55it/s]
 83%|████████▎ | 54472/65536 [9:23:18<1:56:50,  1.58it/s]
 83%|████████▎ | 54473/65536 [9:23:18<1:55:09,  1.60it/s]
 83%|████████▎ | 54474/65536 [9:23:19<1:55:32,  1.60it/s]
 83%|████████▎ | 54475/65536 [9:23:19<1:53:28,  1.62it/s]
 83%|████████▎ | 54476/65536 [9:23:20<1:57:08,  1.57it/s]
 83%|████████▎ | 54477/65536 [9:23:21<1:56:30,  1.58it/s]
 83%|████████▎ | 54478/65536 [9:23:21<1:57:04,  1.57it/s]
 83%|████████▎ | 54479/65536 [9:23:22<1:55:49,  1.59it/s]
 83%|████████▎ | 54480/65536 [9:23:23<1:53:16,  1.63it/s]
                                                         
{'loss': 1.5497, 'learning_rate': 2.522958246602179e-07, 'epoch': 3362.96}
+
 83%|████████▎ | 54480/65536 [9:23:23<1:53:16,  1.63it/s]
 83%|████████▎ | 54481/65536 [9:23:23<1:55:32,  1.59it/s]
 83%|████████▎ | 54482/65536 [9:23:24<1:55:34,  1.59it/s]
 83%|████████▎ | 54483/65536 [9:23:24<1:54:51,  1.60it/s]
 83%|████████▎ | 54484/65536 [9:23:25<1:57:31,  1.57it/s]
 83%|████████▎ | 54485/65536 [9:23:26<1:56:20,  1.58it/s]
 83%|████████▎ | 54486/65536 [9:23:26<1:53:14,  1.63it/s]
 83%|████████▎ | 54487/65536 [9:23:27<1:51:38,  1.65it/s]
 83%|████████▎ | 54488/65536 [9:23:28<1:51:09,  1.66it/s]
 83%|████████▎ | 54489/65536 [9:23:28<1:55:08,  1.60it/s]
 83%|████████▎ | 54490/65536 [9:23:29<1:51:48,  1.65it/s]
 83%|████████▎ | 54491/65536 [9:23:29<1:51:52,  1.65it/s]
 83%|████████▎ | 54492/65536 [9:23:30<1:52:55,  1.63it/s]
 83%|████████▎ | 54493/65536 [9:23:31<1:56:29,  1.58it/s]
 83%|████████▎ | 54494/65536 [9:23:31<1:53:38,  1.62it/s]
 83%|████████▎ | 54495/65536 [9:23:32<1:54:48,  1.60it/s]
 83%|████████▎ | 54496/65536 [9:23:33<1:53:35,  1.62it/s]
 83%|████████▎ | 54497/65536 [9:23:33<1:57:58,  1.56it/s]
 83%|████████▎ | 54498/65536 [9:23:34<1:58:46,  1.55it/s]
 83%|████████▎ | 54499/65536 [9:23:35<2:04:47,  1.47it/s]
 83%|████████▎ | 54500/65536 [9:23:35<2:01:08,  1.52it/s]
                                                         
{'loss': 1.6113, 'learning_rate': 2.520203257009918e-07, 'epoch': 3364.2}
+
 83%|████████▎ | 54500/65536 [9:23:35<2:01:08,  1.52it/s]
 83%|████████▎ | 54501/65536 [9:23:36<2:02:28,  1.50it/s]
 83%|████████▎ | 54502/65536 [9:23:37<1:58:49,  1.55it/s]
 83%|████████▎ | 54503/65536 [9:23:37<1:57:56,  1.56it/s]
 83%|████████▎ | 54504/65536 [9:23:38<1:55:22,  1.59it/s]
 83%|████████▎ | 54505/65536 [9:23:38<1:55:59,  1.59it/s]
 83%|████████▎ | 54506/65536 [9:23:39<1:55:38,  1.59it/s]
 83%|████████▎ | 54507/65536 [9:23:40<1:54:14,  1.61it/s]
 83%|████████▎ | 54508/65536 [9:23:40<1:51:50,  1.64it/s]
 83%|████████▎ | 54509/65536 [9:23:41<1:50:39,  1.66it/s]
 83%|████████▎ | 54510/65536 [9:23:41<1:54:05,  1.61it/s]
 83%|████████▎ | 54511/65536 [9:23:42<1:52:08,  1.64it/s]
 83%|████████▎ | 54512/65536 [9:23:43<1:50:45,  1.66it/s]
 83%|████████▎ | 54513/65536 [9:23:43<1:50:11,  1.67it/s]
 83%|████████▎ | 54514/65536 [9:23:44<1:55:24,  1.59it/s]
 83%|████████▎ | 54515/65536 [9:23:45<1:53:43,  1.62it/s]
 83%|████████▎ | 54516/65536 [9:23:45<1:52:18,  1.64it/s]
 83%|████████▎ | 54517/65536 [9:23:46<1:52:44,  1.63it/s]
 83%|████████▎ | 54518/65536 [9:23:46<1:53:24,  1.62it/s]
 83%|████████▎ | 54519/65536 [9:23:47<1:53:22,  1.62it/s]
 83%|████████▎ | 54520/65536 [9:23:48<1:55:01,  1.60it/s]
                                                         
{'loss': 1.5709, 'learning_rate': 2.5174482674176564e-07, 'epoch': 3365.43}
+
 83%|████████▎ | 54520/65536 [9:23:48<1:55:01,  1.60it/s]
 83%|██████��█▎ | 54521/65536 [9:23:48<1:57:41,  1.56it/s]
 83%|████████▎ | 54522/65536 [9:23:49<1:54:04,  1.61it/s]
 83%|████████▎ | 54523/65536 [9:23:49<1:52:53,  1.63it/s]
 83%|████████▎ | 54524/65536 [9:23:50<1:53:30,  1.62it/s]
 83%|████████▎ | 54525/65536 [9:23:51<1:54:07,  1.61it/s]
 83%|████████▎ | 54526/65536 [9:23:51<1:54:17,  1.61it/s]
 83%|████████▎ | 54527/65536 [9:23:52<1:56:43,  1.57it/s]
 83%|████████▎ | 54528/65536 [9:23:53<1:55:58,  1.58it/s]
 83%|████████▎ | 54529/65536 [9:23:53<1:51:32,  1.64it/s]
 83%|████████▎ | 54530/65536 [9:23:54<1:56:19,  1.58it/s]
 83%|████████▎ | 54531/65536 [9:23:55<1:55:39,  1.59it/s]
 83%|████████▎ | 54532/65536 [9:23:55<1:58:07,  1.55it/s]
 83%|████████▎ | 54533/65536 [9:23:56<1:55:03,  1.59it/s]
 83%|████████▎ | 54534/65536 [9:23:56<1:52:21,  1.63it/s]
 83%|████████▎ | 54535/65536 [9:23:57<1:54:52,  1.60it/s]
 83%|████████▎ | 54536/65536 [9:23:58<1:53:23,  1.62it/s]
 83%|████████▎ | 54537/65536 [9:23:58<1:53:06,  1.62it/s]
 83%|████████▎ | 54538/65536 [9:23:59<1:52:16,  1.63it/s]
 83%|████████▎ | 54539/65536 [9:23:59<1:52:29,  1.63it/s]
 83%|████████▎ | 54540/65536 [9:24:00<1:54:24,  1.60it/s]
                                                         
{'loss': 1.5189, 'learning_rate': 2.514693277825395e-07, 'epoch': 3366.67}
+
 83%|████████▎ | 54540/65536 [9:24:00<1:54:24,  1.60it/s]
 83%|████████▎ | 54541/65536 [9:24:01<1:53:06,  1.62it/s]
 83%|████████▎ | 54542/65536 [9:24:01<1:52:35,  1.63it/s]
 83%|████████▎ | 54543/65536 [9:24:02<1:54:00,  1.61it/s]
 83%|████████▎ | 54544/65536 [9:24:03<1:52:09,  1.63it/s]
 83%|████████▎ | 54545/65536 [9:24:03<1:50:15,  1.66it/s]
 83%|████████▎ | 54546/65536 [9:24:04<1:56:42,  1.57it/s]
 83%|████████▎ | 54547/65536 [9:24:04<1:58:25,  1.55it/s]
 83%|████████▎ | 54548/65536 [9:24:05<1:54:24,  1.60it/s]
 83%|████████▎ | 54549/65536 [9:24:06<1:54:29,  1.60it/s]
 83%|████████▎ | 54550/65536 [9:24:06<1:56:53,  1.57it/s]
 83%|████████▎ | 54551/65536 [9:24:07<1:56:24,  1.57it/s]
 83%|████████▎ | 54552/65536 [9:24:08<1:57:40,  1.56it/s]
 83%|████████▎ | 54553/65536 [9:24:08<1:55:34,  1.58it/s]
 83%|████████▎ | 54554/65536 [9:24:09<1:54:52,  1.59it/s]
 83%|████████▎ | 54555/65536 [9:24:09<1:51:45,  1.64it/s]
 83%|████████▎ | 54556/65536 [9:24:10<1:49:25,  1.67it/s]
 83%|████████▎ | 54557/65536 [9:24:11<1:52:12,  1.63it/s]
 83%|████████▎ | 54558/65536 [9:24:11<1:52:31,  1.63it/s]
 83%|████████▎ | 54559/65536 [9:24:12<1:52:31,  1.63it/s]
 83%|████████▎ | 54560/65536 [9:24:12<1:50:28,  1.66it/s]
                                                         
{'loss': 1.5557, 'learning_rate': 2.5119382882331336e-07, 'epoch': 3367.9}
+
 83%|████████▎ | 54560/65536 [9:24:12<1:50:28,  1.66it/s]
 83%|████████▎ | 54561/65536 [9:24:13<1:52:38,  1.62it/s]
 83%|████████▎ | 54562/65536 [9:24:14<1:55:13,  1.59it/s]
 83%|████████▎ | 54563/65536 [9:24:14<1:52:40,  1.62it/s]
 83%|████████▎ | 54564/65536 [9:24:15<1:53:58,  1.60it/s]
 83%|████████▎ | 54565/65536 [9:24:16<1:54:44,  1.59it/s]
 83%|████████▎ | 54566/65536 [9:24:16<1:54:50,  1.59it/s]
 83%|████████▎ | 54567/65536 [9:24:17<1:52:21,  1.63it/s]
 83%|████████▎ | 54568/65536 [9:24:17<1:51:22,  1.64it/s]
 83%|████████▎ | 54569/65536 [9:24:18<1:53:31,  1.61it/s]
 83%|████████▎ | 54570/65536 [9:24:19<1:55:53,  1.58it/s]
 83%|████████▎ | 54571/65536 [9:24:19<1:53:56,  1.60it/s]
 83%|████████▎ | 54572/65536 [9:24:20<1:52:33,  1.62it/s]
 83%|████████▎ | 54573/65536 [9:24:21<1:55:43,  1.58it/s]
 83%|████████▎ | 54574/65536 [9:24:21<1:55:58,  1.58it/s]
 83%|████████▎ | 54575/65536 [9:24:22<1:56:35,  1.57it/s]
 83%|████████▎ | 54576/65536 [9:24:23<1:55:01,  1.59it/s]
 83%|████████▎ | 54577/65536 [9:24:23<1:56:20,  1.57it/s]
 83%|████████▎ | 54578/65536 [9:24:24<1:58:07,  1.55it/s]
 83%|████████▎ | 54579/65536 [9:24:24<1:56:16,  1.57it/s]
 83%|████████▎ | 54580/65536 [9:24:25<1:53:38,  1.61it/s]
                                                         
{'loss': 1.5679, 'learning_rate': 2.509183298640871e-07, 'epoch': 3369.14}
+
 83%|███████��▎ | 54580/65536 [9:24:25<1:53:38,  1.61it/s]
 83%|████████▎ | 54581/65536 [9:24:26<1:53:21,  1.61it/s]
 83%|████████▎ | 54582/65536 [9:24:26<1:50:50,  1.65it/s]
 83%|████████▎ | 54583/65536 [9:24:27<1:50:08,  1.66it/s]
 83%|████████▎ | 54584/65536 [9:24:27<1:51:13,  1.64it/s]
 83%|████████▎ | 54585/65536 [9:24:28<1:49:23,  1.67it/s]
 83%|████████▎ | 54586/65536 [9:24:29<1:50:02,  1.66it/s]
 83%|████████▎ | 54587/65536 [9:24:29<1:53:59,  1.60it/s]
 83%|████████▎ | 54588/65536 [9:24:30<1:52:49,  1.62it/s]
 83%|████████▎ | 54589/65536 [9:24:31<1:58:12,  1.54it/s]
 83%|████████▎ | 54590/65536 [9:24:31<1:56:59,  1.56it/s]
 83%|████████▎ | 54591/65536 [9:24:32<1:54:15,  1.60it/s]
 83%|████████▎ | 54592/65536 [9:24:32<1:54:09,  1.60it/s]
 83%|████████▎ | 54593/65536 [9:24:33<1:52:03,  1.63it/s]
 83%|████████▎ | 54594/65536 [9:24:34<1:52:16,  1.62it/s]
 83%|████████▎ | 54595/65536 [9:24:34<1:55:27,  1.58it/s]
 83%|████████▎ | 54596/65536 [9:24:35<1:54:24,  1.59it/s]
 83%|████████▎ | 54597/65536 [9:24:36<1:55:51,  1.57it/s]
 83%|████████▎ | 54598/65536 [9:24:36<1:58:19,  1.54it/s]
 83%|████████▎ | 54599/65536 [9:24:37<1:54:24,  1.59it/s]
 83%|████████▎ | 54600/65536 [9:24:38<1:54:27,  1.59it/s]
                                                         
{'loss': 1.5589, 'learning_rate': 2.5064283090486097e-07, 'epoch': 3370.37}
+
 83%|████████▎ | 54600/65536 [9:24:38<1:54:27,  1.59it/s]
 83%|████████▎ | 54601/65536 [9:24:38<1:53:33,  1.60it/s]
 83%|████████▎ | 54602/65536 [9:24:39<1:53:29,  1.61it/s]
 83%|████████▎ | 54603/65536 [9:24:39<1:51:06,  1.64it/s]
 83%|████████▎ | 54604/65536 [9:24:40<1:51:13,  1.64it/s]
 83%|████████▎ | 54605/65536 [9:24:41<1:53:09,  1.61it/s]
 83%|████████▎ | 54606/65536 [9:24:41<1:53:40,  1.60it/s]
 83%|████████▎ | 54607/65536 [9:24:42<1:52:08,  1.62it/s]
 83%|████████▎ | 54608/65536 [9:24:42<1:52:39,  1.62it/s]
 83%|████████▎ | 54609/65536 [9:24:43<1:57:34,  1.55it/s]
 83%|████████▎ | 54610/65536 [9:24:44<1:55:55,  1.57it/s]
 83%|████████▎ | 54611/65536 [9:24:44<1:59:32,  1.52it/s]
 83%|████████▎ | 54612/65536 [9:24:45<1:58:40,  1.53it/s]
 83%|████████▎ | 54613/65536 [9:24:46<1:57:50,  1.54it/s]
 83%|████████▎ | 54614/65536 [9:24:46<1:59:05,  1.53it/s]
 83%|████████▎ | 54615/65536 [9:24:47<1:56:05,  1.57it/s]
 83%|████████▎ | 54616/65536 [9:24:48<1:52:49,  1.61it/s]
 83%|████████▎ | 54617/65536 [9:24:48<1:53:58,  1.60it/s]
 83%|████████▎ | 54618/65536 [9:24:49<1:51:39,  1.63it/s]
 83%|████████▎ | 54619/65536 [9:24:49<1:51:52,  1.63it/s]
 83%|████████▎ | 54620/65536 [9:24:50<1:53:42,  1.60it/s]
                                                         
{'loss': 1.5288, 'learning_rate': 2.503673319456348e-07, 'epoch': 3371.6}
+
 83%|████████▎ | 54620/65536 [9:24:50<1:53:42,  1.60it/s]
 83%|████████▎ | 54621/65536 [9:24:51<1:51:56,  1.63it/s]
 83%|████████▎ | 54622/65536 [9:24:51<1:55:07,  1.58it/s]
 83%|████████▎ | 54623/65536 [9:24:52<1:53:56,  1.60it/s]
 83%|████████▎ | 54624/65536 [9:24:53<1:54:11,  1.59it/s]
 83%|████████▎ | 54625/65536 [9:24:53<1:53:55,  1.60it/s]
 83%|████████▎ | 54626/65536 [9:24:54<1:52:27,  1.62it/s]
 83%|████████▎ | 54627/65536 [9:24:54<1:55:14,  1.58it/s]
 83%|████████▎ | 54628/65536 [9:24:55<1:55:39,  1.57it/s]
 83%|████████▎ | 54629/65536 [9:24:56<1:53:21,  1.60it/s]
 83%|████████▎ | 54630/65536 [9:24:56<1:52:49,  1.61it/s]
 83%|████████▎ | 54631/65536 [9:24:57<1:53:21,  1.60it/s]
 83%|████████▎ | 54632/65536 [9:24:58<1:55:12,  1.58it/s]
 83%|████████▎ | 54633/65536 [9:24:58<1:55:05,  1.58it/s]
 83%|████████▎ | 54634/65536 [9:24:59<1:53:45,  1.60it/s]
 83%|████████▎ | 54635/65536 [9:24:59<1:51:11,  1.63it/s]
 83%|████████▎ | 54636/65536 [9:25:00<1:49:00,  1.67it/s]
 83%|████████▎ | 54637/65536 [9:25:01<1:47:39,  1.69it/s]
 83%|████████▎ | 54638/65536 [9:25:01<1:51:41,  1.63it/s]
 83%|████████▎ | 54639/65536 [9:25:02<1:50:23,  1.65it/s]
 83%|████████▎ | 54640/65536 [9:25:02<1:51:19,  1.63it/s]
                                                         
{'loss': 1.5986, 'learning_rate': 2.500918329864087e-07, 'epoch': 3372.84}
+
 83%|████████▎ | 54640/65536 [9:25:02<1:51:19,  1.63it/s]
 83%|████████▎ | 54641/65536 [9:25:03<1:53:57,  1.59it/s]
 83%|████████▎ | 54642/65536 [9:25:04<1:53:34,  1.60it/s]
 83%|████████▎ | 54643/65536 [9:25:04<1:54:31,  1.59it/s]
 83%|████████▎ | 54644/65536 [9:25:05<1:53:33,  1.60it/s]
 83%|████████▎ | 54645/65536 [9:25:06<1:53:04,  1.61it/s]
 83%|████████▎ | 54646/65536 [9:25:06<1:52:47,  1.61it/s]
 83%|████████▎ | 54647/65536 [9:25:07<1:53:20,  1.60it/s]
 83%|████████▎ | 54648/65536 [9:25:08<1:53:46,  1.59it/s]
 83%|████████▎ | 54649/65536 [9:25:08<1:55:30,  1.57it/s]
 83%|████████▎ | 54650/65536 [9:25:09<1:53:54,  1.59it/s]
 83%|████████▎ | 54651/65536 [9:25:09<1:57:12,  1.55it/s]
 83%|████████▎ | 54652/65536 [9:25:10<1:55:15,  1.57it/s]
 83%|████████▎ | 54653/65536 [9:25:11<1:52:36,  1.61it/s]
 83%|████████▎ | 54654/65536 [9:25:11<1:52:14,  1.62it/s]
 83%|████████▎ | 54655/65536 [9:25:12<1:52:34,  1.61it/s]
 83%|████████▎ | 54656/65536 [9:25:13<1:50:57,  1.63it/s]
 83%|████████▎ | 54657/65536 [9:25:13<1:50:41,  1.64it/s]
 83%|████████▎ | 54658/65536 [9:25:14<1:50:27,  1.64it/s]
 83%|████████▎ | 54659/65536 [9:25:14<1:53:25,  1.60it/s]
 83%|████████▎ | 54660/65536 [9:25:15<1:53:17,  1.60it/s]
                                                         
{'loss': 1.5805, 'learning_rate': 2.4981633402718254e-07, 'epoch': 3374.07}
+
 83%|████████▎ | 54660/65536 [9:25:15<1:53:17,  1.60it/s]
 83%|████████▎ | 54661/65536 [9:25:16<1:56:07,  1.56it/s]
 83%|████████▎ | 54662/65536 [9:25:16<1:57:26,  1.54it/s]
 83%|████████▎ | 54663/65536 [9:25:17<1:55:22,  1.57it/s]
 83%|████████▎ | 54664/65536 [9:25:18<1:54:23,  1.58it/s]
 83%|████████▎ | 54665/65536 [9:25:18<1:53:23,  1.60it/s]
 83%|████████▎ | 54666/65536 [9:25:19<1:51:58,  1.62it/s]
 83%|████████▎ | 54667/65536 [9:25:19<1:50:57,  1.63it/s]
 83%|████████▎ | 54668/65536 [9:25:20<1:49:52,  1.65it/s]
 83%|████████▎ | 54669/65536 [9:25:21<1:52:20,  1.61it/s]
 83%|████████▎ | 54670/65536 [9:25:21<1:52:20,  1.61it/s]
 83%|████████▎ | 54671/65536 [9:25:22<1:51:20,  1.63it/s]
 83%|████████▎ | 54672/65536 [9:25:22<1:52:47,  1.61it/s]
 83%|████████▎ | 54673/65536 [9:25:23<1:55:13,  1.57it/s]
 83%|████████▎ | 54674/65536 [9:25:24<1:54:25,  1.58it/s]
 83%|████████▎ | 54675/65536 [9:25:24<1:51:51,  1.62it/s]
 83%|████████▎ | 54676/65536 [9:25:25<1:56:11,  1.56it/s]
 83%|████████▎ | 54677/65536 [9:25:26<1:55:03,  1.57it/s]
 83%|████████▎ | 54678/65536 [9:25:26<1:54:58,  1.57it/s]
 83%|████████▎ | 54679/65536 [9:25:27<1:53:51,  1.59it/s]
 83%|████████▎ | 54680/65536 [9:25:28<1:52:59,  1.60it/s]
                                                         
{'loss': 1.5344, 'learning_rate': 2.495408350679564e-07, 'epoch': 3375.31}
+
 83%|████████▎ | 54680/65536 [9:25:28<1:52:59,  1.60it/s]
 83%|████████▎ | 54681/65536 [9:25:28<1:50:13,  1.64it/s]
 83%|████████▎ | 54682/65536 [9:25:29<1:49:24,  1.65it/s]
 83%|████████▎ | 54683/65536 [9:25:29<1:50:09,  1.64it/s]
 83%|████████▎ | 54684/65536 [9:25:30<1:48:45,  1.66it/s]
 83%|████████▎ | 54685/65536 [9:25:31<1:48:17,  1.67it/s]
 83%|████████▎ | 54686/65536 [9:25:31<1:47:30,  1.68it/s]
 83%|████████▎ | 54687/65536 [9:25:32<1:48:13,  1.67it/s]
 83%|████████▎ | 54688/65536 [9:25:32<1:52:10,  1.61it/s]
 83%|████████▎ | 54689/65536 [9:25:33<1:54:26,  1.58it/s]
 83%|████████▎ | 54690/65536 [9:25:34<1:53:36,  1.59it/s]
 83%|████████▎ | 54691/65536 [9:25:34<1:50:13,  1.64it/s]
 83%|████████▎ | 54692/65536 [9:25:35<1:53:11,  1.60it/s]
 83%|████████▎ | 54693/65536 [9:25:36<1:53:42,  1.59it/s]
 83%|████████▎ | 54694/65536 [9:25:36<1:53:00,  1.60it/s]
 83%|████████▎ | 54695/65536 [9:25:37<1:51:50,  1.62it/s]
 83%|████████▎ | 54696/65536 [9:25:37<1:48:54,  1.66it/s]
 83%|████████▎ | 54697/65536 [9:25:38<1:50:18,  1.64it/s]
 83%|████████▎ | 54698/65536 [9:25:39<1:50:28,  1.64it/s]
 83%|████████▎ | 54699/65536 [9:25:39<1:49:55,  1.64it/s]
 83%|████████▎ | 54700/65536 [9:25:40<1:50:03,  1.64it/s]
                                                         
{'loss': 1.5753, 'learning_rate': 2.4926533610873026e-07, 'epoch': 3376.54}
+
 83%|████████▎ | 54700/65536 [9:25:40<1:50:03,  1.64it/s]
 83%|████████▎ | 54701/65536 [9:25:40<1:48:37,  1.66it/s]
 83%|████████▎ | 54702/65536 [9:25:41<1:49:24,  1.65it/s]
 83%|████████▎ | 54703/65536 [9:25:42<1:49:18,  1.65it/s]
 83%|████████▎ | 54704/65536 [9:25:42<1:48:41,  1.66it/s]
 83%|████████▎ | 54705/65536 [9:25:43<1:51:25,  1.62it/s]
 83%|████████▎ | 54706/65536 [9:25:43<1:54:30,  1.58it/s]
 83%|████████▎ | 54707/65536 [9:25:44<1:52:04,  1.61it/s]
 83%|████████▎ | 54708/65536 [9:25:45<1:52:31,  1.60it/s]
 83%|████████▎ | 54709/65536 [9:25:45<1:49:19,  1.65it/s]
 83%|████████▎ | 54710/65536 [9:25:46<1:52:05,  1.61it/s]
 83%|████████▎ | 54711/65536 [9:25:47<1:51:56,  1.61it/s]
 83%|████████▎ | 54712/65536 [9:25:47<1:53:27,  1.59it/s]
 83%|████████▎ | 54713/65536 [9:25:48<1:52:33,  1.60it/s]
 83%|████████▎ | 54714/65536 [9:25:48<1:51:49,  1.61it/s]
 83%|████████▎ | 54715/65536 [9:25:49<1:48:58,  1.65it/s]
 83%|████████▎ | 54716/65536 [9:25:50<1:50:03,  1.64it/s]
 83%|████████▎ | 54717/65536 [9:25:50<1:52:20,  1.61it/s]
 83%|████████▎ | 54718/65536 [9:25:51<1:54:05,  1.58it/s]
 83%|████████▎ | 54719/65536 [9:25:52<1:58:14,  1.52it/s]
 83%|████████▎ | 54720/65536 [9:25:52<2:01:39,  1.48it/s]
                                                         
{'loss': 1.538, 'learning_rate': 2.48989837149504e-07, 'epoch': 3377.78}
+
 83%|████████▎ | 54720/65536 [9:25:52<2:01:39,  1.48it/s]
 83%|████████▎ | 54721/65536 [9:25:53<2:01:03,  1.49it/s]
 83%|████████▎ | 54722/65536 [9:25:54<1:56:48,  1.54it/s]
 84%|████████▎ | 54723/65536 [9:25:54<1:54:53,  1.57it/s]
 84%|████████▎ | 54724/65536 [9:25:55<1:56:36,  1.55it/s]
 84%|████████▎ | 54725/65536 [9:25:56<1:55:28,  1.56it/s]
 84%|████████▎ | 54726/65536 [9:25:56<1:53:19,  1.59it/s]
 84%|████████▎ | 54727/65536 [9:25:57<1:52:19,  1.60it/s]
 84%|████████▎ | 54728/65536 [9:25:57<1:49:16,  1.65it/s]
 84%|████████▎ | 54729/65536 [9:25:58<1:50:03,  1.64it/s]
 84%|████████▎ | 54730/65536 [9:25:59<1:52:28,  1.60it/s]
 84%|████████▎ | 54731/65536 [9:25:59<1:50:13,  1.63it/s]
 84%|████████▎ | 54732/65536 [9:26:00<1:52:54,  1.59it/s]
 84%|████████▎ | 54733/65536 [9:26:00<1:53:13,  1.59it/s]
 84%|████████▎ | 54734/65536 [9:26:01<1:51:31,  1.61it/s]
 84%|████████▎ | 54735/65536 [9:26:02<1:50:07,  1.63it/s]
 84%|████████▎ | 54736/65536 [9:26:02<1:48:19,  1.66it/s]
 84%|████████▎ | 54737/65536 [9:26:03<1:50:19,  1.63it/s]
 84%|████████▎ | 54738/65536 [9:26:03<1:50:59,  1.62it/s]
 84%|████████▎ | 54739/65536 [9:26:04<1:49:02,  1.65it/s]
 84%|████████▎ | 54740/65536 [9:26:05<1:52:00,  1.61it/s]
                                                         
{'loss': 1.615, 'learning_rate': 2.4871433819027787e-07, 'epoch': 3379.01}
+
 84%|████████▎ | 54740/65536 [9:26:05<1:52:00,  1.61it/s]
 84%|████████▎ | 54741/65536 [9:26:05<1:52:17,  1.60it/s]
 84%|████████▎ | 54742/65536 [9:26:06<1:55:20,  1.56it/s]
 84%|████████▎ | 54743/65536 [9:26:07<1:52:18,  1.60it/s]
 84%|████████▎ | 54744/65536 [9:26:07<1:52:01,  1.61it/s]
 84%|████████▎ | 54745/65536 [9:26:08<1:54:05,  1.58it/s]
 84%|████████▎ | 54746/65536 [9:26:08<1:51:33,  1.61it/s]
 84%|████████▎ | 54747/65536 [9:26:09<1:51:36,  1.61it/s]
 84%|████████▎ | 54748/65536 [9:26:10<1:52:24,  1.60it/s]
 84%|████████▎ | 54749/65536 [9:26:10<1:48:45,  1.65it/s]
 84%|████████▎ | 54750/65536 [9:26:11<1:51:54,  1.61it/s]
 84%|████████▎ | 54751/65536 [9:26:12<1:53:13,  1.59it/s]
 84%|████████▎ | 54752/65536 [9:26:12<1:51:27,  1.61it/s]
 84%|████████▎ | 54753/65536 [9:26:13<1:49:58,  1.63it/s]
 84%|████████▎ | 54754/65536 [9:26:13<1:48:41,  1.65it/s]
 84%|████████▎ | 54755/65536 [9:26:14<1:49:27,  1.64it/s]
 84%|████████▎ | 54756/65536 [9:26:15<1:49:04,  1.65it/s]
 84%|████████▎ | 54757/65536 [9:26:15<1:51:54,  1.61it/s]
 84%|████████▎ | 54758/65536 [9:26:16<1:51:11,  1.62it/s]
 84%|████████▎ | 54759/65536 [9:26:17<1:51:19,  1.61it/s]
 84%|████████▎ | 54760/65536 [9:26:17<1:48:09,  1.66it/s]
                                                         
{'loss': 1.5935, 'learning_rate': 2.4843883923105173e-07, 'epoch': 3380.25}
+
 84%|████████▎ | 54760/65536 [9:26:17<1:48:09,  1.66it/s]
 84%|████████▎ | 54761/65536 [9:26:18<1:49:23,  1.64it/s]
 84%|████████▎ | 54762/65536 [9:26:18<1:47:44,  1.67it/s]
 84%|████████▎ | 54763/65536 [9:26:19<1:46:49,  1.68it/s]
 84%|████████▎ | 54764/65536 [9:26:19<1:49:00,  1.65it/s]
 84%|████████▎ | 54765/65536 [9:26:20<1:48:23,  1.66it/s]
 84%|████████▎ | 54766/65536 [9:26:21<1:52:28,  1.60it/s]
 84%|████████▎ | 54767/65536 [9:26:21<1:51:13,  1.61it/s]
 84%|████████▎ | 54768/65536 [9:26:22<1:52:04,  1.60it/s]
 84%|████████▎ | 54769/65536 [9:26:23<1:52:27,  1.60it/s]
 84%|████████▎ | 54770/65536 [9:26:23<1:55:47,  1.55it/s]
 84%|████████▎ | 54771/65536 [9:26:24<1:54:07,  1.57it/s]
 84%|████████▎ | 54772/65536 [9:26:25<1:51:11,  1.61it/s]
 84%|████████▎ | 54773/65536 [9:26:25<1:57:43,  1.52it/s]
 84%|████████▎ | 54774/65536 [9:26:26<1:56:39,  1.54it/s]
 84%|████████▎ | 54775/65536 [9:26:27<1:55:27,  1.55it/s]
 84%|████████▎ | 54776/65536 [9:26:27<1:52:45,  1.59it/s]
 84%|████████▎ | 54777/65536 [9:26:28<1:56:30,  1.54it/s]
 84%|████████▎ | 54778/65536 [9:26:28<1:52:04,  1.60it/s]
 84%|████████▎ | 54779/65536 [9:26:29<1:50:53,  1.62it/s]
 84%|████████▎ | 54780/65536 [9:26:30<1:52:34,  1.59it/s]
                                                         
{'loss': 1.5214, 'learning_rate': 2.481633402718256e-07, 'epoch': 3381.48}
+
 84%|████████▎ | 54780/65536 [9:26:30<1:52:34,  1.59it/s]
 84%|████████▎ | 54781/65536 [9:26:30<1:53:44,  1.58it/s]
 84%|████████▎ | 54782/65536 [9:26:31<1:52:08,  1.60it/s]
 84%|████████▎ | 54783/65536 [9:26:32<1:51:26,  1.61it/s]
 84%|████████▎ | 54784/65536 [9:26:32<1:50:50,  1.62it/s]
 84%|████████▎ | 54785/65536 [9:26:33<1:51:02,  1.61it/s]
 84%|████████▎ | 54786/65536 [9:26:33<1:50:07,  1.63it/s]
 84%|████████▎ | 54787/65536 [9:26:34<1:47:59,  1.66it/s]
 84%|████████▎ | 54788/65536 [9:26:35<1:48:40,  1.65it/s]
 84%|████████▎ | 54789/65536 [9:26:35<1:52:53,  1.59it/s]
 84%|████████▎ | 54790/65536 [9:26:36<1:52:12,  1.60it/s]
 84%|████████▎ | 54791/65536 [9:26:36<1:49:43,  1.63it/s]
 84%|████████▎ | 54792/65536 [9:26:37<1:49:48,  1.63it/s]
 84%|████████▎ | 54793/65536 [9:26:38<1:49:38,  1.63it/s]
 84%|████████▎ | 54794/65536 [9:26:38<1:52:02,  1.60it/s]
 84%|████████▎ | 54795/65536 [9:26:39<1:50:11,  1.62it/s]
 84%|████████▎ | 54796/65536 [9:26:40<1:51:47,  1.60it/s]
 84%|████████▎ | 54797/65536 [9:26:40<1:52:22,  1.59it/s]
 84%|████████▎ | 54798/65536 [9:26:41<1:51:04,  1.61it/s]
 84%|████████▎ | 54799/65536 [9:26:41<1:49:14,  1.64it/s]
 84%|████████▎ | 54800/65536 [9:26:42<1:49:35,  1.63it/s]
                                                         
{'loss': 1.5463, 'learning_rate': 2.4788784131259945e-07, 'epoch': 3382.72}
+
 84%|████████▎ | 54800/65536 [9:26:42<1:49:35,  1.63it/s]
 84%|████████▎ | 54801/65536 [9:26:43<1:50:52,  1.61it/s]
 84%|████████▎ | 54802/65536 [9:26:43<1:47:54,  1.66it/s]
 84%|████████▎ | 54803/65536 [9:26:44<1:46:23,  1.68it/s]
 84%|████████▎ | 54804/65536 [9:26:44<1:48:07,  1.65it/s]
 84%|████████▎ | 54805/65536 [9:26:45<1:52:33,  1.59it/s]
 84%|████████▎ | 54806/65536 [9:26:46<1:51:54,  1.60it/s]
 84%|████████▎ | 54807/65536 [9:26:46<1:51:33,  1.60it/s]
 84%|████████▎ | 54808/65536 [9:26:47<1:51:52,  1.60it/s]
 84%|████████▎ | 54809/65536 [9:26:48<1:50:34,  1.62it/s]
 84%|████████▎ | 54810/65536 [9:26:48<1:50:13,  1.62it/s]
 84%|████████▎ | 54811/65536 [9:26:49<1:49:37,  1.63it/s]
 84%|████████▎ | 54812/65536 [9:26:49<1:51:49,  1.60it/s]
 84%|████████▎ | 54813/65536 [9:26:50<1:50:42,  1.61it/s]
 84%|████████▎ | 54814/65536 [9:26:51<1:49:26,  1.63it/s]
 84%|████████▎ | 54815/65536 [9:26:51<1:46:46,  1.67it/s]
 84%|████████▎ | 54816/65536 [9:26:52<1:47:43,  1.66it/s]
 84%|████████▎ | 54817/65536 [9:26:52<1:49:42,  1.63it/s]
 84%|████████▎ | 54818/65536 [9:26:53<1:52:11,  1.59it/s]
 84%|█████���██▎ | 54819/65536 [9:26:54<1:50:58,  1.61it/s]
 84%|████████▎ | 54820/65536 [9:26:54<1:51:16,  1.60it/s]
                                                         
{'loss': 1.5678, 'learning_rate': 2.4761234235337336e-07, 'epoch': 3383.95}
+
 84%|████████▎ | 54820/65536 [9:26:54<1:51:16,  1.60it/s]
 84%|████████▎ | 54821/65536 [9:26:55<1:55:53,  1.54it/s]
 84%|████████▎ | 54822/65536 [9:26:56<1:54:53,  1.55it/s]
 84%|████████▎ | 54823/65536 [9:26:56<1:52:49,  1.58it/s]
 84%|████████▎ | 54824/65536 [9:26:57<1:50:57,  1.61it/s]
 84%|████████▎ | 54825/65536 [9:26:57<1:48:55,  1.64it/s]
 84%|████████▎ | 54826/65536 [9:26:58<1:48:29,  1.65it/s]
 84%|████████▎ | 54827/65536 [9:26:59<1:53:07,  1.58it/s]
 84%|████████▎ | 54828/65536 [9:26:59<1:55:13,  1.55it/s]
 84%|████████▎ | 54829/65536 [9:27:00<1:52:43,  1.58it/s]
 84%|████████▎ | 54830/65536 [9:27:01<1:50:12,  1.62it/s]
 84%|████████▎ | 54831/65536 [9:27:01<1:50:38,  1.61it/s]
 84%|████████▎ | 54832/65536 [9:27:02<1:52:17,  1.59it/s]
 84%|████████▎ | 54833/65536 [9:27:03<1:52:16,  1.59it/s]
 84%|████████▎ | 54834/65536 [9:27:03<1:49:56,  1.62it/s]
 84%|████████▎ | 54835/65536 [9:27:04<1:50:25,  1.62it/s]
 84%|████████▎ | 54836/65536 [9:27:04<1:50:27,  1.61it/s]
 84%|████████▎ | 54837/65536 [9:27:05<1:48:16,  1.65it/s]
 84%|████████▎ | 54838/65536 [9:27:06<1:54:20,  1.56it/s]
 84%|████████▎ | 54839/65536 [9:27:06<1:52:40,  1.58it/s]
 84%|████████▎ | 54840/65536 [9:27:07<1:51:13,  1.60it/s]
                                                         
{'loss': 1.5581, 'learning_rate': 2.473368433941472e-07, 'epoch': 3385.19}
+
 84%|████████▎ | 54840/65536 [9:27:07<1:51:13,  1.60it/s]
 84%|████████▎ | 54841/65536 [9:27:08<1:52:13,  1.59it/s]
 84%|████████▎ | 54842/65536 [9:27:08<1:51:20,  1.60it/s]
 84%|████████▎ | 54843/65536 [9:27:09<1:49:58,  1.62it/s]
 84%|████████▎ | 54844/65536 [9:27:09<1:48:07,  1.65it/s]
 84%|████████▎ | 54845/65536 [9:27:10<1:50:39,  1.61it/s]
 84%|████████▎ | 54846/65536 [9:27:11<1:49:01,  1.63it/s]
 84%|████████▎ | 54847/65536 [9:27:11<1:49:00,  1.63it/s]
 84%|████████▎ | 54848/65536 [9:27:12<1:49:23,  1.63it/s]
 84%|████████▎ | 54849/65536 [9:27:12<1:47:47,  1.65it/s]
 84%|████████▎ | 54850/65536 [9:27:13<1:46:48,  1.67it/s]
 84%|████████▎ | 54851/65536 [9:27:14<1:47:59,  1.65it/s]
 84%|████████▎ | 54852/65536 [9:27:14<1:48:27,  1.64it/s]
 84%|████████▎ | 54853/65536 [9:27:15<1:48:43,  1.64it/s]
 84%|████████▎ | 54854/65536 [9:27:16<1:54:26,  1.56it/s]
 84%|████████▎ | 54855/65536 [9:27:16<1:52:11,  1.59it/s]
 84%|████████▎ | 54856/65536 [9:27:17<1:48:05,  1.65it/s]
 84%|████████▎ | 54857/65536 [9:27:17<1:49:12,  1.63it/s]
 84%|████████▎ | 54858/65536 [9:27:18<1:50:06,  1.62it/s]
 84%|████████▎ | 54859/65536 [9:27:19<1:48:26,  1.64it/s]
 84%|████████▎ | 54860/65536 [9:27:19<1:48:38,  1.64it/s]
                                                         
{'loss': 1.5805, 'learning_rate': 2.4706134443492097e-07, 'epoch': 3386.42}
+
 84%|████████▎ | 54860/65536 [9:27:19<1:48:38,  1.64it/s]
 84%|████████▎ | 54861/65536 [9:27:20<1:48:37,  1.64it/s]
 84%|████████▎ | 54862/65536 [9:27:20<1:51:14,  1.60it/s]
 84%|████████▎ | 54863/65536 [9:27:21<1:52:43,  1.58it/s]
 84%|████████▎ | 54864/65536 [9:27:22<1:49:48,  1.62it/s]
 84%|████████▎ | 54865/65536 [9:27:22<1:48:32,  1.64it/s]
 84%|████████▎ | 54866/65536 [9:27:23<1:50:39,  1.61it/s]
 84%|████████▎ | 54867/65536 [9:27:23<1:48:42,  1.64it/s]
 84%|████████▎ | 54868/65536 [9:27:24<1:49:27,  1.62it/s]
 84%|████████▎ | 54869/65536 [9:27:25<1:49:53,  1.62it/s]
 84%|████████▎ | 54870/65536 [9:27:25<1:54:03,  1.56it/s]
 84%|████████▎ | 54871/65536 [9:27:26<1:54:22,  1.55it/s]
 84%|████████▎ | 54872/65536 [9:27:27<1:51:58,  1.59it/s]
 84%|████████▎ | 54873/65536 [9:27:27<1:51:40,  1.59it/s]
 84%|████████▎ | 54874/65536 [9:27:28<1:48:29,  1.64it/s]
 84%|████████▎ | 54875/65536 [9:27:28<1:49:46,  1.62it/s]
 84%|████████▎ | 54876/65536 [9:27:29<1:50:10,  1.61it/s]
 84%|████████▎ | 54877/65536 [9:27:30<1:54:45,  1.55it/s]
 84%|█████��██▎ | 54878/65536 [9:27:30<1:52:49,  1.57it/s]
 84%|████████▎ | 54879/65536 [9:27:31<1:52:26,  1.58it/s]
 84%|████████▎ | 54880/65536 [9:27:32<1:52:00,  1.59it/s]
                                                         
{'loss': 1.506, 'learning_rate': 2.4678584547569483e-07, 'epoch': 3387.65}
+
 84%|████████▎ | 54880/65536 [9:27:32<1:52:00,  1.59it/s]
 84%|████████▎ | 54881/65536 [9:27:32<1:50:51,  1.60it/s]
 84%|████████▎ | 54882/65536 [9:27:33<1:51:57,  1.59it/s]
 84%|████████▎ | 54883/65536 [9:27:34<1:50:13,  1.61it/s]
 84%|████████▎ | 54884/65536 [9:27:34<1:49:10,  1.63it/s]
 84%|████████▎ | 54885/65536 [9:27:35<1:48:07,  1.64it/s]
 84%|████████▎ | 54886/65536 [9:27:35<1:48:55,  1.63it/s]
 84%|████████▍ | 54887/65536 [9:27:36<1:50:03,  1.61it/s]
 84%|████████▍ | 54888/65536 [9:27:37<1:52:34,  1.58it/s]
 84%|████████▍ | 54889/65536 [9:27:37<1:50:02,  1.61it/s]
 84%|████████▍ | 54890/65536 [9:27:38<1:50:07,  1.61it/s]
 84%|████████▍ | 54891/65536 [9:27:39<1:51:16,  1.59it/s]
 84%|████████▍ | 54892/65536 [9:27:39<1:48:09,  1.64it/s]
 84%|████████▍ | 54893/65536 [9:27:40<1:48:41,  1.63it/s]
 84%|████████▍ | 54894/65536 [9:27:40<1:50:02,  1.61it/s]
 84%|████████▍ | 54895/65536 [9:27:41<1:48:47,  1.63it/s]
 84%|████████▍ | 54896/65536 [9:27:42<1:51:22,  1.59it/s]
 84%|████████▍ | 54897/65536 [9:27:42<1:49:20,  1.62it/s]
 84%|████████▍ | 54898/65536 [9:27:43<1:48:03,  1.64it/s]
 84%|████████▍ | 54899/65536 [9:27:43<1:52:55,  1.57it/s]
 84%|████████▍ | 54900/65536 [9:27:44<1:49:37,  1.62it/s]
                                                         
{'loss': 1.5686, 'learning_rate': 2.465103465164687e-07, 'epoch': 3388.89}
+
 84%|████████▍ | 54900/65536 [9:27:44<1:49:37,  1.62it/s]
 84%|████████▍ | 54901/65536 [9:27:45<1:50:16,  1.61it/s]
 84%|████████▍ | 54902/65536 [9:27:45<1:55:11,  1.54it/s]
 84%|████████▍ | 54903/65536 [9:27:46<1:55:04,  1.54it/s]
 84%|████████▍ | 54904/65536 [9:27:47<1:56:25,  1.52it/s]
 84%|████████▍ | 54905/65536 [9:27:47<1:52:13,  1.58it/s]
 84%|████████▍ | 54906/65536 [9:27:48<1:48:11,  1.64it/s]
 84%|████████▍ | 54907/65536 [9:27:48<1:48:26,  1.63it/s]
 84%|████████▍ | 54908/65536 [9:27:49<1:49:31,  1.62it/s]
 84%|████████▍ | 54909/65536 [9:27:50<1:48:12,  1.64it/s]
 84%|████████▍ | 54910/65536 [9:27:50<1:46:55,  1.66it/s]
 84%|████████▍ | 54911/65536 [9:27:51<1:47:18,  1.65it/s]
 84%|████████▍ | 54912/65536 [9:27:51<1:44:50,  1.69it/s]
 84%|████████▍ | 54913/65536 [9:27:52<1:45:46,  1.67it/s]
 84%|████████▍ | 54914/65536 [9:27:53<1:53:48,  1.56it/s]
 84%|████████▍ | 54915/65536 [9:27:53<1:51:22,  1.59it/s]
 84%|████████▍ | 54916/65536 [9:27:54<1:49:23,  1.62it/s]
 84%|████████▍ | 54917/65536 [9:27:55<1:49:28,  1.62it/s]
 84%|████████▍ | 54918/65536 [9:27:55<1:48:27,  1.63it/s]
 84%|████████▍ | 54919/65536 [9:27:56<1:51:26,  1.59it/s]
 84%|████████▍ | 54920/65536 [9:27:57<1:51:17,  1.59it/s]
                                                         
{'loss': 1.5814, 'learning_rate': 2.4623484755724255e-07, 'epoch': 3390.12}
+
 84%|████████▍ | 54920/65536 [9:27:57<1:51:17,  1.59it/s]
 84%|████████▍ | 54921/65536 [9:27:57<1:53:55,  1.55it/s]
 84%|████████▍ | 54922/65536 [9:27:58<1:54:36,  1.54it/s]
 84%|████████▍ | 54923/65536 [9:27:58<1:50:53,  1.60it/s]
 84%|████████▍ | 54924/65536 [9:27:59<1:48:46,  1.63it/s]
 84%|████████▍ | 54925/65536 [9:28:00<1:51:40,  1.58it/s]
 84%|████████▍ | 54926/65536 [9:28:00<1:50:20,  1.60it/s]
 84%|████████▍ | 54927/65536 [9:28:01<1:50:15,  1.60it/s]
 84%|████████▍ | 54928/65536 [9:28:02<1:48:53,  1.62it/s]
 84%|████████▍ | 54929/65536 [9:28:02<1:50:21,  1.60it/s]
 84%|████████▍ | 54930/65536 [9:28:03<1:50:18,  1.60it/s]
 84%|████████▍ | 54931/65536 [9:28:03<1:48:08,  1.63it/s]
 84%|████████▍ | 54932/65536 [9:28:04<1:48:12,  1.63it/s]
 84%|████████▍ | 54933/65536 [9:28:05<1:50:35,  1.60it/s]
 84%|████████▍ | 54934/65536 [9:28:05<1:49:44,  1.61it/s]
 84%|████████▍ | 54935/65536 [9:28:06<1:56:18,  1.52it/s]
 84%|████████▍ | 54936/65536 [9:28:07<1:53:34,  1.56it/s]
 84%|██████���█▍ | 54937/65536 [9:28:07<1:51:51,  1.58it/s]
 84%|████████▍ | 54938/65536 [9:28:08<1:49:38,  1.61it/s]
 84%|████████▍ | 54939/65536 [9:28:08<1:48:38,  1.63it/s]
 84%|████████▍ | 54940/65536 [9:28:09<1:47:43,  1.64it/s]
                                                         
{'loss': 1.5282, 'learning_rate': 2.459593485980164e-07, 'epoch': 3391.36}
+
 84%|████████▍ | 54940/65536 [9:28:09<1:47:43,  1.64it/s]
 84%|████████▍ | 54941/65536 [9:28:10<1:48:07,  1.63it/s]
 84%|████████▍ | 54942/65536 [9:28:10<1:47:43,  1.64it/s]
 84%|████████▍ | 54943/65536 [9:28:11<1:45:45,  1.67it/s]
 84%|████████▍ | 54944/65536 [9:28:11<1:45:25,  1.67it/s]
 84%|████████▍ | 54945/65536 [9:28:12<1:48:30,  1.63it/s]
 84%|████████▍ | 54946/65536 [9:28:13<1:52:15,  1.57it/s]
 84%|████████▍ | 54947/65536 [9:28:13<1:53:04,  1.56it/s]
 84%|████████▍ | 54948/65536 [9:28:14<1:49:31,  1.61it/s]
 84%|████████▍ | 54949/65536 [9:28:15<1:50:54,  1.59it/s]
 84%|████████▍ | 54950/65536 [9:28:15<1:49:24,  1.61it/s]
 84%|████████▍ | 54951/65536 [9:28:16<1:56:40,  1.51it/s]
 84%|████████▍ | 54952/65536 [9:28:17<1:54:16,  1.54it/s]
 84%|████████▍ | 54953/65536 [9:28:17<1:51:02,  1.59it/s]
 84%|████████▍ | 54954/65536 [9:28:18<1:48:15,  1.63it/s]
 84%|████████▍ | 54955/65536 [9:28:18<1:47:37,  1.64it/s]
 84%|████████▍ | 54956/65536 [9:28:19<1:51:41,  1.58it/s]
 84%|████████▍ | 54957/65536 [9:28:20<1:51:38,  1.58it/s]
 84%|████████▍ | 54958/65536 [9:28:20<1:50:28,  1.60it/s]
 84%|████████▍ | 54959/65536 [9:28:21<1:48:18,  1.63it/s]
 84%|████████▍ | 54960/65536 [9:28:21<1:48:23,  1.63it/s]
                                                         
{'loss': 1.5412, 'learning_rate': 2.4568384963879027e-07, 'epoch': 3392.59}
+
 84%|████████▍ | 54960/65536 [9:28:21<1:48:23,  1.63it/s]
 84%|████████▍ | 54961/65536 [9:28:22<1:47:49,  1.63it/s]
 84%|████████▍ | 54962/65536 [9:28:23<1:45:30,  1.67it/s]
 84%|████████▍ | 54963/65536 [9:28:23<1:46:22,  1.66it/s]
 84%|████████▍ | 54964/65536 [9:28:24<1:48:34,  1.62it/s]
 84%|████████▍ | 54965/65536 [9:28:25<1:48:49,  1.62it/s]
 84%|████████▍ | 54966/65536 [9:28:25<1:49:13,  1.61it/s]
 84%|████████▍ | 54967/65536 [9:28:26<1:55:38,  1.52it/s]
 84%|████████▍ | 54968/65536 [9:28:27<1:53:28,  1.55it/s]
 84%|████████▍ | 54969/65536 [9:28:27<1:52:51,  1.56it/s]
 84%|████████▍ | 54970/65536 [9:28:28<1:55:15,  1.53it/s]
 84%|████████▍ | 54971/65536 [9:28:28<1:54:43,  1.53it/s]
 84%|████████▍ | 54972/65536 [9:28:29<1:55:18,  1.53it/s]
 84%|████████▍ | 54973/65536 [9:28:30<1:53:06,  1.56it/s]
 84%|████████▍ | 54974/65536 [9:28:30<1:52:12,  1.57it/s]
 84%|████████▍ | 54975/65536 [9:28:31<1:50:12,  1.60it/s]
 84%|████████▍ | 54976/65536 [9:28:32<1:48:39,  1.62it/s]
 84%|████████▍ | 54977/65536 [9:28:32<1:47:51,  1.63it/s]
 84%|████████▍ | 54978/65536 [9:28:33<1:46:16,  1.66it/s]
 84%|████████▍ | 54979/65536 [9:28:33<1:45:53,  1.66it/s]
 84%|████████▍ | 54980/65536 [9:28:34<1:46:39,  1.65it/s]
                                                         
{'loss': 1.5542, 'learning_rate': 2.454083506795641e-07, 'epoch': 3393.83}
+
 84%|████████▍ | 54980/65536 [9:28:34<1:46:39,  1.65it/s]
 84%|████████▍ | 54981/65536 [9:28:35<1:47:24,  1.64it/s]
 84%|████████▍ | 54982/65536 [9:28:35<1:45:31,  1.67it/s]
 84%|████████▍ | 54983/65536 [9:28:36<1:53:10,  1.55it/s]
 84%|████████▍ | 54984/65536 [9:28:37<1:53:55,  1.54it/s]
 84%|████████▍ | 54985/65536 [9:28:37<1:54:41,  1.53it/s]
 84%|████████▍ | 54986/65536 [9:28:38<1:52:45,  1.56it/s]
 84%|████████▍ | 54987/65536 [9:28:39<1:53:32,  1.55it/s]
 84%|████████▍ | 54988/65536 [9:28:39<1:52:01,  1.57it/s]
 84%|████████▍ | 54989/65536 [9:28:40<1:51:34,  1.58it/s]
 84%|████████▍ | 54990/65536 [9:28:40<1:48:51,  1.61it/s]
 84%|████████▍ | 54991/65536 [9:28:41<1:47:58,  1.63it/s]
 84%|████████▍ | 54992/65536 [9:28:42<1:47:06,  1.64it/s]
 84%|████████▍ | 54993/65536 [9:28:42<1:45:09,  1.67it/s]
 84%|████████▍ | 54994/65536 [9:28:43<1:45:03,  1.67it/s]
 84%|████████▍ | 54995/65536 [9:28:43<1:45:58,  1.66it/s]
 84%|████████▍ | 54996/65536 [9:28:44<1:44:27,  1.68it/s]
 84%|████████▍ | 54997/65536 [9:28:45<1:44:48,  1.68it/s]
 84%|████████▍ | 54998/65536 [9:28:45<1:45:10,  1.67it/s]
 84%|████████▍ | 54999/65536 [9:28:46<1:46:56,  1.64it/s]
 84%|████████▍ | 55000/65536 [9:28:46<1:49:21,  1.61it/s]
                                                         
{'loss': 1.5828, 'learning_rate': 2.4513285172033793e-07, 'epoch': 3395.06}
+
 84%|████████▍ | 55000/65536 [9:28:46<1:49:21,  1.61it/s]
 84%|████████▍ | 55001/65536 [9:28:47<1:50:20,  1.59it/s]
 84%|████████▍ | 55002/65536 [9:28:48<1:47:49,  1.63it/s]
 84%|████████▍ | 55003/65536 [9:28:48<1:48:55,  1.61it/s]
 84%|████████▍ | 55004/65536 [9:28:49<1:50:08,  1.59it/s]
 84%|████████▍ | 55005/65536 [9:28:50<1:53:23,  1.55it/s]
 84%|████████▍ | 55006/65536 [9:28:50<1:50:46,  1.58it/s]
 84%|████████▍ | 55007/65536 [9:28:51<1:50:42,  1.58it/s]
 84%|████████▍ | 55008/65536 [9:28:51<1:49:01,  1.61it/s]
 84%|████████▍ | 55009/65536 [9:28:52<1:47:14,  1.64it/s]
 84%|████████▍ | 55010/65536 [9:28:53<1:50:22,  1.59it/s]
 84%|████████▍ | 55011/65536 [9:28:53<1:49:29,  1.60it/s]
 84%|████████▍ | 55012/65536 [9:28:54<1:48:35,  1.62it/s]
 84%|████████▍ | 55013/65536 [9:28:55<1:49:35,  1.60it/s]
 84%|████████▍ | 55014/65536 [9:28:55<1:47:20,  1.63it/s]
 84%|████████▍ | 55015/65536 [9:28:56<1:47:30,  1.63it/s]
 84%|████████▍ | 55016/65536 [9:28:56<1:51:17,  1.58it/s]
 84%|████████▍ | 55017/65536 [9:28:57<1:48:50,  1.61it/s]
 84%|████████▍ | 55018/65536 [9:28:58<1:47:28,  1.63it/s]
 84%|████████▍ | 55019/65536 [9:28:58<1:49:11,  1.61it/s]
 84%|████████▍ | 55020/65536 [9:28:59<1:52:49,  1.55it/s]
                                                         
{'loss': 1.5865, 'learning_rate': 2.448573527611118e-07, 'epoch': 3396.3}
+
 84%|████████▍ | 55020/65536 [9:28:59<1:52:49,  1.55it/s]
 84%|████████▍ | 55021/65536 [9:29:00<1:53:47,  1.54it/s]
 84%|████████▍ | 55022/65536 [9:29:00<1:50:52,  1.58it/s]
 84%|████████▍ | 55023/65536 [9:29:01<1:47:52,  1.62it/s]
 84%|████████▍ | 55024/65536 [9:29:01<1:47:50,  1.62it/s]
 84%|████████▍ | 55025/65536 [9:29:02<1:46:00,  1.65it/s]
 84%|████████▍ | 55026/65536 [9:29:03<1:46:09,  1.65it/s]
 84%|████████▍ | 55027/65536 [9:29:03<1:44:43,  1.67it/s]
 84%|████████▍ | 55028/65536 [9:29:04<1:44:19,  1.68it/s]
 84%|████████▍ | 55029/65536 [9:29:04<1:45:26,  1.66it/s]
 84%|████████▍ | 55030/65536 [9:29:05<1:48:40,  1.61it/s]
 84%|████████▍ | 55031/65536 [9:29:06<1:48:05,  1.62it/s]
 84%|████████▍ | 55032/65536 [9:29:06<1:51:36,  1.57it/s]
 84%|████████▍ | 55033/65536 [9:29:07<1:53:08,  1.55it/s]
 84%|████████▍ | 55034/65536 [9:29:08<1:50:25,  1.59it/s]
 84%|████████▍ | 55035/65536 [9:29:08<1:48:47,  1.61it/s]
 84%|████████▍ | 55036/65536 [9:29:09<1:48:45,  1.61it/s]
 84%|████████▍ | 55037/65536 [9:29:09<1:47:13,  1.63it/s]
 84%|████████▍ | 55038/65536 [9:29:10<1:46:11,  1.65it/s]
 84%|████████▍ | 55039/65536 [9:29:11<1:47:39,  1.63it/s]
 84%|████████▍ | 55040/65536 [9:29:11<1:47:08,  1.63it/s]
                                                         
{'loss': 1.5806, 'learning_rate': 2.4458185380188565e-07, 'epoch': 3397.53}
+
 84%|████████▍ | 55040/65536 [9:29:11<1:47:08,  1.63it/s]
 84%|████████▍ | 55041/65536 [9:29:12<1:47:52,  1.62it/s]
 84%|████████▍ | 55042/65536 [9:29:12<1:46:48,  1.64it/s]
 84%|████████▍ | 55043/65536 [9:29:13<1:45:16,  1.66it/s]
 84%|████████▍ | 55044/65536 [9:29:14<1:47:44,  1.62it/s]
 84%|████████▍ | 55045/65536 [9:29:14<1:47:05,  1.63it/s]
 84%|████████▍ | 55046/65536 [9:29:15<1:46:16,  1.65it/s]
 84%|████████▍ | 55047/65536 [9:29:15<1:45:54,  1.65it/s]
 84%|████████▍ | 55048/65536 [9:29:16<1:50:49,  1.58it/s]
 84%|████████▍ | 55049/65536 [9:29:17<1:50:46,  1.58it/s]
 84%|████████▍ | 55050/65536 [9:29:17<1:51:25,  1.57it/s]
 84%|████████▍ | 55051/65536 [9:29:18<1:48:27,  1.61it/s]
 84%|████████▍ | 55052/65536 [9:29:19<1:45:48,  1.65it/s]
 84%|████████▍ | 55053/65536 [9:29:19<1:48:10,  1.62it/s]
 84%|████████▍ | 55054/65536 [9:29:20<1:46:16,  1.64it/s]
 84%|███████��▍ | 55055/65536 [9:29:20<1:47:22,  1.63it/s]
 84%|████████▍ | 55056/65536 [9:29:21<1:49:27,  1.60it/s]
 84%|████████▍ | 55057/65536 [9:29:22<1:47:35,  1.62it/s]
 84%|████████▍ | 55058/65536 [9:29:22<1:46:35,  1.64it/s]
 84%|████████▍ | 55059/65536 [9:29:23<1:49:19,  1.60it/s]
 84%|████████▍ | 55060/65536 [9:29:24<1:46:19,  1.64it/s]
                                                         
{'loss': 1.5568, 'learning_rate': 2.443063548426595e-07, 'epoch': 3398.77}
+
 84%|████████▍ | 55060/65536 [9:29:24<1:46:19,  1.64it/s]
 84%|████████▍ | 55061/65536 [9:29:24<1:46:40,  1.64it/s]
 84%|████████▍ | 55062/65536 [9:29:25<1:46:51,  1.63it/s]
 84%|████████▍ | 55063/65536 [9:29:25<1:45:11,  1.66it/s]
 84%|████████▍ | 55064/65536 [9:29:26<1:50:30,  1.58it/s]
 84%|████████▍ | 55065/65536 [9:29:27<1:50:08,  1.58it/s]
 84%|████████▍ | 55066/65536 [9:29:27<1:47:48,  1.62it/s]
 84%|████████▍ | 55067/65536 [9:29:28<1:47:36,  1.62it/s]
 84%|████████▍ | 55068/65536 [9:29:28<1:46:13,  1.64it/s]
 84%|████████▍ | 55069/65536 [9:29:29<1:46:05,  1.64it/s]
 84%|████████▍ | 55070/65536 [9:29:30<1:49:29,  1.59it/s]
 84%|████████▍ | 55071/65536 [9:29:30<1:54:38,  1.52it/s]
 84%|████████▍ | 55072/65536 [9:29:31<1:51:34,  1.56it/s]
 84%|████████▍ | 55073/65536 [9:29:32<1:49:02,  1.60it/s]
 84%|████████▍ | 55074/65536 [9:29:32<1:47:14,  1.63it/s]
 84%|████████▍ | 55075/65536 [9:29:33<1:45:54,  1.65it/s]
 84%|████████▍ | 55076/65536 [9:29:33<1:44:45,  1.66it/s]
 84%|████████▍ | 55077/65536 [9:29:34<1:45:48,  1.65it/s]
 84%|████████▍ | 55078/65536 [9:29:35<1:47:55,  1.61it/s]
 84%|████████▍ | 55079/65536 [9:29:35<1:46:43,  1.63it/s]
 84%|████████▍ | 55080/65536 [9:29:36<1:47:40,  1.62it/s]
                                                         
{'loss': 1.618, 'learning_rate': 2.4403085588343337e-07, 'epoch': 3400.0}
+
 84%|████████▍ | 55080/65536 [9:29:36<1:47:40,  1.62it/s]
 84%|████████▍ | 55081/65536 [9:29:37<1:53:50,  1.53it/s]
 84%|████████▍ | 55082/65536 [9:29:37<1:51:10,  1.57it/s]
 84%|████████▍ | 55083/65536 [9:29:38<1:49:53,  1.59it/s]
 84%|████████▍ | 55084/65536 [9:29:39<1:50:04,  1.58it/s]
 84%|████████▍ | 55085/65536 [9:29:39<1:50:24,  1.58it/s]
 84%|████████▍ | 55086/65536 [9:29:40<1:50:11,  1.58it/s]
 84%|████████▍ | 55087/65536 [9:29:41<1:54:57,  1.51it/s]
 84%|████████▍ | 55088/65536 [9:29:41<1:51:26,  1.56it/s]
 84%|████████▍ | 55089/65536 [9:29:42<1:51:07,  1.57it/s]
 84%|████████▍ | 55090/65536 [9:29:42<1:50:23,  1.58it/s]
 84%|████████▍ | 55091/65536 [9:29:43<1:47:24,  1.62it/s]
 84%|████████▍ | 55092/65536 [9:29:44<1:45:08,  1.66it/s]
 84%|████████▍ | 55093/65536 [9:29:44<1:45:11,  1.65it/s]
 84%|████████▍ | 55094/65536 [9:29:45<1:43:32,  1.68it/s]
 84%|████████▍ | 55095/65536 [9:29:45<1:41:53,  1.71it/s]
 84%|████████▍ | 55096/65536 [9:29:46<1:44:32,  1.66it/s]
 84%|████████▍ | 55097/65536 [9:29:47<1:49:11,  1.59it/s]
 84%|████████▍ | 55098/65536 [9:29:47<1:47:11,  1.62it/s]
 84%|████████▍ | 55099/65536 [9:29:48<1:47:50,  1.61it/s]
 84%|████████▍ | 55100/65536 [9:29:48<1:47:23,  1.62it/s]
                                                         
{'loss': 1.5494, 'learning_rate': 2.437553569242072e-07, 'epoch': 3401.23}
+
 84%|████████▍ | 55100/65536 [9:29:48<1:47:23,  1.62it/s]
 84%|████████▍ | 55101/65536 [9:29:49<1:46:50,  1.63it/s]
 84%|████████▍ | 55102/65536 [9:29:50<1:44:56,  1.66it/s]
 84%|████████▍ | 55103/65536 [9:29:50<1:45:41,  1.65it/s]
 84%|████████▍ | 55104/65536 [9:29:51<1:46:14,  1.64it/s]
 84%|████████▍ | 55105/65536 [9:29:51<1:45:43,  1.64it/s]
 84%|████████▍ | 55106/65536 [9:29:52<1:47:11,  1.62it/s]
 84%|████████▍ | 55107/65536 [9:29:53<1:50:16,  1.58it/s]
 84%|████████▍ | 55108/65536 [9:29:53<1:49:36,  1.59it/s]
 84%|████████▍ | 55109/65536 [9:29:54<1:49:26,  1.59it/s]
 84%|████████▍ | 55110/65536 [9:29:55<1:49:28,  1.59it/s]
 84%|████████▍ | 55111/65536 [9:29:55<1:47:25,  1.62it/s]
 84%|████████▍ | 55112/65536 [9:29:56<1:44:57,  1.66it/s]
 84%|████████▍ | 55113/65536 [9:29:56<1:48:15,  1.60it/s]
 84%|████████▍ | 55114/65536 [9:29:57<1:49:08,  1.59it/s]
 84%|████████▍ | 55115/65536 [9:29:58<1:48:34,  1.60it/s]
 84%|████████▍ | 55116/65536 [9:29:58<1:51:48,  1.55it/s]
 84%|████████▍ | 55117/65536 [9:29:59<1:49:03,  1.59it/s]
 84%|████████▍ | 55118/65536 [9:30:00<1:47:58,  1.61it/s]
 84%|████████▍ | 55119/65536 [9:30:00<1:45:31,  1.65it/s]
 84%|████████▍ | 55120/65536 [9:30:01<1:46:25,  1.63it/s]
                                                         
{'loss': 1.5472, 'learning_rate': 2.43479857964981e-07, 'epoch': 3402.47}
+
 84%|████████▍ | 55120/65536 [9:30:01<1:46:25,  1.63it/s]
 84%|████████▍ | 55121/65536 [9:30:02<1:50:17,  1.57it/s]
 84%|████████▍ | 55122/65536 [9:30:02<1:48:14,  1.60it/s]
 84%|████████▍ | 55123/65536 [9:30:03<1:46:44,  1.63it/s]
 84%|████████▍ | 55124/65536 [9:30:03<1:50:59,  1.56it/s]
 84%|████████▍ | 55125/65536 [9:30:04<1:48:36,  1.60it/s]
 84%|████████▍ | 55126/65536 [9:30:05<1:47:37,  1.61it/s]
 84%|████████▍ | 55127/65536 [9:30:05<1:48:23,  1.60it/s]
 84%|████████▍ | 55128/65536 [9:30:06<1:46:34,  1.63it/s]
 84%|████████▍ | 55129/65536 [9:30:06<1:49:49,  1.58it/s]
 84%|████████▍ | 55130/65536 [9:30:07<1:49:30,  1.58it/s]
 84%|████████▍ | 55131/65536 [9:30:08<1:50:48,  1.56it/s]
 84%|████████▍ | 55132/65536 [9:30:08<1:50:52,  1.56it/s]
 84%|████████▍ | 55133/65536 [9:30:09<1:49:41,  1.58it/s]
 84%|████████▍ | 55134/65536 [9:30:10<1:50:12,  1.57it/s]
 84%|████████▍ | 55135/65536 [9:30:10<1:47:55,  1.61it/s]
 84%|████████▍ | 55136/65536 [9:30:11<1:45:54,  1.64it/s]
 84%|████████▍ | 55137/65536 [9:30:11<1:45:40,  1.64it/s]
 84%|████████▍ | 55138/65536 [9:30:12<1:48:28,  1.60it/s]
 84%|████████▍ | 55139/65536 [9:30:13<1:45:41,  1.64it/s]
 84%|████████▍ | 55140/65536 [9:30:13<1:44:32,  1.66it/s]
                                                         
{'loss': 1.5314, 'learning_rate': 2.4320435900575484e-07, 'epoch': 3403.7}
+
 84%|████████▍ | 55140/65536 [9:30:13<1:44:32,  1.66it/s]
 84%|████████▍ | 55141/65536 [9:30:14<1:45:59,  1.63it/s]
 84%|████████▍ | 55142/65536 [9:30:15<1:45:20,  1.64it/s]
 84%|████████▍ | 55143/65536 [9:30:15<1:46:48,  1.62it/s]
 84%|████████▍ | 55144/65536 [9:30:16<1:45:59,  1.63it/s]
 84%|████████▍ | 55145/65536 [9:30:16<1:48:38,  1.59it/s]
 84%|████████▍ | 55146/65536 [9:30:17<1:47:30,  1.61it/s]
 84%|████████▍ | 55147/65536 [9:30:18<1:45:43,  1.64it/s]
 84%|████████▍ | 55148/65536 [9:30:18<1:46:38,  1.62it/s]
 84%|████████▍ | 55149/65536 [9:30:19<1:48:08,  1.60it/s]
 84%|████████▍ | 55150/65536 [9:30:19<1:46:46,  1.62it/s]
 84%|████████▍ | 55151/65536 [9:30:20<1:45:28,  1.64it/s]
 84%|████████▍ | 55152/65536 [9:30:21<1:45:32,  1.64it/s]
 84%|████████▍ | 55153/65536 [9:30:21<1:47:35,  1.61it/s]
 84%|████████▍ | 55154/65536 [9:30:22<1:47:31,  1.61it/s]
 84%|████████▍ | 55155/65536 [9:30:23<1:48:05,  1.60it/s]
 84%|████████▍ | 55156/65536 [9:30:23<1:46:38,  1.62it/s]
 84%|████████▍ | 55157/65536 [9:30:24<1:47:00,  1.62it/s]
 84%|████████▍ | 55158/65536 [9:30:24<1:48:30,  1.59it/s]
 84%|████████▍ | 55159/65536 [9:30:25<1:48:52,  1.59it/s]
 84%|████████▍ | 55160/65536 [9:30:26<1:46:46,  1.62it/s]
                                                         
{'loss': 1.5775, 'learning_rate': 2.429288600465287e-07, 'epoch': 3404.94}
+
 84%|████████▍ | 55160/65536 [9:30:26<1:46:46,  1.62it/s]
 84%|████████▍ | 55161/65536 [9:30:26<1:46:34,  1.62it/s]
 84%|████████▍ | 55162/65536 [9:30:27<1:51:04,  1.56it/s]
 84%|████████▍ | 55163/65536 [9:30:28<1:48:56,  1.59it/s]
 84%|████████▍ | 55164/65536 [9:30:28<1:52:06,  1.54it/s]
 84%|████████▍ | 55165/65536 [9:30:29<1:50:52,  1.56it/s]
 84%|████████▍ | 55166/65536 [9:30:30<1:48:25,  1.59it/s]
 84%|████████▍ | 55167/65536 [9:30:30<1:48:35,  1.59it/s]
 84%|████████▍ | 55168/65536 [9:30:31<1:48:34,  1.59it/s]
 84%|████████▍ | 55169/65536 [9:30:31<1:47:23,  1.61it/s]
 84%|████████▍ | 55170/65536 [9:30:32<1:45:58,  1.63it/s]
 84%|████████▍ | 55171/65536 [9:30:33<1:45:32,  1.64it/s]
 84%|████████▍ | 55172/65536 [9:30:33<1:50:39,  1.56it/s]
 84%|████████▍ | 55173/65536 [9:30:34<1:46:25,  1.62it/s]
 84%|████████▍ | 55174/65536 [9:30:34<1:44:41,  1.65it/s]
 84%|████████▍ | 55175/65536 [9:30:35<1:45:05,  1.64it/s]
 84%|████████▍ | 55176/65536 [9:30:36<1:46:46,  1.62it/s]
 84%|████████▍ | 55177/65536 [9:30:36<1:45:21,  1.64it/s]
 84%|████████▍ | 55178/65536 [9:30:37<1:51:28,  1.55it/s]
 84%|████████▍ | 55179/65536 [9:30:38<1:49:15,  1.58it/s]
 84%|████████▍ | 55180/65536 [9:30:38<1:47:19,  1.61it/s]
                                                         
{'loss': 1.5817, 'learning_rate': 2.4265336108730256e-07, 'epoch': 3406.17}
+
 84%|████████▍ | 55180/65536 [9:30:38<1:47:19,  1.61it/s]
 84%|████████▍ | 55181/65536 [9:30:39<1:45:46,  1.63it/s]
 84%|████████▍ | 55182/65536 [9:30:39<1:47:50,  1.60it/s]
 84%|████████▍ | 55183/65536 [9:30:40<1:47:15,  1.61it/s]
 84%|████████▍ | 55184/65536 [9:30:41<1:48:47,  1.59it/s]
 84%|████████▍ | 55185/65536 [9:30:41<1:51:52,  1.54it/s]
 84%|████████▍ | 55186/65536 [9:30:42<1:50:38,  1.56it/s]
 84%|████████▍ | 55187/65536 [9:30:43<1:47:08,  1.61it/s]
 84%|████████▍ | 55188/65536 [9:30:43<1:45:58,  1.63it/s]
 84%|████████▍ | 55189/65536 [9:30:44<1:46:10,  1.62it/s]
 84%|████████▍ | 55190/65536 [9:30:44<1:46:16,  1.62it/s]
 84%|████████▍ | 55191/65536 [9:30:45<1:45:03,  1.64it/s]
 84%|████████▍ | 55192/65536 [9:30:46<1:45:11,  1.64it/s]
 84%|████████▍ | 55193/65536 [9:30:46<1:43:19,  1.67it/s]
 84%|████████▍ | 55194/65536 [9:30:47<1:46:19,  1.62it/s]
 84%|████████▍ | 55195/65536 [9:30:48<1:49:32,  1.57it/s]
 84%|████████▍ | 55196/65536 [9:30:48<1:48:29,  1.59it/s]
 84%|████████▍ | 55197/65536 [9:30:49<1:51:09,  1.55it/s]
 84%|████████▍ | 55198/65536 [9:30:49<1:50:13,  1.56it/s]
 84%|████████▍ | 55199/65536 [9:30:50<1:48:08,  1.59it/s]
 84%|████████▍ | 55200/65536 [9:30:51<1:46:45,  1.61it/s]
                                                         
{'loss': 1.5537, 'learning_rate': 2.423778621280764e-07, 'epoch': 3407.41}
+
 84%|████████▍ | 55200/65536 [9:30:51<1:46:45,  1.61it/s]
 84%|████████▍ | 55201/65536 [9:30:51<1:48:10,  1.59it/s]
 84%|████████▍ | 55202/65536 [9:30:52<1:46:17,  1.62it/s]
 84%|████████▍ | 55203/65536 [9:30:53<1:45:18,  1.64it/s]
 84%|████████▍ | 55204/65536 [9:30:53<1:44:06,  1.65it/s]
 84%|████████▍ | 55205/65536 [9:30:54<1:44:01,  1.66it/s]
 84%|████████▍ | 55206/65536 [9:30:54<1:44:51,  1.64it/s]
 84%|████████▍ | 55207/65536 [9:30:55<1:44:51,  1.64it/s]
 84%|████████▍ | 55208/65536 [9:30:56<1:43:09,  1.67it/s]
 84%|████████▍ | 55209/65536 [9:30:56<1:41:32,  1.70it/s]
 84%|████████▍ | 55210/65536 [9:30:57<1:43:41,  1.66it/s]
 84%|████████▍ | 55211/65536 [9:30:57<1:42:58,  1.67it/s]
 84%|████████▍ | 55212/65536 [9:30:58<1:40:53,  1.71it/s]
 84%|████████▍ | 55213/65536 [9:30:58<1:42:21,  1.68it/s]
 84%|████████▍ | 55214/65536 [9:30:59<1:43:39,  1.66it/s]
 84%|████████▍ | 55215/65536 [9:31:00<1:44:06,  1.65it/s]
 84%|████████▍ | 55216/65536 [9:31:00<1:44:49,  1.64it/s]
 84%|████████▍ | 55217/65536 [9:31:01<1:44:18,  1.65it/s]
 84%|████████▍ | 55218/65536 [9:31:02<1:46:39,  1.61it/s]
 84%|████████▍ | 55219/65536 [9:31:02<1:46:46,  1.61it/s]
 84%|████████▍ | 55220/65536 [9:31:03<1:48:38,  1.58it/s]
                                                         
{'loss': 1.5873, 'learning_rate': 2.421023631688503e-07, 'epoch': 3408.64}
+
 84%|████████▍ | 55220/65536 [9:31:03<1:48:38,  1.58it/s]
 84%|████████▍ | 55221/65536 [9:31:03<1:48:19,  1.59it/s]
 84%|████████▍ | 55222/65536 [9:31:04<1:46:02,  1.62it/s]
 84%|████████▍ | 55223/65536 [9:31:05<1:48:12,  1.59it/s]
 84%|████████▍ | 55224/65536 [9:31:05<1:47:28,  1.60it/s]
 84%|████████▍ | 55225/65536 [9:31:06<1:47:32,  1.60it/s]
 84%|████████▍ | 55226/65536 [9:31:07<1:48:32,  1.58it/s]
 84%|████████▍ | 55227/65536 [9:31:07<1:46:01,  1.62it/s]
 84%|████████▍ | 55228/65536 [9:31:08<1:45:42,  1.63it/s]
 84%|████████▍ | 55229/65536 [9:31:08<1:46:12,  1.62it/s]
 84%|████████▍ | 55230/65536 [9:31:09<1:45:18,  1.63it/s]
 84%|████████▍ | 55231/65536 [9:31:10<1:44:11,  1.65it/s]
 84%|████████▍ | 55232/65536 [9:31:10<1:47:56,  1.59it/s]
 84%|████████▍ | 55233/65536 [9:31:11<1:50:40,  1.55it/s]
 84%|████████▍ | 55234/65536 [9:31:12<1:47:33,  1.60it/s]
 84%|████████▍ | 55235/65536 [9:31:12<1:48:01,  1.59it/s]
 84%|████████▍ | 55236/65536 [9:31:13<1:46:30,  1.61it/s]
 84%|████████▍ | 55237/65536 [9:31:13<1:45:09,  1.63it/s]
 84%|████████▍ | 55238/65536 [9:31:14<1:46:11,  1.62it/s]
 84%|████████▍ | 55239/65536 [9:31:15<1:47:21,  1.60it/s]
 84%|████████▍ | 55240/65536 [9:31:15<1:46:06,  1.62it/s]
                                                         
{'loss': 1.5728, 'learning_rate': 2.4182686420962413e-07, 'epoch': 3409.88}
+
 84%|████████▍ | 55240/65536 [9:31:15<1:46:06,  1.62it/s]
 84%|████████▍ | 55241/65536 [9:31:16<1:47:02,  1.60it/s]
 84%|████████▍ | 55242/65536 [9:31:17<1:45:26,  1.63it/s]
 84%|████████▍ | 55243/65536 [9:31:17<1:50:00,  1.56it/s]
 84%|████████▍ | 55244/65536 [9:31:18<1:47:13,  1.60it/s]
 84%|████████▍ | 55245/65536 [9:31:18<1:47:09,  1.60it/s]
 84%|████████▍ | 55246/65536 [9:31:19<1:44:04,  1.65it/s]
 84%|████████▍ | 55247/65536 [9:31:20<1:46:36,  1.61it/s]
 84%|████████▍ | 55248/65536 [9:31:20<1:46:05,  1.62it/s]
 84%|████████▍ | 55249/65536 [9:31:21<1:45:38,  1.62it/s]
 84%|████████▍ | 55250/65536 [9:31:21<1:44:10,  1.65it/s]
 84%|████████▍ | 55251/65536 [9:31:22<1:45:15,  1.63it/s]
 84%|████████▍ | 55252/65536 [9:31:23<1:43:22,  1.66it/s]
 84%|████████▍ | 55253/65536 [9:31:23<1:44:29,  1.64it/s]
 84%|████████▍ | 55254/65536 [9:31:24<1:43:59,  1.65it/s]
 84%|████████▍ | 55255/65536 [9:31:24<1:43:25,  1.66it/s]
 84%|████████▍ | 55256/65536 [9:31:25<1:45:11,  1.63it/s]
 84%|████████▍ | 55257/65536 [9:31:26<1:48:30,  1.58it/s]
 84%|████████▍ | 55258/65536 [9:31:26<1:45:11,  1.63it/s]
 84%|████████▍ | 55259/65536 [9:31:27<1:48:19,  1.58it/s]
 84%|████████▍ | 55260/65536 [9:31:28<1:47:24,  1.59it/s]
                                                         
{'loss': 1.5896, 'learning_rate': 2.415513652503979e-07, 'epoch': 3411.11}
+
 84%|████████▍ | 55260/65536 [9:31:28<1:47:24,  1.59it/s]
 84%|████████▍ | 55261/65536 [9:31:28<1:50:05,  1.56it/s]
 84%|████████▍ | 55262/65536 [9:31:29<1:47:13,  1.60it/s]
 84%|████████▍ | 55263/65536 [9:31:30<1:47:45,  1.59it/s]
 84%|████████▍ | 55264/65536 [9:31:30<1:46:16,  1.61it/s]
 84%|████████▍ | 55265/65536 [9:31:31<1:47:24,  1.59it/s]
 84%|████████▍ | 55266/65536 [9:31:31<1:46:16,  1.61it/s]
 84%|████████▍ | 55267/65536 [9:31:32<1:46:46,  1.60it/s]
 84%|████████▍ | 55268/65536 [9:31:33<1:45:22,  1.62it/s]
 84%|████████▍ | 55269/65536 [9:31:33<1:44:27,  1.64it/s]
 84%|████████▍ | 55270/65536 [9:31:34<1:44:54,  1.63it/s]
 84%|████████▍ | 55271/65536 [9:31:34<1:41:42,  1.68it/s]
 84%|████████▍ | 55272/65536 [9:31:35<1:45:48,  1.62it/s]
 84%|████████▍ | 55273/65536 [9:31:36<1:46:52,  1.60it/s]
 84%|████████▍ | 55274/65536 [9:31:36<1:45:53,  1.62it/s]
 84%|████████▍ | 55275/65536 [9:31:37<1:49:19,  1.56it/s]
 84%|████████▍ | 55276/65536 [9:31:38<1:45:47,  1.62it/s]
 84%|████████▍ | 55277/65536 [9:31:38<1:46:26,  1.61it/s]
 84%|████████▍ | 55278/65536 [9:31:39<1:44:41,  1.63it/s]
 84%|████████▍ | 55279/65536 [9:31:39<1:45:17,  1.62it/s]
 84%|████████▍ | 55280/65536 [9:31:40<1:45:48,  1.62it/s]
                                                         
{'loss': 1.5592, 'learning_rate': 2.4127586629117174e-07, 'epoch': 3412.35}
+
 84%|████████▍ | 55280/65536 [9:31:40<1:45:48,  1.62it/s]
 84%|████████▍ | 55281/65536 [9:31:41<1:48:49,  1.57it/s]
 84%|████████▍ | 55282/65536 [9:31:41<1:47:20,  1.59it/s]
 84%|████████▍ | 55283/65536 [9:31:42<1:45:44,  1.62it/s]
 84%|████████▍ | 55284/65536 [9:31:43<1:48:44,  1.57it/s]
 84%|████████▍ | 55285/65536 [9:31:43<1:48:09,  1.58it/s]
 84%|████████▍ | 55286/65536 [9:31:44<1:47:02,  1.60it/s]
 84%|████████▍ | 55287/65536 [9:31:44<1:44:35,  1.63it/s]
 84%|████████▍ | 55288/65536 [9:31:45<1:43:23,  1.65it/s]
 84%|████████▍ | 55289/65536 [9:31:46<1:45:08,  1.62it/s]
 84%|████████▍ | 55290/65536 [9:31:46<1:44:30,  1.63it/s]
 84%|████████▍ | 55291/65536 [9:31:47<1:48:02,  1.58it/s]
 84%|████████▍ | 55292/65536 [9:31:48<1:47:20,  1.59it/s]
 84%|████████▍ | 55293/65536 [9:31:48<1:44:55,  1.63it/s]
 84%|████████▍ | 55294/65536 [9:31:49<1:44:47,  1.63it/s]
 84%|████████▍ | 55295/65536 [9:31:49<1:44:30,  1.63it/s]
 84%|████████▍ | 55296/65536 [9:31:50<1:45:27,  1.62it/s]
 84%|████████▍ | 55297/65536 [9:31:51<1:47:58,  1.58it/s]
 84%|████████▍ | 55298/65536 [9:31:51<1:49:34,  1.56it/s]
 84%|████████▍ | 55299/65536 [9:31:52<1:48:18,  1.58it/s]
 84%|████████▍ | 55300/65536 [9:31:53<1:46:46,  1.60it/s]
                                                         
{'loss': 1.5229, 'learning_rate': 2.410003673319456e-07, 'epoch': 3413.58}
+
 84%|████████▍ | 55300/65536 [9:31:53<1:46:46,  1.60it/s]
 84%|████████▍ | 55301/65536 [9:31:53<1:45:45,  1.61it/s]
 84%|████████▍ | 55302/65536 [9:31:54<1:45:43,  1.61it/s]
 84%|████████▍ | 55303/65536 [9:31:54<1:44:19,  1.63it/s]
 84%|████████▍ | 55304/65536 [9:31:55<1:42:57,  1.66it/s]
 84%|████████▍ | 55305/65536 [9:31:56<1:45:56,  1.61it/s]
 84%|████████▍ | 55306/65536 [9:31:56<1:44:02,  1.64it/s]
 84%|████████▍ | 55307/65536 [9:31:57<1:46:41,  1.60it/s]
 84%|████████▍ | 55308/65536 [9:31:57<1:46:36,  1.60it/s]
 84%|████████▍ | 55309/65536 [9:31:58<1:44:56,  1.62it/s]
 84%|████████▍ | 55310/65536 [9:31:59<1:47:02,  1.59it/s]
 84%|████████▍ | 55311/65536 [9:31:59<1:47:54,  1.58it/s]
 84%|████████▍ | 55312/65536 [9:32:00<1:45:36,  1.61it/s]
 84%|████████▍ | 55313/65536 [9:32:01<1:49:21,  1.56it/s]
 84%|████████▍ | 55314/65536 [9:32:01<1:49:46,  1.55it/s]
 84%|████████▍ | 55315/65536 [9:32:02<1:47:56,  1.58it/s]
 84%|████████▍ | 55316/65536 [9:32:03<1:45:39,  1.61it/s]
 84%|████████▍ | 55317/65536 [9:32:03<1:48:18,  1.57it/s]
 84%|████████▍ | 55318/65536 [9:32:04<1:45:12,  1.62it/s]
 84%|████████▍ | 55319/65536 [9:32:04<1:43:18,  1.65it/s]
 84%|████████▍ | 55320/65536 [9:32:05<1:43:20,  1.65it/s]
                                                         
{'loss': 1.5571, 'learning_rate': 2.4072486837271946e-07, 'epoch': 3414.81}
+
 84%|████████▍ | 55320/65536 [9:32:05<1:43:20,  1.65it/s]
 84%|████████▍ | 55321/65536 [9:32:06<1:42:46,  1.66it/s]
 84%|████████▍ | 55322/65536 [9:32:06<1:41:42,  1.67it/s]
 84%|████████▍ | 55323/65536 [9:32:07<1:44:01,  1.64it/s]
 84%|████████▍ | 55324/65536 [9:32:08<2:16:03,  1.25it/s]
 84%|████████▍ | 55325/65536 [9:32:09<2:08:08,  1.33it/s]
 84%|████████▍ | 55326/65536 [9:32:09<2:00:51,  1.41it/s]
 84%|████████▍ | 55327/65536 [9:32:10<1:55:04,  1.48it/s]
 84%|████████▍ | 55328/65536 [9:32:10<1:51:50,  1.52it/s]
 84%|████████▍ | 55329/65536 [9:32:11<1:51:35,  1.52it/s]
 84%|████████▍ | 55330/65536 [9:32:12<1:49:54,  1.55it/s]
 84%|████████▍ | 55331/65536 [9:32:12<1:47:11,  1.59it/s]
 84%|████████▍ | 55332/65536 [9:32:13<1:46:08,  1.60it/s]
 84%|████████▍ | 55333/65536 [9:32:13<1:41:36,  1.67it/s]
 84%|████████▍ | 55334/65536 [9:32:14<1:43:19,  1.65it/s]
 84%|████████▍ | 55335/65536 [9:32:15<1:42:23,  1.66it/s]
 84%|████████▍ | 55336/65536 [9:32:15<1:42:57,  1.65it/s]
 84%|████████▍ | 55337/65536 [9:32:16<1:43:58,  1.63it/s]
 84%|████████▍ | 55338/65536 [9:32:17<1:44:53,  1.62it/s]
 84%|████████▍ | 55339/65536 [9:32:17<1:44:35,  1.62it/s]
 84%|████████▍ | 55340/65536 [9:32:18<1:50:24,  1.54it/s]
                                                         
{'loss': 1.5525, 'learning_rate': 2.404493694134933e-07, 'epoch': 3416.05}
+
 84%|████████▍ | 55340/65536 [9:32:18<1:50:24,  1.54it/s]
 84%|████████▍ | 55341/65536 [9:32:19<1:49:33,  1.55it/s]
 84%|████████▍ | 55342/65536 [9:32:19<1:49:32,  1.55it/s]
 84%|████████▍ | 55343/65536 [9:32:20<1:48:40,  1.56it/s]
 84%|████████▍ | 55344/65536 [9:32:20<1:47:38,  1.58it/s]
 84%|████████▍ | 55345/65536 [9:32:21<1:44:55,  1.62it/s]
 84%|████████▍ | 55346/65536 [9:32:22<1:44:28,  1.63it/s]
 84%|████████▍ | 55347/65536 [9:32:22<1:42:03,  1.66it/s]
 84%|████████▍ | 55348/65536 [9:32:23<1:43:55,  1.63it/s]
 84%|████████▍ | 55349/65536 [9:32:23<1:43:13,  1.64it/s]
 84%|████████▍ | 55350/65536 [9:32:24<1:45:07,  1.62it/s]
 84%|████████▍ | 55351/65536 [9:32:25<1:44:06,  1.63it/s]
 84%|████████▍ | 55352/65536 [9:32:25<1:42:45,  1.65it/s]
 84%|████████▍ | 55353/65536 [9:32:26<1:44:39,  1.62it/s]
 84%|████████▍ | 55354/65536 [9:32:27<1:43:59,  1.63it/s]
 84%|████████▍ | 55355/65536 [9:32:27<1:45:23,  1.61it/s]
 84%|████████▍ | 55356/65536 [9:32:28<1:47:51,  1.57it/s]
 84%|████████▍ | 55357/65536 [9:32:28<1:47:00,  1.59it/s]
 84%|████████▍ | 55358/65536 [9:32:29<1:44:50,  1.62it/s]
 84%|████████▍ | 55359/65536 [9:32:30<1:46:22,  1.59it/s]
 84%|████████▍ | 55360/65536 [9:32:30<1:46:31,  1.59it/s]
                                                         
{'loss': 1.5809, 'learning_rate': 2.401738704542672e-07, 'epoch': 3417.28}
+
 84%|████████▍ | 55360/65536 [9:32:30<1:46:31,  1.59it/s]
 84%|████████▍ | 55361/65536 [9:32:31<1:52:14,  1.51it/s]
 84%|████████▍ | 55362/65536 [9:32:32<1:49:04,  1.55it/s]
 84%|████████▍ | 55363/65536 [9:32:32<1:47:42,  1.57it/s]
 84%|████████▍ | 55364/65536 [9:32:33<1:46:40,  1.59it/s]
 84%|████████▍ | 55365/65536 [9:32:34<1:48:40,  1.56it/s]
 84%|████████▍ | 55366/65536 [9:32:34<1:47:14,  1.58it/s]
 84%|████████▍ | 55367/65536 [9:32:35<1:45:36,  1.60it/s]
 84%|████████▍ | 55368/65536 [9:32:35<1:44:20,  1.62it/s]
 84%|████████▍ | 55369/65536 [9:32:36<1:45:33,  1.61it/s]
 84%|████████▍ | 55370/65536 [9:32:37<1:50:03,  1.54it/s]
 84%|████████▍ | 55371/65536 [9:32:37<1:48:16,  1.56it/s]
 84%|████████▍ | 55372/65536 [9:32:38<1:52:05,  1.51it/s]
 84%|████████▍ | 55373/65536 [9:32:39<1:52:14,  1.51it/s]
 84%|████████▍ | 55374/65536 [9:32:39<1:51:46,  1.52it/s]
 84%|████████▍ | 55375/65536 [9:32:40<1:49:56,  1.54it/s]
 84%|████████▍ | 55376/65536 [9:32:41<1:46:46,  1.59it/s]
 84%|████████▍ | 55377/65536 [9:32:41<1:48:22,  1.56it/s]
 85%|████████▍ | 55378/65536 [9:32:42<1:45:27,  1.61it/s]
 85%|████████▍ | 55379/65536 [9:32:42<1:45:48,  1.60it/s]
 85%|████████▍ | 55380/65536 [9:32:43<1:45:06,  1.61it/s]
                                                         
{'loss': 1.5348, 'learning_rate': 2.3989837149504104e-07, 'epoch': 3418.52}
+
 85%|████████▍ | 55380/65536 [9:32:43<1:45:06,  1.61it/s]
 85%|████████▍ | 55381/65536 [9:32:44<1:45:54,  1.60it/s]
 85%|████████▍ | 55382/65536 [9:32:44<1:43:12,  1.64it/s]
 85%|████████▍ | 55383/65536 [9:32:45<1:44:37,  1.62it/s]
 85%|████████▍ | 55384/65536 [9:32:46<1:47:25,  1.57it/s]
 85%|████████▍ | 55385/65536 [9:32:46<1:45:26,  1.60it/s]
 85%|████████▍ | 55386/65536 [9:32:47<1:44:06,  1.62it/s]
 85%|████████▍ | 55387/65536 [9:32:47<1:45:00,  1.61it/s]
 85%|████████▍ | 55388/65536 [9:32:48<1:46:46,  1.58it/s]
 85%|████████▍ | 55389/65536 [9:32:49<1:46:47,  1.58it/s]
 85%|████████▍ | 55390/65536 [9:32:49<1:44:49,  1.61it/s]
 85%|████████▍ | 55391/65536 [9:32:50<1:47:03,  1.58it/s]
 85%|████████▍ | 55392/65536 [9:32:51<1:44:20,  1.62it/s]
 85%|████████▍ | 55393/65536 [9:32:51<1:43:11,  1.64it/s]
 85%|████████▍ | 55394/65536 [9:32:52<1:41:31,  1.66it/s]
 85%|████████▍ | 55395/65536 [9:32:52<1:42:55,  1.64it/s]
 85%|████████▍ | 55396/65536 [9:32:53<1:45:02,  1.61it/s]
 85%|████████▍ | 55397/65536 [9:32:54<1:47:27,  1.57it/s]
 85%|████████▍ | 55398/65536 [9:32:54<1:45:41,  1.60it/s]
 85%|████████▍ | 55399/65536 [9:32:55<1:49:40,  1.54it/s]
 85%|████████▍ | 55400/65536 [9:32:56<1:47:16,  1.57it/s]
                                                         
{'loss': 1.5412, 'learning_rate': 2.396228725358148e-07, 'epoch': 3419.75}
+
 85%|████████▍ | 55400/65536 [9:32:56<1:47:16,  1.57it/s]
 85%|████████▍ | 55401/65536 [9:32:56<1:47:33,  1.57it/s]
 85%|████████▍ | 55402/65536 [9:32:57<1:48:17,  1.56it/s]
 85%|████████▍ | 55403/65536 [9:32:57<1:47:47,  1.57it/s]
 85%|████████▍ | 55404/65536 [9:32:58<1:46:33,  1.58it/s]
 85%|████████▍ | 55405/65536 [9:32:59<1:52:02,  1.51it/s]
 85%|████████▍ | 55406/65536 [9:32:59<1:48:23,  1.56it/s]
 85%|████████▍ | 55407/65536 [9:33:00<1:48:02,  1.56it/s]
 85%|████████▍ | 55408/65536 [9:33:01<1:44:59,  1.61it/s]
 85%|████████▍ | 55409/65536 [9:33:01<1:43:44,  1.63it/s]
 85%|████████▍ | 55410/65536 [9:33:02<1:42:39,  1.64it/s]
 85%|████████▍ | 55411/65536 [9:33:02<1:40:25,  1.68it/s]
 85%|████████▍ | 55412/65536 [9:33:03<1:44:14,  1.62it/s]
 85%|████████▍ | 55413/65536 [9:33:04<1:45:46,  1.59it/s]
 85%|████████▍ | 55414/65536 [9:33:04<1:45:36,  1.60it/s]
 85%|████████▍ | 55415/65536 [9:33:05<1:44:57,  1.61it/s]
 85%|████████▍ | 55416/65536 [9:33:06<1:44:13,  1.62it/s]
 85%|████████▍ | 55417/65536 [9:33:06<1:42:12,  1.65it/s]
 85%|████████▍ | 55418/65536 [9:33:07<1:42:18,  1.65it/s]
 85%|████████▍ | 55419/65536 [9:33:07<1:41:41,  1.66it/s]
 85%|████████▍ | 55420/65536 [9:33:08<1:44:03,  1.62it/s]
                                                         
{'loss': 1.5556, 'learning_rate': 2.3934737357658865e-07, 'epoch': 3420.99}
+
 85%|████████▍ | 55420/65536 [9:33:08<1:44:03,  1.62it/s]
 85%|████████▍ | 55421/65536 [9:33:09<1:48:28,  1.55it/s]
 85%|████████▍ | 55422/65536 [9:33:09<1:48:00,  1.56it/s]
 85%|████████▍ | 55423/65536 [9:33:10<1:45:53,  1.59it/s]
 85%|████████▍ | 55424/65536 [9:33:11<1:47:01,  1.57it/s]
 85%|████████▍ | 55425/65536 [9:33:11<1:45:12,  1.60it/s]
 85%|████████▍ | 55426/65536 [9:33:12<1:45:40,  1.59it/s]
 85%|████████▍ | 55427/65536 [9:33:12<1:45:20,  1.60it/s]
 85%|████████▍ | 55428/65536 [9:33:13<1:44:04,  1.62it/s]
 85%|████████▍ | 55429/65536 [9:33:14<1:43:00,  1.64it/s]
 85%|████████▍ | 55430/65536 [9:33:14<1:42:01,  1.65it/s]
 85%|████████▍ | 55431/65536 [9:33:15<1:41:46,  1.65it/s]
 85%|████████▍ | 55432/65536 [9:33:16<1:47:10,  1.57it/s]
 85%|████████▍ | 55433/65536 [9:33:16<1:45:23,  1.60it/s]
 85%|████████▍ | 55434/65536 [9:33:17<1:45:41,  1.59it/s]
 85%|████████▍ | 55435/65536 [9:33:17<1:43:55,  1.62it/s]
 85%|████████▍ | 55436/65536 [9:33:18<1:44:30,  1.61it/s]
 85%|████████▍ | 55437/65536 [9:33:19<1:47:11,  1.57it/s]
 85%|████████▍ | 55438/65536 [9:33:19<1:45:03,  1.60it/s]
 85%|████████▍ | 55439/65536 [9:33:20<1:44:08,  1.62it/s]
 85%|████████▍ | 55440/65536 [9:33:21<1:45:37,  1.59it/s]
                                                         
{'loss': 1.5618, 'learning_rate': 2.390718746173625e-07, 'epoch': 3422.22}
+
 85%|████████▍ | 55440/65536 [9:33:21<1:45:37,  1.59it/s]
 85%|████████▍ | 55441/65536 [9:33:21<1:44:37,  1.61it/s]
 85%|████████▍ | 55442/65536 [9:33:22<1:44:02,  1.62it/s]
 85%|████████▍ | 55443/65536 [9:33:22<1:44:49,  1.60it/s]
 85%|████████▍ | 55444/65536 [9:33:23<1:43:52,  1.62it/s]
 85%|████████▍ | 55445/65536 [9:33:24<1:43:52,  1.62it/s]
 85%|████████▍ | 55446/65536 [9:33:24<1:46:36,  1.58it/s]
 85%|████████▍ | 55447/65536 [9:33:25<1:46:51,  1.57it/s]
 85%|████████▍ | 55448/65536 [9:33:26<1:45:19,  1.60it/s]
 85%|████████▍ | 55449/65536 [9:33:26<1:44:43,  1.61it/s]
 85%|████████▍ | 55450/65536 [9:33:27<1:43:25,  1.63it/s]
 85%|████████▍ | 55451/65536 [9:33:27<1:41:55,  1.65it/s]
 85%|████████▍ | 55452/65536 [9:33:28<1:41:13,  1.66it/s]
 85%|████████▍ | 55453/65536 [9:33:29<1:45:08,  1.60it/s]
 85%|████████▍ | 55454/65536 [9:33:29<1:46:01,  1.58it/s]
 85%|████████▍ | 55455/65536 [9:33:30<1:45:23,  1.59it/s]
 85%|████████▍ | 55456/65536 [9:33:30<1:44:22,  1.61it/s]
 85%|████████▍ | 55457/65536 [9:33:31<1:45:43,  1.59it/s]
 85%|████████▍ | 55458/65536 [9:33:32<1:45:52,  1.59it/s]
 85%|████████▍ | 55459/65536 [9:33:32<1:45:34,  1.59it/s]
 85%|████████▍ | 55460/65536 [9:33:33<1:44:36,  1.61it/s]
                                                         
{'loss': 1.5525, 'learning_rate': 2.3879637565813637e-07, 'epoch': 3423.46}
+
 85%|████████▍ | 55460/65536 [9:33:33<1:44:36,  1.61it/s]
 85%|████████▍ | 55461/65536 [9:33:34<1:44:01,  1.61it/s]
 85%|████████▍ | 55462/65536 [9:33:34<1:43:55,  1.62it/s]
 85%|████████▍ | 55463/65536 [9:33:35<1:43:02,  1.63it/s]
 85%|████████▍ | 55464/65536 [9:33:35<1:45:41,  1.59it/s]
 85%|████████▍ | 55465/65536 [9:33:36<1:45:17,  1.59it/s]
 85%|████████▍ | 55466/65536 [9:33:37<1:42:03,  1.64it/s]
 85%|████████▍ | 55467/65536 [9:33:37<1:42:07,  1.64it/s]
 85%|████████▍ | 55468/65536 [9:33:38<1:43:31,  1.62it/s]
 85%|████████▍ | 55469/65536 [9:33:39<1:46:05,  1.58it/s]
 85%|████████▍ | 55470/65536 [9:33:39<1:45:37,  1.59it/s]
 85%|████████▍ | 55471/65536 [9:33:40<1:45:24,  1.59it/s]
 85%|████████▍ | 55472/65536 [9:33:40<1:47:07,  1.57it/s]
 85%|████████▍ | 55473/65536 [9:33:41<1:47:21,  1.56it/s]
 85%|████████▍ | 55474/65536 [9:33:42<1:45:45,  1.59it/s]
 85%|████████▍ | 55475/65536 [9:33:42<1:48:32,  1.54it/s]
 85%|████████▍ | 55476/65536 [9:33:43<1:46:38,  1.57it/s]
 85%|████████▍ | 55477/65536 [9:33:44<1:46:48,  1.57it/s]
 85%|████████▍ | 55478/65536 [9:33:44<1:45:03,  1.60it/s]
 85%|████████▍ | 55479/65536 [9:33:45<1:42:25,  1.64it/s]
 85%|████████▍ | 55480/65536 [9:33:45<1:43:41,  1.62it/s]
                                                         
{'loss': 1.5701, 'learning_rate': 2.3852087669891023e-07, 'epoch': 3424.69}
+
 85%|████████▍ | 55480/65536 [9:33:45<1:43:41,  1.62it/s]
 85%|████████▍ | 55481/65536 [9:33:46<1:44:13,  1.61it/s]
 85%|████████▍ | 55482/65536 [9:33:47<1:44:45,  1.60it/s]
 85%|████████▍ | 55483/65536 [9:33:47<1:44:08,  1.61it/s]
 85%|████████▍ | 55484/65536 [9:33:48<1:44:23,  1.60it/s]
 85%|████████▍ | 55485/65536 [9:33:49<1:43:12,  1.62it/s]
 85%|████████▍ | 55486/65536 [9:33:49<1:51:13,  1.51it/s]
 85%|████████▍ | 55487/65536 [9:33:50<1:47:48,  1.55it/s]
 85%|████████▍ | 55488/65536 [9:33:51<1:45:09,  1.59it/s]
 85%|████████▍ | 55489/65536 [9:33:51<1:44:03,  1.61it/s]
 85%|████████▍ | 55490/65536 [9:33:52<1:44:46,  1.60it/s]
 85%|████████▍ | 55491/65536 [9:33:52<1:44:36,  1.60it/s]
 85%|████████▍ | 55492/65536 [9:33:53<1:43:35,  1.62it/s]
 85%|████████▍ | 55493/65536 [9:33:54<1:43:37,  1.62it/s]
 85%|████████▍ | 55494/65536 [9:33:54<1:42:59,  1.63it/s]
 85%|████████▍ | 55495/65536 [9:33:55<1:42:13,  1.64it/s]
 85%|████████▍ | 55496/65536 [9:33:56<1:45:16,  1.59it/s]
 85%|████████▍ | 55497/65536 [9:33:56<1:45:00,  1.59it/s]
 85%|████████▍ | 55498/65536 [9:33:57<1:44:44,  1.60it/s]
 85%|████████▍ | 55499/65536 [9:33:57<1:43:13,  1.62it/s]
 85%|████████▍ | 55500/65536 [9:33:58<1:41:26,  1.65it/s]
                                                         
{'loss': 1.5665, 'learning_rate': 2.382453777396841e-07, 'epoch': 3425.93}
+
 85%|████████▍ | 55500/65536 [9:33:58<1:41:26,  1.65it/s]
 85%|████████▍ | 55501/65536 [9:33:59<1:42:15,  1.64it/s]
 85%|████████▍ | 55502/65536 [9:33:59<1:47:16,  1.56it/s]
 85%|████████▍ | 55503/65536 [9:34:00<1:44:52,  1.59it/s]
 85%|████████▍ | 55504/65536 [9:34:00<1:43:50,  1.61it/s]
 85%|████████▍ | 55505/65536 [9:34:01<1:42:19,  1.63it/s]
 85%|████████▍ | 55506/65536 [9:34:02<1:44:35,  1.60it/s]
 85%|████████▍ | 55507/65536 [9:34:02<1:42:31,  1.63it/s]
 85%|████████▍ | 55508/65536 [9:34:03<1:44:09,  1.60it/s]
 85%|████████▍ | 55509/65536 [9:34:04<1:43:16,  1.62it/s]
 85%|████████▍ | 55510/65536 [9:34:04<1:44:00,  1.61it/s]
 85%|████████▍ | 55511/65536 [9:34:05<1:43:37,  1.61it/s]
 85%|████████▍ | 55512/65536 [9:34:05<1:46:26,  1.57it/s]
 85%|████████▍ | 55513/65536 [9:34:06<1:46:17,  1.57it/s]
 85%|████████▍ | 55514/65536 [9:34:07<1:47:53,  1.55it/s]
 85%|████████▍ | 55515/65536 [9:34:07<1:46:26,  1.57it/s]
 85%|████████▍ | 55516/65536 [9:34:08<1:43:29,  1.61it/s]
 85%|████████▍ | 55517/65536 [9:34:09<1:43:00,  1.62it/s]
 85%|████████▍ | 55518/65536 [9:34:09<1:45:16,  1.59it/s]
 85%|████████▍ | 55519/65536 [9:34:10<1:43:56,  1.61it/s]
 85%|████████▍ | 55520/65536 [9:34:10<1:43:51,  1.61it/s]
                                                         
{'loss': 1.5478, 'learning_rate': 2.3796987878045797e-07, 'epoch': 3427.16}
+
 85%|████████▍ | 55520/65536 [9:34:10<1:43:51,  1.61it/s]
 85%|████████▍ | 55521/65536 [9:34:11<1:42:47,  1.62it/s]
 85%|████████▍ | 55522/65536 [9:34:12<1:41:49,  1.64it/s]
 85%|████████▍ | 55523/65536 [9:34:12<1:41:30,  1.64it/s]
 85%|████████▍ | 55524/65536 [9:34:13<1:43:04,  1.62it/s]
 85%|████████▍ | 55525/65536 [9:34:13<1:40:01,  1.67it/s]
 85%|████████▍ | 55526/65536 [9:34:14<1:39:16,  1.68it/s]
 85%|████████▍ | 55527/65536 [9:34:15<1:40:58,  1.65it/s]
 85%|████████▍ | 55528/65536 [9:34:15<1:43:02,  1.62it/s]
 85%|████████▍ | 55529/65536 [9:34:16<1:44:50,  1.59it/s]
 85%|████████▍ | 55530/65536 [9:34:17<1:48:37,  1.54it/s]
 85%|████████▍ | 55531/65536 [9:34:17<1:47:05,  1.56it/s]
 85%|████████▍ | 55532/65536 [9:34:18<1:47:24,  1.55it/s]
 85%|████████▍ | 55533/65536 [9:34:19<1:46:26,  1.57it/s]
 85%|████████▍ | 55534/65536 [9:34:19<1:48:26,  1.54it/s]
 85%|████████▍ | 55535/65536 [9:34:20<1:48:49,  1.53it/s]
 85%|████████▍ | 55536/65536 [9:34:21<1:45:37,  1.58it/s]
 85%|████████▍ | 55537/65536 [9:34:21<1:43:33,  1.61it/s]
 85%|████████▍ | 55538/65536 [9:34:22<1:41:26,  1.64it/s]
 85%|████████▍ | 55539/65536 [9:34:22<1:42:58,  1.62it/s]
 85%|████████▍ | 55540/65536 [9:34:23<1:43:12,  1.61it/s]
                                                         
{'loss': 1.5384, 'learning_rate': 2.3769437982123175e-07, 'epoch': 3428.4}
+
 85%|████████▍ | 55540/65536 [9:34:23<1:43:12,  1.61it/s]
 85%|████████▍ | 55541/65536 [9:34:24<1:43:01,  1.62it/s]
 85%|████████▍ | 55542/65536 [9:34:24<1:42:32,  1.62it/s]
 85%|████████▍ | 55543/65536 [9:34:25<1:43:37,  1.61it/s]
 85%|████████▍ | 55544/65536 [9:34:25<1:43:05,  1.62it/s]
 85%|████████▍ | 55545/65536 [9:34:26<1:43:32,  1.61it/s]
 85%|████████▍ | 55546/65536 [9:34:27<1:41:59,  1.63it/s]
 85%|████████▍ | 55547/65536 [9:34:27<1:42:13,  1.63it/s]
 85%|████████▍ | 55548/65536 [9:34:28<1:43:36,  1.61it/s]
 85%|████████▍ | 55549/65536 [9:34:29<1:45:59,  1.57it/s]
 85%|████████▍ | 55550/65536 [9:34:29<1:47:08,  1.55it/s]
 85%|████████▍ | 55551/65536 [9:34:30<1:43:07,  1.61it/s]
 85%|████████▍ | 55552/65536 [9:34:30<1:45:22,  1.58it/s]
 85%|████████▍ | 55553/65536 [9:34:31<1:44:53,  1.59it/s]
 85%|████████▍ | 55554/65536 [9:34:32<1:45:25,  1.58it/s]
 85%|████████▍ | 55555/65536 [9:34:32<1:43:41,  1.60it/s]
 85%|████████▍ | 55556/65536 [9:34:33<1:44:38,  1.59it/s]
 85%|████████▍ | 55557/65536 [9:34:34<1:42:54,  1.62it/s]
 85%|████████▍ | 55558/65536 [9:34:34<1:42:17,  1.63it/s]
 85%|████████▍ | 55559/65536 [9:34:35<1:42:27,  1.62it/s]
 85%|████████▍ | 55560/65536 [9:34:35<1:41:09,  1.64it/s]
                                                         
{'loss': 1.5703, 'learning_rate': 2.374188808620056e-07, 'epoch': 3429.63}
+
 85%|████████▍ | 55560/65536 [9:34:35<1:41:09,  1.64it/s]
 85%|████████▍ | 55561/65536 [9:34:36<1:40:51,  1.65it/s]
 85%|████████▍ | 55562/65536 [9:34:37<1:40:25,  1.66it/s]
 85%|████████▍ | 55563/65536 [9:34:37<1:40:07,  1.66it/s]
 85%|████████▍ | 55564/65536 [9:34:38<1:38:34,  1.69it/s]
 85%|████████▍ | 55565/65536 [9:34:38<1:42:21,  1.62it/s]
 85%|████████▍ | 55566/65536 [9:34:39<1:40:57,  1.65it/s]
 85%|████████▍ | 55567/65536 [9:34:40<1:44:27,  1.59it/s]
 85%|████████▍ | 55568/65536 [9:34:40<1:41:59,  1.63it/s]
 85%|████████▍ | 55569/65536 [9:34:41<1:40:46,  1.65it/s]
 85%|████████▍ | 55570/65536 [9:34:42<1:43:59,  1.60it/s]
 85%|████████▍ | 55571/65536 [9:34:42<1:44:57,  1.58it/s]
 85%|████████▍ | 55572/65536 [9:34:43<1:47:41,  1.54it/s]
 85%|████████▍ | 55573/65536 [9:34:43<1:46:13,  1.56it/s]
 85%|████████▍ | 55574/65536 [9:34:44<1:45:09,  1.58it/s]
 85%|████████▍ | 55575/65536 [9:34:45<1:44:19,  1.59it/s]
 85%|████████▍ | 55576/65536 [9:34:45<1:41:02,  1.64it/s]
 85%|████████▍ | 55577/65536 [9:34:46<1:42:50,  1.61it/s]
 85%|████████▍ | 55578/65536 [9:34:47<1:43:41,  1.60it/s]
 85%|████████▍ | 55579/65536 [9:34:47<1:43:53,  1.60it/s]
 85%|████████▍ | 55580/65536 [9:34:48<1:44:03,  1.59it/s]
                                                         
{'loss': 1.569, 'learning_rate': 2.3714338190277947e-07, 'epoch': 3430.86}
+
 85%|████████▍ | 55580/65536 [9:34:48<1:44:03,  1.59it/s]
 85%|████████▍ | 55581/65536 [9:34:48<1:44:03,  1.59it/s]
 85%|████████▍ | 55582/65536 [9:34:49<1:45:15,  1.58it/s]
 85%|████████▍ | 55583/65536 [9:34:50<1:47:12,  1.55it/s]
 85%|████████▍ | 55584/65536 [9:34:50<1:46:23,  1.56it/s]
 85%|████████▍ | 55585/65536 [9:34:51<1:48:58,  1.52it/s]
 85%|████████▍ | 55586/65536 [9:34:52<1:49:02,  1.52it/s]
 85%|████████▍ | 55587/65536 [9:34:52<1:46:39,  1.55it/s]
 85%|████████▍ | 55588/65536 [9:34:53<1:46:19,  1.56it/s]
 85%|████████▍ | 55589/65536 [9:34:54<1:45:21,  1.57it/s]
 85%|████████▍ | 55590/65536 [9:34:54<1:44:10,  1.59it/s]
 85%|████████▍ | 55591/65536 [9:34:55<1:43:19,  1.60it/s]
 85%|████████▍ | 55592/65536 [9:34:55<1:41:18,  1.64it/s]
 85%|████████▍ | 55593/65536 [9:34:56<1:41:49,  1.63it/s]
 85%|████████▍ | 55594/65536 [9:34:57<1:42:54,  1.61it/s]
 85%|████████▍ | 55595/65536 [9:34:57<1:41:41,  1.63it/s]
 85%|████████▍ | 55596/65536 [9:34:58<1:39:12,  1.67it/s]
 85%|████████▍ | 55597/65536 [9:34:58<1:38:57,  1.67it/s]
 85%|████████▍ | 55598/65536 [9:34:59<1:38:38,  1.68it/s]
 85%|████████▍ | 55599/65536 [9:35:00<1:44:58,  1.58it/s]
 85%|████████▍ | 55600/65536 [9:35:00<1:43:27,  1.60it/s]
                                                         
{'loss': 1.5612, 'learning_rate': 2.3686788294355333e-07, 'epoch': 3432.1}
+
 85%|████████▍ | 55600/65536 [9:35:00<1:43:27,  1.60it/s]
 85%|████████▍ | 55601/65536 [9:35:01<1:42:54,  1.61it/s]
 85%|████████▍ | 55602/65536 [9:35:02<1:41:01,  1.64it/s]
 85%|████████▍ | 55603/65536 [9:35:02<1:40:37,  1.65it/s]
 85%|████████▍ | 55604/65536 [9:35:03<1:41:04,  1.64it/s]
 85%|████████▍ | 55605/65536 [9:35:03<1:42:09,  1.62it/s]
 85%|████████▍ | 55606/65536 [9:35:04<1:40:10,  1.65it/s]
 85%|████████▍ | 55607/65536 [9:35:05<1:39:11,  1.67it/s]
 85%|████████▍ | 55608/65536 [9:35:05<1:38:59,  1.67it/s]
 85%|████████▍ | 55609/65536 [9:35:06<1:41:15,  1.63it/s]
 85%|████████▍ | 55610/65536 [9:35:06<1:39:18,  1.67it/s]
 85%|████████▍ | 55611/65536 [9:35:07<1:38:54,  1.67it/s]
 85%|████████▍ | 55612/65536 [9:35:08<1:39:44,  1.66it/s]
 85%|████████▍ | 55613/65536 [9:35:08<1:43:15,  1.60it/s]
 85%|████████▍ | 55614/65536 [9:35:09<1:43:28,  1.60it/s]
 85%|████████▍ | 55615/65536 [9:35:10<1:46:47,  1.55it/s]
 85%|████████▍ | 55616/65536 [9:35:10<1:44:38,  1.58it/s]
 85%|████████▍ | 55617/65536 [9:35:11<1:43:54,  1.59it/s]
 85%|████████▍ | 55618/65536 [9:35:11<1:43:29,  1.60it/s]
 85%|████████▍ | 55619/65536 [9:35:12<1:41:46,  1.62it/s]
 85%|████████▍ | 55620/65536 [9:35:13<1:41:15,  1.63it/s]
                                                         
{'loss': 1.5785, 'learning_rate': 2.3659238398432719e-07, 'epoch': 3433.33}
+
 85%|████████▍ | 55620/65536 [9:35:13<1:41:15,  1.63it/s]
 85%|████████▍ | 55621/65536 [9:35:13<1:41:16,  1.63it/s]
 85%|████████▍ | 55622/65536 [9:35:14<1:42:29,  1.61it/s]
 85%|████████▍ | 55623/65536 [9:35:15<1:43:13,  1.60it/s]
 85%|████████▍ | 55624/65536 [9:35:15<1:46:13,  1.56it/s]
 85%|████████▍ | 55625/65536 [9:35:16<1:46:20,  1.55it/s]
 85%|████████▍ | 55626/65536 [9:35:16<1:43:28,  1.60it/s]
 85%|████████▍ | 55627/65536 [9:35:17<1:41:25,  1.63it/s]
 85%|████████▍ | 55628/65536 [9:35:18<1:40:40,  1.64it/s]
 85%|████████▍ | 55629/65536 [9:35:18<1:44:29,  1.58it/s]
 85%|████████▍ | 55630/65536 [9:35:19<1:43:26,  1.60it/s]
 85%|████████▍ | 55631/65536 [9:35:20<1:47:30,  1.54it/s]
 85%|████████▍ | 55632/65536 [9:35:20<1:45:55,  1.56it/s]
 85%|████████▍ | 55633/65536 [9:35:21<1:42:44,  1.61it/s]
 85%|████████▍ | 55634/65536 [9:35:21<1:43:01,  1.60it/s]
 85%|████████▍ | 55635/65536 [9:35:22<1:46:58,  1.54it/s]
 85%|████████▍ | 55636/65536 [9:35:23<1:46:01,  1.56it/s]
 85%|████████▍ | 55637/65536 [9:35:23<1:43:33,  1.59it/s]
 85%|████████▍ | 55638/65536 [9:35:24<1:41:53,  1.62it/s]
 85%|████████▍ | 55639/65536 [9:35:25<1:41:59,  1.62it/s]
 85%|████████▍ | 55640/65536 [9:35:25<1:44:35,  1.58it/s]
                                                         
{'loss': 1.5189, 'learning_rate': 2.3631688502510104e-07, 'epoch': 3434.57}
+
 85%|████████▍ | 55640/65536 [9:35:25<1:44:35,  1.58it/s]
 85%|████████▍ | 55641/65536 [9:35:26<1:43:43,  1.59it/s]
 85%|████████▍ | 55642/65536 [9:35:27<1:43:41,  1.59it/s]
 85%|████████▍ | 55643/65536 [9:35:27<1:41:34,  1.62it/s]
 85%|████████▍ | 55644/65536 [9:35:28<1:40:11,  1.65it/s]
 85%|████████▍ | 55645/65536 [9:35:28<1:40:19,  1.64it/s]
 85%|████████▍ | 55646/65536 [9:35:29<1:38:49,  1.67it/s]
 85%|████████▍ | 55647/65536 [9:35:29<1:38:41,  1.67it/s]
 85%|████████▍ | 55648/65536 [9:35:30<1:42:54,  1.60it/s]
 85%|████████▍ | 55649/65536 [9:35:31<1:43:26,  1.59it/s]
 85%|████████▍ | 55650/65536 [9:35:31<1:43:25,  1.59it/s]
 85%|████████▍ | 55651/65536 [9:35:32<1:42:38,  1.61it/s]
 85%|████████▍ | 55652/65536 [9:35:33<1:44:38,  1.57it/s]
 85%|████████▍ | 55653/65536 [9:35:33<1:43:14,  1.60it/s]
 85%|████████▍ | 55654/65536 [9:35:34<1:45:26,  1.56it/s]
 85%|████████▍ | 55655/65536 [9:35:35<1:43:49,  1.59it/s]
 85%|████████▍ | 55656/65536 [9:35:35<1:42:27,  1.61it/s]
 85%|████████▍ | 55657/65536 [9:35:36<1:40:55,  1.63it/s]
 85%|████████▍ | 55658/65536 [9:35:36<1:40:49,  1.63it/s]
 85%|████████▍ | 55659/65536 [9:35:37<1:40:28,  1.64it/s]
 85%|████████▍ | 55660/65536 [9:35:38<1:39:43,  1.65it/s]
                                                         
{'loss': 1.5885, 'learning_rate': 2.360413860658748e-07, 'epoch': 3435.8}
+
 85%|████████▍ | 55660/65536 [9:35:38<1:39:43,  1.65it/s]
 85%|████████▍ | 55661/65536 [9:35:38<1:40:35,  1.64it/s]
 85%|████████▍ | 55662/65536 [9:35:39<1:39:53,  1.65it/s]
 85%|████████▍ | 55663/65536 [9:35:39<1:37:00,  1.70it/s]
 85%|████████▍ | 55664/65536 [9:35:40<1:42:43,  1.60it/s]
 85%|████████▍ | 55665/65536 [9:35:41<1:42:33,  1.60it/s]
 85%|████████▍ | 55666/65536 [9:35:41<1:42:29,  1.61it/s]
 85%|████████▍ | 55667/65536 [9:35:42<1:42:11,  1.61it/s]
 85%|████████▍ | 55668/65536 [9:35:42<1:39:12,  1.66it/s]
 85%|████████▍ | 55669/65536 [9:35:43<1:40:53,  1.63it/s]
 85%|████████▍ | 55670/65536 [9:35:44<1:40:44,  1.63it/s]
 85%|████████▍ | 55671/65536 [9:35:44<1:39:25,  1.65it/s]
 85%|████████▍ | 55672/65536 [9:35:45<1:40:47,  1.63it/s]
 85%|████████▍ | 55673/65536 [9:35:46<1:42:52,  1.60it/s]
 85%|████████▍ | 55674/65536 [9:35:46<1:42:39,  1.60it/s]
 85%|████████▍ | 55675/65536 [9:35:47<1:41:41,  1.62it/s]
 85%|████████▍ | 55676/65536 [9:35:47<1:43:29,  1.59it/s]
 85%|████████▍ | 55677/65536 [9:35:48<1:41:29,  1.62it/s]
 85%|████████▍ | 55678/65536 [9:35:49<1:41:11,  1.62it/s]
 85%|████████▍ | 55679/65536 [9:35:49<1:41:44,  1.61it/s]
 85%|████████▍ | 55680/65536 [9:35:50<1:45:18,  1.56it/s]
                                                         
{'loss': 1.585, 'learning_rate': 2.3576588710664866e-07, 'epoch': 3437.04}
+
 85%|████████▍ | 55680/65536 [9:35:50<1:45:18,  1.56it/s]
 85%|████████▍ | 55681/65536 [9:35:51<1:47:22,  1.53it/s]
 85%|████████▍ | 55682/65536 [9:35:51<1:46:28,  1.54it/s]
 85%|████████▍ | 55683/65536 [9:35:52<1:43:49,  1.58it/s]
 85%|████████▍ | 55684/65536 [9:35:53<1:42:21,  1.60it/s]
 85%|████████▍ | 55685/65536 [9:35:53<1:42:28,  1.60it/s]
 85%|████████▍ | 55686/65536 [9:35:54<1:45:48,  1.55it/s]
 85%|████████▍ | 55687/65536 [9:35:54<1:44:52,  1.57it/s]
 85%|████████▍ | 55688/65536 [9:35:55<1:46:29,  1.54it/s]
 85%|████████▍ | 55689/65536 [9:35:56<1:44:30,  1.57it/s]
 85%|████████▍ | 55690/65536 [9:35:56<1:41:29,  1.62it/s]
 85%|████████▍ | 55691/65536 [9:35:57<1:39:52,  1.64it/s]
 85%|████████▍ | 55692/65536 [9:35:58<1:40:24,  1.63it/s]
 85%|████████▍ | 55693/65536 [9:35:58<1:39:45,  1.64it/s]
 85%|████████▍ | 55694/65536 [9:35:59<1:39:48,  1.64it/s]
 85%|████████▍ | 55695/65536 [9:35:59<1:39:52,  1.64it/s]
 85%|████████▍ | 55696/65536 [9:36:00<1:45:06,  1.56it/s]
 85%|████████▍ | 55697/65536 [9:36:01<1:40:53,  1.63it/s]
 85%|████████▍ | 55698/65536 [9:36:01<1:40:25,  1.63it/s]
 85%|████████▍ | 55699/65536 [9:36:02<1:38:31,  1.66it/s]
 85%|████████▍ | 55700/65536 [9:36:02<1:40:17,  1.63it/s]
                                                         
{'loss': 1.5307, 'learning_rate': 2.3549038814742252e-07, 'epoch': 3438.27}
+
 85%|████████▍ | 55700/65536 [9:36:02<1:40:17,  1.63it/s]
 85%|████████▍ | 55701/65536 [9:36:03<1:41:58,  1.61it/s]
 85%|████████▍ | 55702/65536 [9:36:04<1:42:12,  1.60it/s]
 85%|████████▍ | 55703/65536 [9:36:04<1:40:05,  1.64it/s]
 85%|████████▍ | 55704/65536 [9:36:05<1:39:39,  1.64it/s]
 85%|████████▍ | 55705/65536 [9:36:06<1:42:22,  1.60it/s]
 85%|████████▌ | 55706/65536 [9:36:06<1:42:25,  1.60it/s]
 85%|████████▌ | 55707/65536 [9:36:07<1:41:28,  1.61it/s]
 85%|████████▌ | 55708/65536 [9:36:07<1:39:44,  1.64it/s]
 85%|████████▌ | 55709/65536 [9:36:08<1:42:22,  1.60it/s]
 85%|████████▌ | 55710/65536 [9:36:09<1:40:00,  1.64it/s]
 85%|████████▌ | 55711/65536 [9:36:09<1:38:20,  1.67it/s]
 85%|████████▌ | 55712/65536 [9:36:10<1:43:23,  1.58it/s]
 85%|████████▌ | 55713/65536 [9:36:11<1:42:45,  1.59it/s]
 85%|████████▌ | 55714/65536 [9:36:11<1:43:32,  1.58it/s]
 85%|████████▌ | 55715/65536 [9:36:12<1:43:15,  1.59it/s]
 85%|████████▌ | 55716/65536 [9:36:12<1:45:41,  1.55it/s]
 85%|████████▌ | 55717/65536 [9:36:13<1:45:19,  1.55it/s]
 85%|████████▌ | 55718/65536 [9:36:14<1:40:39,  1.63it/s]
 85%|████████▌ | 55719/65536 [9:36:14<1:40:01,  1.64it/s]
 85%|████████▌ | 55720/65536 [9:36:15<1:40:48,  1.62it/s]
                                                         
{'loss': 1.5353, 'learning_rate': 2.3521488918819637e-07, 'epoch': 3439.51}
+
 85%|████████▌ | 55720/65536 [9:36:15<1:40:48,  1.62it/s]
 85%|████████▌ | 55721/65536 [9:36:15<1:38:52,  1.65it/s]
 85%|████████▌ | 55722/65536 [9:36:16<1:40:46,  1.62it/s]
 85%|████████▌ | 55723/65536 [9:36:17<1:42:28,  1.60it/s]
 85%|████████▌ | 55724/65536 [9:36:17<1:40:24,  1.63it/s]
 85%|████████▌ | 55725/65536 [9:36:18<1:39:09,  1.65it/s]
 85%|████████▌ | 55726/65536 [9:36:19<1:38:11,  1.67it/s]
 85%|████████▌ | 55727/65536 [9:36:19<1:37:56,  1.67it/s]
 85%|████████▌ | 55728/65536 [9:36:20<1:39:27,  1.64it/s]
 85%|████████▌ | 55729/65536 [9:36:20<1:43:17,  1.58it/s]
 85%|████████▌ | 55730/65536 [9:36:21<1:41:56,  1.60it/s]
 85%|████████▌ | 55731/65536 [9:36:22<1:40:39,  1.62it/s]
 85%|████████▌ | 55732/65536 [9:36:22<1:42:33,  1.59it/s]
 85%|████████▌ | 55733/65536 [9:36:23<1:41:13,  1.61it/s]
 85%|████████▌ | 55734/65536 [9:36:23<1:39:05,  1.65it/s]
 85%|████████▌ | 55735/65536 [9:36:24<1:39:21,  1.64it/s]
 85%|████████▌ | 55736/65536 [9:36:25<1:44:09,  1.57it/s]
 85%|████████▌ | 55737/65536 [9:36:25<1:43:16,  1.58it/s]
 85%|████████▌ | 55738/65536 [9:36:26<1:41:50,  1.60it/s]
 85%|████████▌ | 55739/65536 [9:36:27<1:41:23,  1.61it/s]
 85%|████████▌ | 55740/65536 [9:36:27<1:39:12,  1.65it/s]
                                                         
{'loss': 1.551, 'learning_rate': 2.3493939022897026e-07, 'epoch': 3440.74}
+
 85%|████████▌ | 55740/65536 [9:36:27<1:39:12,  1.65it/s]
 85%|████████▌ | 55741/65536 [9:36:28<1:40:57,  1.62it/s]
 85%|████████▌ | 55742/65536 [9:36:28<1:39:09,  1.65it/s]
 85%|████████▌ | 55743/65536 [9:36:29<1:40:26,  1.63it/s]
 85%|████████▌ | 55744/65536 [9:36:30<1:38:39,  1.65it/s]
 85%|████████▌ | 55745/65536 [9:36:30<1:45:22,  1.55it/s]
 85%|████████▌ | 55746/65536 [9:36:31<1:44:57,  1.55it/s]
 85%|████████▌ | 55747/65536 [9:36:32<1:43:09,  1.58it/s]
 85%|████████▌ | 55748/65536 [9:36:32<1:42:40,  1.59it/s]
 85%|████████▌ | 55749/65536 [9:36:33<1:41:34,  1.61it/s]
 85%|████████▌ | 55750/65536 [9:36:33<1:41:39,  1.60it/s]
 85%|████████▌ | 55751/65536 [9:36:34<1:44:55,  1.55it/s]
 85%|████████▌ | 55752/65536 [9:36:35<1:42:19,  1.59it/s]
 85%|████████▌ | 55753/65536 [9:36:35<1:40:52,  1.62it/s]
 85%|████████▌ | 55754/65536 [9:36:36<1:42:10,  1.60it/s]
 85%|████████▌ | 55755/65536 [9:36:37<1:38:53,  1.65it/s]
 85%|████████▌ | 55756/65536 [9:36:37<1:38:22,  1.66it/s]
 85%|████████▌ | 55757/65536 [9:36:38<1:39:33,  1.64it/s]
 85%|████████▌ | 55758/65536 [9:36:38<1:38:42,  1.65it/s]
 85%|████████▌ | 55759/65536 [9:36:39<1:39:45,  1.63it/s]
 85%|████████▌ | 55760/65536 [9:36:40<1:39:26,  1.64it/s]
                                                         
{'loss': 1.5697, 'learning_rate': 2.3466389126974412e-07, 'epoch': 3441.98}
+
 85%|████████▌ | 55760/65536 [9:36:40<1:39:26,  1.64it/s]
 85%|████████▌ | 55761/65536 [9:36:40<1:42:33,  1.59it/s]
 85%|████████▌ | 55762/65536 [9:36:41<1:41:02,  1.61it/s]
 85%|████████▌ | 55763/65536 [9:36:41<1:41:09,  1.61it/s]
 85%|████████▌ | 55764/65536 [9:36:42<1:38:30,  1.65it/s]
 85%|████████▌ | 55765/65536 [9:36:43<1:40:01,  1.63it/s]
 85%|████████▌ | 55766/65536 [9:36:43<1:38:58,  1.65it/s]
 85%|████████▌ | 55767/65536 [9:36:44<1:40:12,  1.62it/s]
 85%|████████▌ | 55768/65536 [9:36:45<1:40:27,  1.62it/s]
 85%|████████▌ | 55769/65536 [9:36:45<1:40:48,  1.61it/s]
 85%|████████▌ | 55770/65536 [9:36:46<1:39:34,  1.63it/s]
 85%|████████▌ | 55771/65536 [9:36:46<1:38:17,  1.66it/s]
 85%|████████▌ | 55772/65536 [9:36:47<1:39:42,  1.63it/s]
 85%|████████▌ | 55773/65536 [9:36:48<1:39:49,  1.63it/s]
 85%|████████▌ | 55774/65536 [9:36:48<1:42:59,  1.58it/s]
 85%|████████▌ | 55775/65536 [9:36:49<1:41:57,  1.60it/s]
 85%|████████▌ | 55776/65536 [9:36:50<1:43:18,  1.57it/s]
 85%|████████▌ | 55777/65536 [9:36:50<1:46:44,  1.52it/s]
 85%|████████▌ | 55778/65536 [9:36:51<1:46:03,  1.53it/s]
 85%|████████▌ | 55779/65536 [9:36:52<1:45:35,  1.54it/s]
 85%|████████▌ | 55780/65536 [9:36:52<1:44:03,  1.56it/s]
                                                         
{'loss': 1.5602, 'learning_rate': 2.34388392310518e-07, 'epoch': 3443.21}
+
 85%|████████▌ | 55780/65536 [9:36:52<1:44:03,  1.56it/s]
 85%|████████▌ | 55781/65536 [9:36:53<1:42:34,  1.59it/s]
 85%|████████▌ | 55782/65536 [9:36:53<1:39:09,  1.64it/s]
 85%|████████▌ | 55783/65536 [9:36:54<1:41:50,  1.60it/s]
 85%|████████▌ | 55784/65536 [9:36:55<1:41:13,  1.61it/s]
 85%|████████▌ | 55785/65536 [9:36:55<1:42:36,  1.58it/s]
 85%|████████▌ | 55786/65536 [9:36:56<1:41:05,  1.61it/s]
 85%|████████▌ | 55787/65536 [9:36:56<1:40:32,  1.62it/s]
 85%|████████▌ | 55788/65536 [9:36:57<1:42:14,  1.59it/s]
 85%|████████▌ | 55789/65536 [9:36:58<1:39:14,  1.64it/s]
 85%|████████▌ | 55790/65536 [9:36:58<1:37:37,  1.66it/s]
 85%|████████▌ | 55791/65536 [9:36:59<1:37:52,  1.66it/s]
 85%|████████▌ | 55792/65536 [9:36:59<1:37:56,  1.66it/s]
 85%|████████▌ | 55793/65536 [9:37:00<1:42:37,  1.58it/s]
 85%|████████▌ | 55794/65536 [9:37:01<1:41:20,  1.60it/s]
 85%|████████▌ | 55795/65536 [9:37:01<1:41:42,  1.60it/s]
 85%|████████▌ | 55796/65536 [9:37:02<1:43:28,  1.57it/s]
 85%|████████▌ | 55797/65536 [9:37:03<1:41:35,  1.60it/s]
 85%|████████▌ | 55798/65536 [9:37:03<1:40:35,  1.61it/s]
 85%|████████▌ | 55799/65536 [9:37:04<1:38:39,  1.64it/s]
 85%|████████▌ | 55800/65536 [9:37:04<1:37:36,  1.66it/s]
                                                         
{'loss': 1.5776, 'learning_rate': 2.3411289335129173e-07, 'epoch': 3444.44}
+
 85%|████████▌ | 55800/65536 [9:37:04<1:37:36,  1.66it/s]
 85%|████████▌ | 55801/65536 [9:37:05<1:37:12,  1.67it/s]
 85%|████████▌ | 55802/65536 [9:37:06<1:38:38,  1.64it/s]
 85%|████████▌ | 55803/65536 [9:37:06<1:41:08,  1.60it/s]
 85%|████████▌ | 55804/65536 [9:37:07<1:40:24,  1.62it/s]
 85%|████████▌ | 55805/65536 [9:37:08<1:41:57,  1.59it/s]
 85%|████████▌ | 55806/65536 [9:37:08<1:40:24,  1.62it/s]
 85%|████████▌ | 55807/65536 [9:37:09<1:40:38,  1.61it/s]
 85%|████████▌ | 55808/65536 [9:37:09<1:40:50,  1.61it/s]
 85%|████████▌ | 55809/65536 [9:37:10<1:40:34,  1.61it/s]
 85%|████████▌ | 55810/65536 [9:37:11<1:43:09,  1.57it/s]
 85%|████████▌ | 55811/65536 [9:37:11<1:43:05,  1.57it/s]
 85%|████████▌ | 55812/65536 [9:37:12<1:44:34,  1.55it/s]
 85%|████████▌ | 55813/65536 [9:37:13<1:42:05,  1.59it/s]
 85%|████████▌ | 55814/65536 [9:37:13<1:44:21,  1.55it/s]
 85%|████████▌ | 55815/65536 [9:37:14<1:44:38,  1.55it/s]
 85%|████████▌ | 55816/65536 [9:37:15<1:42:28,  1.58it/s]
 85%|████████▌ | 55817/65536 [9:37:15<1:40:42,  1.61it/s]
 85%|████████▌ | 55818/65536 [9:37:16<1:39:55,  1.62it/s]
 85%|████████▌ | 55819/65536 [9:37:16<1:37:54,  1.65it/s]
 85%|████████▌ | 55820/65536 [9:37:17<1:38:31,  1.64it/s]
                                                         
{'loss': 1.5562, 'learning_rate': 2.338373943920656e-07, 'epoch': 3445.68}
+
 85%|████████▌ | 55820/65536 [9:37:17<1:38:31,  1.64it/s]
 85%|████████▌ | 55821/65536 [9:37:18<1:39:38,  1.63it/s]
 85%|████████▌ | 55822/65536 [9:37:18<1:37:32,  1.66it/s]
 85%|████████▌ | 55823/65536 [9:37:19<1:38:32,  1.64it/s]
 85%|████████▌ | 55824/65536 [9:37:19<1:38:14,  1.65it/s]
 85%|████████▌ | 55825/65536 [9:37:20<1:37:16,  1.66it/s]
 85%|████████▌ | 55826/65536 [9:37:21<1:40:35,  1.61it/s]
 85%|████████▌ | 55827/65536 [9:37:21<1:40:23,  1.61it/s]
 85%|████████▌ | 55828/65536 [9:37:22<1:39:59,  1.62it/s]
 85%|████████▌ | 55829/65536 [9:37:23<1:40:54,  1.60it/s]
 85%|████████▌ | 55830/65536 [9:37:23<1:38:23,  1.64it/s]
 85%|████████▌ | 55831/65536 [9:37:24<1:36:54,  1.67it/s]
 85%|████████▌ | 55832/65536 [9:37:24<1:38:31,  1.64it/s]
 85%|████████▌ | 55833/65536 [9:37:25<1:38:31,  1.64it/s]
 85%|████████▌ | 55834/65536 [9:37:25<1:36:39,  1.67it/s]
 85%|████████▌ | 55835/65536 [9:37:26<1:38:52,  1.64it/s]
 85%|████████▌ | 55836/65536 [9:37:27<1:38:51,  1.64it/s]
 85%|████████▌ | 55837/65536 [9:37:27<1:39:22,  1.63it/s]
 85%|████████▌ | 55838/65536 [9:37:28<1:42:06,  1.58it/s]
 85%|████████▌ | 55839/65536 [9:37:29<1:44:48,  1.54it/s]
 85%|████████▌ | 55840/65536 [9:37:29<1:43:33,  1.56it/s]
                                                         
{'loss': 1.5396, 'learning_rate': 2.3356189543283945e-07, 'epoch': 3446.91}
+
 85%|████████▌ | 55840/65536 [9:37:29<1:43:33,  1.56it/s]
 85%|████████▌ | 55841/65536 [9:37:30<1:42:54,  1.57it/s]
 85%|████████▌ | 55842/65536 [9:37:31<1:46:37,  1.52it/s]
 85%|████████▌ | 55843/65536 [9:37:31<1:43:32,  1.56it/s]
 85%|████████▌ | 55844/65536 [9:37:32<1:41:44,  1.59it/s]
 85%|████████▌ | 55845/65536 [9:37:32<1:40:45,  1.60it/s]
 85%|████████▌ | 55846/65536 [9:37:33<1:42:53,  1.57it/s]
 85%|████████▌ | 55847/65536 [9:37:34<1:43:13,  1.56it/s]
 85%|████████▌ | 55848/65536 [9:37:34<1:42:40,  1.57it/s]
 85%|████████▌ | 55849/65536 [9:37:35<1:40:06,  1.61it/s]
 85%|████████▌ | 55850/65536 [9:37:36<1:40:58,  1.60it/s]
 85%|████████▌ | 55851/65536 [9:37:36<1:38:59,  1.63it/s]
 85%|████████▌ | 55852/65536 [9:37:37<1:38:26,  1.64it/s]
 85%|████████▌ | 55853/65536 [9:37:37<1:37:40,  1.65it/s]
 85%|████████▌ | 55854/65536 [9:37:38<1:39:34,  1.62it/s]
 85%|████████▌ | 55855/65536 [9:37:39<1:39:01,  1.63it/s]
 85%|████████▌ | 55856/65536 [9:37:39<1:39:25,  1.62it/s]
 85%|████████▌ | 55857/65536 [9:37:40<1:38:31,  1.64it/s]
 85%|████████▌ | 55858/65536 [9:37:41<1:42:54,  1.57it/s]
 85%|████████▌ | 55859/65536 [9:37:41<1:41:37,  1.59it/s]
 85%|████████▌ | 55860/65536 [9:37:42<1:40:40,  1.60it/s]
                                                         
{'loss': 1.5721, 'learning_rate': 2.332863964736133e-07, 'epoch': 3448.15}
+
 85%|████████▌ | 55860/65536 [9:37:42<1:40:40,  1.60it/s]
 85%|████████▌ | 55861/65536 [9:37:42<1:38:51,  1.63it/s]
 85%|████████▌ | 55862/65536 [9:37:43<1:36:55,  1.66it/s]
 85%|████████▌ | 55863/65536 [9:37:44<1:36:12,  1.68it/s]
 85%|████████▌ | 55864/65536 [9:37:44<1:37:19,  1.66it/s]
 85%|████████▌ | 55865/65536 [9:37:45<1:37:59,  1.64it/s]
 85%|████████▌ | 55866/65536 [9:37:45<1:38:59,  1.63it/s]
 85%|████████▌ | 55867/65536 [9:37:46<1:42:33,  1.57it/s]
 85%|████████▌ | 55868/65536 [9:37:47<1:40:29,  1.60it/s]
 85%|████████▌ | 55869/65536 [9:37:47<1:38:36,  1.63it/s]
 85%|████████▌ | 55870/65536 [9:37:48<1:37:45,  1.65it/s]
 85%|████████▌ | 55871/65536 [9:37:48<1:36:51,  1.66it/s]
 85%|████████▌ | 55872/65536 [9:37:49<1:37:05,  1.66it/s]
 85%|████████▌ | 55873/65536 [9:37:50<1:41:47,  1.58it/s]
 85%|████████▌ | 55874/65536 [9:37:51<1:47:40,  1.50it/s]
 85%|████████▌ | 55875/65536 [9:37:51<1:47:55,  1.49it/s]
 85%|████████▌ | 55876/65536 [9:37:52<1:43:52,  1.55it/s]
 85%|████████▌ | 55877/65536 [9:37:52<1:43:43,  1.55it/s]
 85%|████████▌ | 55878/65536 [9:37:53<1:40:23,  1.60it/s]
 85%|████████▌ | 55879/65536 [9:37:54<1:42:54,  1.56it/s]
 85%|████████▌ | 55880/65536 [9:37:54<1:42:21,  1.57it/s]
                                                         
{'loss': 1.5857, 'learning_rate': 2.3301089751438717e-07, 'epoch': 3449.38}
+
 85%|████████▌ | 55880/65536 [9:37:54<1:42:21,  1.57it/s]
 85%|████████▌ | 55881/65536 [9:37:55<1:38:37,  1.63it/s]
 85%|████████▌ | 55882/65536 [9:37:55<1:38:18,  1.64it/s]
 85%|████████▌ | 55883/65536 [9:37:56<1:37:19,  1.65it/s]
 85%|████████▌ | 55884/65536 [9:37:57<1:37:34,  1.65it/s]
 85%|████████▌ | 55885/65536 [9:37:57<1:36:10,  1.67it/s]
 85%|████████▌ | 55886/65536 [9:37:58<1:36:55,  1.66it/s]
 85%|████████▌ | 55887/65536 [9:37:58<1:36:37,  1.66it/s]
 85%|████████▌ | 55888/65536 [9:37:59<1:40:17,  1.60it/s]
 85%|████████▌ | 55889/65536 [9:38:00<1:40:27,  1.60it/s]
 85%|████████▌ | 55890/65536 [9:38:00<1:40:18,  1.60it/s]
 85%|████████▌ | 55891/65536 [9:38:01<1:42:50,  1.56it/s]
 85%|████████▌ | 55892/65536 [9:38:02<1:39:59,  1.61it/s]
 85%|████████▌ | 55893/65536 [9:38:02<1:40:57,  1.59it/s]
 85%|████████▌ | 55894/65536 [9:38:03<1:40:50,  1.59it/s]
 85%|████████▌ | 55895/65536 [9:38:04<1:41:10,  1.59it/s]
 85%|████████▌ | 55896/65536 [9:38:04<1:41:06,  1.59it/s]
 85%|████████▌ | 55897/65536 [9:38:05<1:41:16,  1.59it/s]
 85%|████████▌ | 55898/65536 [9:38:05<1:41:12,  1.59it/s]
 85%|████████▌ | 55899/65536 [9:38:06<1:41:20,  1.58it/s]
 85%|████████▌ | 55900/65536 [9:38:07<1:39:43,  1.61it/s]
                                                         
{'loss': 1.5669, 'learning_rate': 2.3273539855516102e-07, 'epoch': 3450.62}
+
 85%|████████▌ | 55900/65536 [9:38:07<1:39:43,  1.61it/s]
 85%|████████▌ | 55901/65536 [9:38:07<1:39:07,  1.62it/s]
 85%|████████▌ | 55902/65536 [9:38:08<1:38:38,  1.63it/s]
 85%|████████▌ | 55903/65536 [9:38:09<1:38:22,  1.63it/s]
 85%|████████▌ | 55904/65536 [9:38:09<1:39:33,  1.61it/s]
 85%|████████▌ | 55905/65536 [9:38:10<1:37:38,  1.64it/s]
 85%|████████▌ | 55906/65536 [9:38:10<1:37:55,  1.64it/s]
 85%|████████▌ | 55907/65536 [9:38:11<1:41:40,  1.58it/s]
 85%|████████▌ | 55908/65536 [9:38:12<1:40:38,  1.59it/s]
 85%|████████▌ | 55909/65536 [9:38:12<1:40:22,  1.60it/s]
 85%|████████▌ | 55910/65536 [9:38:13<1:37:51,  1.64it/s]
 85%|████████▌ | 55911/65536 [9:38:13<1:37:34,  1.64it/s]
 85%|████████▌ | 55912/65536 [9:38:14<1:39:03,  1.62it/s]
 85%|████████▌ | 55913/65536 [9:38:15<1:38:23,  1.63it/s]
 85%|████████▌ | 55914/65536 [9:38:15<1:42:35,  1.56it/s]
 85%|████████▌ | 55915/65536 [9:38:16<1:42:15,  1.57it/s]
 85%|████████▌ | 55916/65536 [9:38:17<1:43:20,  1.55it/s]
 85%|████████▌ | 55917/65536 [9:38:17<1:39:53,  1.60it/s]
 85%|████████▌ | 55918/65536 [9:38:18<1:38:27,  1.63it/s]
 85%|████████▌ | 55919/65536 [9:38:18<1:40:16,  1.60it/s]
 85%|████████▌ | 55920/65536 [9:38:19<1:39:54,  1.60it/s]
                                                         
{'loss': 1.5511, 'learning_rate': 2.3245989959593488e-07, 'epoch': 3451.85}
+
 85%|████████▌ | 55920/65536 [9:38:19<1:39:54,  1.60it/s]
 85%|████████▌ | 55921/65536 [9:38:20<1:39:43,  1.61it/s]
 85%|████████▌ | 55922/65536 [9:38:20<1:37:31,  1.64it/s]
 85%|████████▌ | 55923/65536 [9:38:21<1:40:17,  1.60it/s]
 85%|████████▌ | 55924/65536 [9:38:22<1:40:40,  1.59it/s]
 85%|████████▌ | 55925/65536 [9:38:22<1:39:33,  1.61it/s]
 85%|████████▌ | 55926/65536 [9:38:23<1:40:42,  1.59it/s]
 85%|████████▌ | 55927/65536 [9:38:24<1:42:15,  1.57it/s]
 85%|████████▌ | 55928/65536 [9:38:24<1:39:42,  1.61it/s]
 85%|████████▌ | 55929/65536 [9:38:25<1:37:03,  1.65it/s]
 85%|████████▌ | 55930/65536 [9:38:25<1:36:13,  1.66it/s]
 85%|████████▌ | 55931/65536 [9:38:26<1:38:56,  1.62it/s]
 85%|████████▌ | 55932/65536 [9:38:27<1:38:08,  1.63it/s]
 85%|████████▌ | 55933/65536 [9:38:27<1:36:34,  1.66it/s]
 85%|████████▌ | 55934/65536 [9:38:28<1:39:39,  1.61it/s]
 85%|████████▌ | 55935/65536 [9:38:28<1:39:38,  1.61it/s]
 85%|████████▌ | 55936/65536 [9:38:29<1:39:17,  1.61it/s]
 85%|████████▌ | 55937/65536 [9:38:30<1:38:37,  1.62it/s]
 85%|████████▌ | 55938/65536 [9:38:30<1:36:55,  1.65it/s]
 85%|████████▌ | 55939/65536 [9:38:31<1:39:37,  1.61it/s]
 85%|████████▌ | 55940/65536 [9:38:32<1:40:45,  1.59it/s]
                                                         
{'loss': 1.5583, 'learning_rate': 2.3218440063670866e-07, 'epoch': 3453.09}
+
 85%|████████▌ | 55940/65536 [9:38:32<1:40:45,  1.59it/s]
 85%|████████▌ | 55941/65536 [9:38:32<1:38:58,  1.62it/s]
 85%|████████▌ | 55942/65536 [9:38:33<1:36:57,  1.65it/s]
 85%|████████▌ | 55943/65536 [9:38:33<1:36:22,  1.66it/s]
 85%|████████▌ | 55944/65536 [9:38:34<1:35:17,  1.68it/s]
 85%|████████▌ | 55945/65536 [9:38:34<1:36:37,  1.65it/s]
 85%|████████▌ | 55946/65536 [9:38:35<1:38:27,  1.62it/s]
 85%|████████▌ | 55947/65536 [9:38:36<1:41:08,  1.58it/s]
 85%|████████▌ | 55948/65536 [9:38:36<1:39:01,  1.61it/s]
 85%|████████▌ | 55949/65536 [9:38:37<1:41:50,  1.57it/s]
 85%|████████▌ | 55950/65536 [9:38:38<1:39:46,  1.60it/s]
 85%|████████▌ | 55951/65536 [9:38:38<1:40:41,  1.59it/s]
 85%|████████▌ | 55952/65536 [9:38:39<1:39:09,  1.61it/s]
 85%|████████▌ | 55953/65536 [9:38:40<1:40:04,  1.60it/s]
 85%|████████▌ | 55954/65536 [9:38:40<1:38:53,  1.61it/s]
 85%|████████▌ | 55955/65536 [9:38:41<1:41:56,  1.57it/s]
 85%|████████▌ | 55956/65536 [9:38:41<1:40:05,  1.60it/s]
 85%|████████▌ | 55957/65536 [9:38:42<1:41:57,  1.57it/s]
 85%|████████▌ | 55958/65536 [9:38:43<1:39:16,  1.61it/s]
 85%|████████▌ | 55959/65536 [9:38:43<1:38:30,  1.62it/s]
 85%|████████▌ | 55960/65536 [9:38:44<1:39:13,  1.61it/s]
                                                         
{'loss': 1.5677, 'learning_rate': 2.3190890167748252e-07, 'epoch': 3454.32}
+
 85%|████████▌ | 55960/65536 [9:38:44<1:39:13,  1.61it/s]
 85%|████████▌ | 55961/65536 [9:38:45<1:38:33,  1.62it/s]
 85%|████████▌ | 55962/65536 [9:38:45<1:36:08,  1.66it/s]
 85%|████████▌ | 55963/65536 [9:38:46<1:40:13,  1.59it/s]
 85%|████████▌ | 55964/65536 [9:38:46<1:38:53,  1.61it/s]
 85%|████████▌ | 55965/65536 [9:38:47<1:36:40,  1.65it/s]
 85%|████████▌ | 55966/65536 [9:38:48<1:35:50,  1.66it/s]
 85%|████████▌ | 55967/65536 [9:38:48<1:38:49,  1.61it/s]
 85%|████████▌ | 55968/65536 [9:38:49<1:39:22,  1.60it/s]
 85%|████████▌ | 55969/65536 [9:38:49<1:41:19,  1.57it/s]
 85%|████████▌ | 55970/65536 [9:38:50<1:40:38,  1.58it/s]
 85%|████████▌ | 55971/65536 [9:38:51<1:40:58,  1.58it/s]
 85%|████████▌ | 55972/65536 [9:38:51<1:45:47,  1.51it/s]
 85%|████████▌ | 55973/65536 [9:38:52<1:46:16,  1.50it/s]
 85%|████████▌ | 55974/65536 [9:38:53<1:48:06,  1.47it/s]
 85%|████████▌ | 55975/65536 [9:38:53<1:43:30,  1.54it/s]
 85%|████████▌ | 55976/65536 [9:38:54<1:39:44,  1.60it/s]
 85%|████████▌ | 55977/65536 [9:38:55<1:38:30,  1.62it/s]
 85%|████████▌ | 55978/65536 [9:38:55<1:36:49,  1.65it/s]
 85%|████████▌ | 55979/65536 [9:38:56<1:38:04,  1.62it/s]
 85%|████████▌ | 55980/65536 [9:38:57<1:40:13,  1.59it/s]
                                                         
{'loss': 1.5985, 'learning_rate': 2.3163340271825638e-07, 'epoch': 3455.56}
+
 85%|████████▌ | 55980/65536 [9:38:57<1:40:13,  1.59it/s]
 85%|████████▌ | 55981/65536 [9:38:57<1:41:32,  1.57it/s]
 85%|████████▌ | 55982/65536 [9:38:58<1:41:40,  1.57it/s]
 85%|████████▌ | 55983/65536 [9:38:58<1:40:00,  1.59it/s]
 85%|████████▌ | 55984/65536 [9:38:59<1:39:21,  1.60it/s]
 85%|████████▌ | 55985/65536 [9:39:00<1:40:10,  1.59it/s]
 85%|████████▌ | 55986/65536 [9:39:00<1:38:59,  1.61it/s]
 85%|████████▌ | 55987/65536 [9:39:01<1:39:23,  1.60it/s]
 85%|████████▌ | 55988/65536 [9:39:02<1:42:17,  1.56it/s]
 85%|████████▌ | 55989/65536 [9:39:02<1:40:38,  1.58it/s]
 85%|████████▌ | 55990/65536 [9:39:03<1:39:31,  1.60it/s]
 85%|████████▌ | 55991/65536 [9:39:03<1:37:35,  1.63it/s]
 85%|████████▌ | 55992/65536 [9:39:04<1:38:38,  1.61it/s]
 85%|████████▌ | 55993/65536 [9:39:05<1:40:51,  1.58it/s]
 85%|████████▌ | 55994/65536 [9:39:05<1:40:07,  1.59it/s]
 85%|████████▌ | 55995/65536 [9:39:06<1:38:47,  1.61it/s]
 85%|████████▌ | 55996/65536 [9:39:07<1:40:02,  1.59it/s]
 85%|████████▌ | 55997/65536 [9:39:07<1:37:46,  1.63it/s]
 85%|████████▌ | 55998/65536 [9:39:08<1:37:02,  1.64it/s]
 85%|████████▌ | 55999/65536 [9:39:08<1:39:15,  1.60it/s]
 85%|████████▌ | 56000/65536 [9:39:09<1:36:57,  1.64it/s]
                                                         
{'loss': 1.5462, 'learning_rate': 2.3135790375903024e-07, 'epoch': 3456.79}
+
 85%|████████▌ | 56000/65536 [9:39:09<1:36:57,  1.64it/s]
 85%|████████▌ | 56001/65536 [9:39:10<1:37:47,  1.62it/s]
 85%|████████▌ | 56002/65536 [9:39:10<1:36:16,  1.65it/s]
 85%|████████▌ | 56003/65536 [9:39:11<1:36:37,  1.64it/s]
 85%|████████▌ | 56004/65536 [9:39:11<1:40:34,  1.58it/s]
 85%|████████▌ | 56005/65536 [9:39:12<1:39:24,  1.60it/s]
 85%|████████▌ | 56006/65536 [9:39:13<1:39:01,  1.60it/s]
 85%|████████▌ | 56007/65536 [9:39:13<1:39:13,  1.60it/s]
 85%|████████▌ | 56008/65536 [9:39:14<1:39:36,  1.59it/s]
 85%|████████▌ | 56009/65536 [9:39:15<1:37:49,  1.62it/s]
 85%|████████▌ | 56010/65536 [9:39:15<1:36:16,  1.65it/s]
 85%|████████▌ | 56011/65536 [9:39:16<1:37:34,  1.63it/s]
 85%|████████▌ | 56012/65536 [9:39:16<1:36:32,  1.64it/s]
 85%|████████▌ | 56013/65536 [9:39:17<1:36:30,  1.64it/s]
 85%|████████▌ | 56014/65536 [9:39:18<1:36:49,  1.64it/s]
 85%|████████▌ | 56015/65536 [9:39:18<1:37:23,  1.63it/s]
 85%|████████▌ | 56016/65536 [9:39:19<1:38:11,  1.62it/s]
 85%|████████▌ | 56017/65536 [9:39:20<1:42:07,  1.55it/s]
 85%|████████▌ | 56018/65536 [9:39:20<1:40:03,  1.59it/s]
 85%|████████▌ | 56019/65536 [9:39:21<1:40:01,  1.59it/s]
 85%|████████▌ | 56020/65536 [9:39:21<1:39:37,  1.59it/s]
                                                         
{'loss': 1.5744, 'learning_rate': 2.310824047998041e-07, 'epoch': 3458.02}
+
 85%|████████▌ | 56020/65536 [9:39:21<1:39:37,  1.59it/s]
 85%|████████▌ | 56021/65536 [9:39:22<1:38:42,  1.61it/s]
 85%|████████▌ | 56022/65536 [9:39:23<1:38:32,  1.61it/s]
 85%|████████▌ | 56023/65536 [9:39:23<1:41:21,  1.56it/s]
 85%|████████▌ | 56024/65536 [9:39:24<1:44:07,  1.52it/s]
 85%|████████▌ | 56025/65536 [9:39:25<1:42:26,  1.55it/s]
 85%|████████▌ | 56026/65536 [9:39:25<1:43:04,  1.54it/s]
 85%|████████▌ | 56027/65536 [9:39:26<1:41:04,  1.57it/s]
 85%|████████▌ | 56028/65536 [9:39:26<1:38:49,  1.60it/s]
 85%|████████▌ | 56029/65536 [9:39:27<1:38:37,  1.61it/s]
 85%|████████▌ | 56030/65536 [9:39:28<1:37:29,  1.63it/s]
 85%|████████▌ | 56031/65536 [9:39:28<1:38:01,  1.62it/s]
 85%|████████▌ | 56032/65536 [9:39:29<1:37:45,  1.62it/s]
 85%|████████▌ | 56033/65536 [9:39:30<1:38:15,  1.61it/s]
 86%|████████▌ | 56034/65536 [9:39:30<1:38:40,  1.60it/s]
 86%|████████▌ | 56035/65536 [9:39:31<1:40:44,  1.57it/s]
 86%|████████▌ | 56036/65536 [9:39:32<1:43:00,  1.54it/s]
 86%|████████▌ | 56037/65536 [9:39:32<1:41:21,  1.56it/s]
 86%|████████▌ | 56038/65536 [9:39:33<1:40:25,  1.58it/s]
 86%|████████▌ | 56039/65536 [9:39:33<1:41:52,  1.55it/s]
 86%|████████▌ | 56040/65536 [9:39:34<1:39:51,  1.58it/s]
                                                         
{'loss': 1.5822, 'learning_rate': 2.3080690584057796e-07, 'epoch': 3459.26}
+
 86%|████████▌ | 56040/65536 [9:39:34<1:39:51,  1.58it/s]
 86%|████████▌ | 56041/65536 [9:39:35<1:37:58,  1.62it/s]
 86%|████████▌ | 56042/65536 [9:39:35<1:36:14,  1.64it/s]
 86%|████████▌ | 56043/65536 [9:39:36<1:36:55,  1.63it/s]
 86%|████████▌ | 56044/65536 [9:39:36<1:35:19,  1.66it/s]
 86%|████████▌ | 56045/65536 [9:39:37<1:37:19,  1.63it/s]
 86%|████████▌ | 56046/65536 [9:39:38<1:41:04,  1.56it/s]
 86%|████████▌ | 56047/65536 [9:39:38<1:40:26,  1.57it/s]
 86%|████████▌ | 56048/65536 [9:39:39<1:41:43,  1.55it/s]
 86%|████████▌ | 56049/65536 [9:39:40<1:39:20,  1.59it/s]
 86%|████████▌ | 56050/65536 [9:39:40<1:39:20,  1.59it/s]
 86%|████████▌ | 56051/65536 [9:39:41<1:38:54,  1.60it/s]
 86%|████████▌ | 56052/65536 [9:39:42<1:41:13,  1.56it/s]
 86%|████████▌ | 56053/65536 [9:39:42<1:44:04,  1.52it/s]
 86%|████████▌ | 56054/65536 [9:39:43<1:41:15,  1.56it/s]
 86%|████████▌ | 56055/65536 [9:39:44<1:40:06,  1.58it/s]
 86%|████████▌ | 56056/65536 [9:39:44<1:38:09,  1.61it/s]
 86%|████████▌ | 56057/65536 [9:39:45<1:37:26,  1.62it/s]
 86%|████████▌ | 56058/65536 [9:39:45<1:36:27,  1.64it/s]
 86%|████████▌ | 56059/65536 [9:39:46<1:37:56,  1.61it/s]
 86%|████████▌ | 56060/65536 [9:39:47<1:38:08,  1.61it/s]
                                                         
{'loss': 1.5583, 'learning_rate': 2.3053140688135182e-07, 'epoch': 3460.49}
+
 86%|████████▌ | 56060/65536 [9:39:47<1:38:08,  1.61it/s]
 86%|████████▌ | 56061/65536 [9:39:47<1:39:25,  1.59it/s]
 86%|████████▌ | 56062/65536 [9:39:48<1:41:01,  1.56it/s]
 86%|████████▌ | 56063/65536 [9:39:48<1:38:34,  1.60it/s]
 86%|████████▌ | 56064/65536 [9:39:49<1:36:00,  1.64it/s]
 86%|████████▌ | 56065/65536 [9:39:50<1:40:07,  1.58it/s]
 86%|████████▌ | 56066/65536 [9:39:50<1:41:18,  1.56it/s]
 86%|████████▌ | 56067/65536 [9:39:51<1:39:27,  1.59it/s]
 86%|████████▌ | 56068/65536 [9:39:52<1:42:24,  1.54it/s]
 86%|████████▌ | 56069/65536 [9:39:52<1:48:09,  1.46it/s]
 86%|████████▌ | 56070/65536 [9:39:53<1:46:09,  1.49it/s]
 86%|████████▌ | 56071/65536 [9:39:54<1:43:03,  1.53it/s]
 86%|████████▌ | 56072/65536 [9:39:54<1:41:20,  1.56it/s]
 86%|████████▌ | 56073/65536 [9:39:55<1:39:20,  1.59it/s]
 86%|████████▌ | 56074/65536 [9:39:56<1:40:30,  1.57it/s]
 86%|████████▌ | 56075/65536 [9:39:56<1:40:38,  1.57it/s]
 86%|████████▌ | 56076/65536 [9:39:57<1:40:42,  1.57it/s]
 86%|████████▌ | 56077/65536 [9:39:57<1:40:22,  1.57it/s]
 86%|████████▌ | 56078/65536 [9:39:58<1:38:20,  1.60it/s]
 86%|████████▌ | 56079/65536 [9:39:59<1:39:12,  1.59it/s]
 86%|████████▌ | 56080/65536 [9:39:59<1:40:06,  1.57it/s]
                                                         
{'loss': 1.5357, 'learning_rate': 2.3025590792212557e-07, 'epoch': 3461.73}
+
 86%|████████▌ | 56080/65536 [9:39:59<1:40:06,  1.57it/s]
 86%|████████▌ | 56081/65536 [9:40:00<1:42:09,  1.54it/s]
 86%|████████▌ | 56082/65536 [9:40:01<1:41:15,  1.56it/s]
 86%|████████▌ | 56083/65536 [9:40:01<1:39:49,  1.58it/s]
 86%|████████▌ | 56084/65536 [9:40:02<1:37:38,  1.61it/s]
 86%|████████▌ | 56085/65536 [9:40:03<1:41:54,  1.55it/s]
 86%|████████▌ | 56086/65536 [9:40:03<1:40:02,  1.57it/s]
 86%|████████▌ | 56087/65536 [9:40:04<1:39:18,  1.59it/s]
 86%|████████▌ | 56088/65536 [9:40:04<1:40:43,  1.56it/s]
 86%|████████▌ | 56089/65536 [9:40:05<1:39:14,  1.59it/s]
 86%|████████▌ | 56090/65536 [9:40:06<1:40:32,  1.57it/s]
 86%|████████▌ | 56091/65536 [9:40:06<1:39:46,  1.58it/s]
 86%|████████▌ | 56092/65536 [9:40:07<1:40:50,  1.56it/s]
 86%|████████▌ | 56093/65536 [9:40:08<1:41:05,  1.56it/s]
 86%|████████▌ | 56094/65536 [9:40:08<1:39:40,  1.58it/s]
 86%|████████▌ | 56095/65536 [9:40:09<1:36:59,  1.62it/s]
 86%|████████▌ | 56096/65536 [9:40:10<1:38:41,  1.59it/s]
 86%|████████▌ | 56097/65536 [9:40:10<1:37:09,  1.62it/s]
 86%|████████▌ | 56098/65536 [9:40:11<1:35:36,  1.65it/s]
 86%|████████▌ | 56099/65536 [9:40:11<1:37:12,  1.62it/s]
 86%|████████▌ | 56100/65536 [9:40:12<1:37:05,  1.62it/s]
                                                         
{'loss': 1.5823, 'learning_rate': 2.2998040896289945e-07, 'epoch': 3462.96}
+
 86%|████████▌ | 56100/65536 [9:40:12<1:37:05,  1.62it/s]
 86%|████████▌ | 56101/65536 [9:40:13<1:41:51,  1.54it/s]
 86%|████████▌ | 56102/65536 [9:40:13<1:40:52,  1.56it/s]
 86%|████████▌ | 56103/65536 [9:40:14<1:38:19,  1.60it/s]
 86%|████████▌ | 56104/65536 [9:40:15<1:39:22,  1.58it/s]
 86%|████████▌ | 56105/65536 [9:40:15<1:40:09,  1.57it/s]
 86%|████████▌ | 56106/65536 [9:40:16<1:37:30,  1.61it/s]
 86%|████████▌ | 56107/65536 [9:40:16<1:38:33,  1.59it/s]
 86%|████████▌ | 56108/65536 [9:40:17<1:38:53,  1.59it/s]
 86%|████████▌ | 56109/65536 [9:40:18<1:37:04,  1.62it/s]
 86%|████████▌ | 56110/65536 [9:40:18<1:36:53,  1.62it/s]
 86%|████████▌ | 56111/65536 [9:40:19<1:38:44,  1.59it/s]
 86%|████████▌ | 56112/65536 [9:40:20<1:39:50,  1.57it/s]
 86%|████████▌ | 56113/65536 [9:40:20<1:39:09,  1.58it/s]
 86%|████████▌ | 56114/65536 [9:40:21<1:38:02,  1.60it/s]
 86%|████████▌ | 56115/65536 [9:40:21<1:36:21,  1.63it/s]
 86%|████████▌ | 56116/65536 [9:40:22<1:36:59,  1.62it/s]
 86%|████████▌ | 56117/65536 [9:40:23<1:40:57,  1.55it/s]
 86%|████████▌ | 56118/65536 [9:40:23<1:43:17,  1.52it/s]
 86%|████████▌ | 56119/65536 [9:40:24<1:41:28,  1.55it/s]
 86%|████████▌ | 56120/65536 [9:40:25<1:41:43,  1.54it/s]
                                                         
{'loss': 1.5461, 'learning_rate': 2.297049100036733e-07, 'epoch': 3464.2}
+
 86%|████████▌ | 56120/65536 [9:40:25<1:41:43,  1.54it/s]
 86%|████████▌ | 56121/65536 [9:40:25<1:40:16,  1.56it/s]
 86%|████████▌ | 56122/65536 [9:40:26<1:41:49,  1.54it/s]
 86%|████████▌ | 56123/65536 [9:40:27<1:41:32,  1.55it/s]
 86%|████████▌ | 56124/65536 [9:40:27<1:39:32,  1.58it/s]
 86%|████████▌ | 56125/65536 [9:40:28<1:37:12,  1.61it/s]
 86%|████████▌ | 56126/65536 [9:40:28<1:37:27,  1.61it/s]
 86%|████████▌ | 56127/65536 [9:40:29<1:36:36,  1.62it/s]
 86%|████████▌ | 56128/65536 [9:40:30<1:39:00,  1.58it/s]
 86%|████████▌ | 56129/65536 [9:40:30<1:37:17,  1.61it/s]
 86%|████████▌ | 56130/65536 [9:40:31<1:36:45,  1.62it/s]
 86%|████████▌ | 56131/65536 [9:40:32<1:36:39,  1.62it/s]
 86%|████████▌ | 56132/65536 [9:40:32<1:38:23,  1.59it/s]
 86%|████████▌ | 56133/65536 [9:40:33<1:37:07,  1.61it/s]
 86%|████████▌ | 56134/65536 [9:40:33<1:39:50,  1.57it/s]
 86%|████████▌ | 56135/65536 [9:40:34<1:40:10,  1.56it/s]
 86%|████████▌ | 56136/65536 [9:40:35<1:39:30,  1.57it/s]
 86%|████████▌ | 56137/65536 [9:40:35<1:41:30,  1.54it/s]
 86%|████████▌ | 56138/65536 [9:40:36<1:42:31,  1.53it/s]
 86%|████████▌ | 56139/65536 [9:40:37<1:42:45,  1.52it/s]
 86%|████████▌ | 56140/65536 [9:40:37<1:40:20,  1.56it/s]
                                                         
{'loss': 1.5769, 'learning_rate': 2.294294110444472e-07, 'epoch': 3465.43}
+
 86%|████████▌ | 56140/65536 [9:40:37<1:40:20,  1.56it/s]
 86%|████████▌ | 56141/65536 [9:40:38<1:40:01,  1.57it/s]
 86%|████████▌ | 56142/65536 [9:40:39<1:39:46,  1.57it/s]
 86%|████████▌ | 56143/65536 [9:40:39<1:37:15,  1.61it/s]
 86%|████████▌ | 56144/65536 [9:40:40<1:37:09,  1.61it/s]
 86%|████████▌ | 56145/65536 [9:40:40<1:37:45,  1.60it/s]
 86%|████████▌ | 56146/65536 [9:40:41<1:38:41,  1.59it/s]
 86%|████████▌ | 56147/65536 [9:40:42<1:36:57,  1.61it/s]
 86%|████████▌ | 56148/65536 [9:40:42<1:36:37,  1.62it/s]
 86%|████████▌ | 56149/65536 [9:40:43<1:34:51,  1.65it/s]
 86%|████████▌ | 56150/65536 [9:40:44<1:40:09,  1.56it/s]
 86%|████████▌ | 56151/65536 [9:40:44<1:38:22,  1.59it/s]
 86%|████████▌ | 56152/65536 [9:40:45<1:38:13,  1.59it/s]
 86%|████████▌ | 56153/65536 [9:40:45<1:37:13,  1.61it/s]
 86%|████████▌ | 56154/65536 [9:40:46<1:37:49,  1.60it/s]
 86%|████████▌ | 56155/65536 [9:40:47<1:36:53,  1.61it/s]
 86%|████████▌ | 56156/65536 [9:40:47<1:36:12,  1.62it/s]
 86%|████████▌ | 56157/65536 [9:40:48<1:37:54,  1.60it/s]
 86%|████████▌ | 56158/65536 [9:40:49<1:36:48,  1.61it/s]
 86%|████████▌ | 56159/65536 [9:40:49<1:36:08,  1.63it/s]
 86%|████████▌ | 56160/65536 [9:40:50<1:38:04,  1.59it/s]
                                                         
{'loss': 1.5764, 'learning_rate': 2.2915391208522103e-07, 'epoch': 3466.67}
+
 86%|████████▌ | 56160/65536 [9:40:50<1:38:04,  1.59it/s]
 86%|████████▌ | 56161/65536 [9:40:50<1:38:11,  1.59it/s]
 86%|████████▌ | 56162/65536 [9:40:51<1:38:44,  1.58it/s]
 86%|████████▌ | 56163/65536 [9:40:52<1:37:56,  1.60it/s]
 86%|████████▌ | 56164/65536 [9:40:52<1:38:56,  1.58it/s]
 86%|████████▌ | 56165/65536 [9:40:53<1:37:10,  1.61it/s]
 86%|████████▌ | 56166/65536 [9:40:54<1:39:02,  1.58it/s]
 86%|████████▌ | 56167/65536 [9:40:54<1:39:17,  1.57it/s]
 86%|████████▌ | 56168/65536 [9:40:55<1:36:15,  1.62it/s]
 86%|████████▌ | 56169/65536 [9:40:55<1:34:51,  1.65it/s]
 86%|████████▌ | 56170/65536 [9:40:56<1:37:08,  1.61it/s]
 86%|████████▌ | 56171/65536 [9:40:57<1:37:14,  1.61it/s]
 86%|████████▌ | 56172/65536 [9:40:57<1:38:07,  1.59it/s]
 86%|████████▌ | 56173/65536 [9:40:58<1:38:07,  1.59it/s]
 86%|████████▌ | 56174/65536 [9:40:59<1:37:32,  1.60it/s]
 86%|████████▌ | 56175/65536 [9:40:59<1:35:43,  1.63it/s]
 86%|████████▌ | 56176/65536 [9:41:00<1:37:06,  1.61it/s]
 86%|████████▌ | 56177/65536 [9:41:00<1:37:10,  1.61it/s]
 86%|████████▌ | 56178/65536 [9:41:01<1:38:01,  1.59it/s]
 86%|████████▌ | 56179/65536 [9:41:02<1:38:54,  1.58it/s]
 86%|████████▌ | 56180/65536 [9:41:02<1:40:56,  1.54it/s]
                                                         
{'loss': 1.5525, 'learning_rate': 2.288784131259949e-07, 'epoch': 3467.9}
+
 86%|████████▌ | 56180/65536 [9:41:02<1:40:56,  1.54it/s]
 86%|████████▌ | 56181/65536 [9:41:03<1:40:13,  1.56it/s]
 86%|████████▌ | 56182/65536 [9:41:04<1:41:48,  1.53it/s]
 86%|████████▌ | 56183/65536 [9:41:04<1:40:10,  1.56it/s]
 86%|████████▌ | 56184/65536 [9:41:05<1:38:26,  1.58it/s]
 86%|████████▌ | 56185/65536 [9:41:06<1:38:00,  1.59it/s]
 86%|████████▌ | 56186/65536 [9:41:06<1:37:26,  1.60it/s]
 86%|████████▌ | 56187/65536 [9:41:07<1:38:22,  1.58it/s]
 86%|████████▌ | 56188/65536 [9:41:07<1:39:02,  1.57it/s]
 86%|████████▌ | 56189/65536 [9:41:08<1:37:50,  1.59it/s]
 86%|████████▌ | 56190/65536 [9:41:09<1:36:41,  1.61it/s]
 86%|████████▌ | 56191/65536 [9:41:09<1:36:43,  1.61it/s]
 86%|████████▌ | 56192/65536 [9:41:10<1:38:06,  1.59it/s]
 86%|████████▌ | 56193/65536 [9:41:11<1:37:28,  1.60it/s]
 86%|████████▌ | 56194/65536 [9:41:11<1:37:26,  1.60it/s]
 86%|████████▌ | 56195/65536 [9:41:12<1:40:22,  1.55it/s]
 86%|████████▌ | 56196/65536 [9:41:13<1:41:45,  1.53it/s]
 86%|████████▌ | 56197/65536 [9:41:13<1:39:53,  1.56it/s]
 86%|████████▌ | 56198/65536 [9:41:14<1:41:54,  1.53it/s]
 86%|████████▌ | 56199/65536 [9:41:14<1:42:21,  1.52it/s]
 86%|████████▌ | 56200/65536 [9:41:15<1:40:01,  1.56it/s]
                                                         
{'loss': 1.5715, 'learning_rate': 2.2860291416676864e-07, 'epoch': 3469.14}
+
 86%|████████▌ | 56200/65536 [9:41:15<1:40:01,  1.56it/s]
 86%|████████▌ | 56201/65536 [9:41:16<1:38:51,  1.57it/s]
 86%|████████▌ | 56202/65536 [9:41:16<1:40:01,  1.56it/s]
 86%|████████▌ | 56203/65536 [9:41:17<1:38:46,  1.57it/s]
 86%|████████▌ | 56204/65536 [9:41:18<1:37:11,  1.60it/s]
 86%|████████▌ | 56205/65536 [9:41:18<1:36:07,  1.62it/s]
 86%|████████▌ | 56206/65536 [9:41:19<1:35:56,  1.62it/s]
 86%|████████▌ | 56207/65536 [9:41:19<1:33:19,  1.67it/s]
 86%|████████▌ | 56208/65536 [9:41:20<1:35:32,  1.63it/s]
 86%|████████▌ | 56209/65536 [9:41:21<1:36:01,  1.62it/s]
 86%|████████▌ | 56210/65536 [9:41:21<1:37:32,  1.59it/s]
 86%|████████▌ | 56211/65536 [9:41:22<1:37:31,  1.59it/s]
 86%|████████▌ | 56212/65536 [9:41:23<1:38:17,  1.58it/s]
 86%|████████▌ | 56213/65536 [9:41:23<1:39:10,  1.57it/s]
 86%|████████▌ | 56214/65536 [9:41:24<1:36:12,  1.61it/s]
 86%|████████▌ | 56215/65536 [9:41:25<1:41:15,  1.53it/s]
 86%|████████▌ | 56216/65536 [9:41:25<1:38:45,  1.57it/s]
 86%|████████▌ | 56217/65536 [9:41:26<1:38:24,  1.58it/s]
 86%|████████▌ | 56218/65536 [9:41:26<1:37:21,  1.60it/s]
 86%|████████▌ | 56219/65536 [9:41:27<1:37:51,  1.59it/s]
 86%|████████▌ | 56220/65536 [9:41:28<1:36:47,  1.60it/s]
                                                         
{'loss': 1.6001, 'learning_rate': 2.283274152075425e-07, 'epoch': 3470.37}
+
 86%|████████▌ | 56220/65536 [9:41:28<1:36:47,  1.60it/s]
 86%|████████▌ | 56221/65536 [9:41:28<1:36:00,  1.62it/s]
 86%|████████▌ | 56222/65536 [9:41:29<1:38:32,  1.58it/s]
 86%|████████▌ | 56223/65536 [9:41:29<1:37:18,  1.60it/s]
 86%|████████▌ | 56224/65536 [9:41:30<1:38:12,  1.58it/s]
 86%|████████▌ | 56225/65536 [9:41:31<1:39:05,  1.57it/s]
 86%|████████▌ | 56226/65536 [9:41:31<1:38:45,  1.57it/s]
 86%|████████▌ | 56227/65536 [9:41:32<1:38:28,  1.58it/s]
 86%|████████▌ | 56228/65536 [9:41:33<1:41:06,  1.53it/s]
 86%|████████▌ | 56229/65536 [9:41:33<1:37:17,  1.59it/s]
 86%|████████▌ | 56230/65536 [9:41:34<1:36:21,  1.61it/s]
 86%|████████▌ | 56231/65536 [9:41:35<1:41:59,  1.52it/s]
 86%|████████▌ | 56232/65536 [9:41:35<1:39:29,  1.56it/s]
 86%|████████▌ | 56233/65536 [9:41:36<1:37:39,  1.59it/s]
 86%|████████▌ | 56234/65536 [9:41:37<1:39:00,  1.57it/s]
 86%|████████▌ | 56235/65536 [9:41:37<1:37:34,  1.59it/s]
 86%|████████▌ | 56236/65536 [9:41:38<1:36:08,  1.61it/s]
 86%|████████▌ | 56237/65536 [9:41:38<1:35:26,  1.62it/s]
 86%|████████▌ | 56238/65536 [9:41:39<1:34:57,  1.63it/s]
 86%|████████▌ | 56239/65536 [9:41:40<1:36:17,  1.61it/s]
 86%|████████▌ | 56240/65536 [9:41:40<1:36:11,  1.61it/s]
                                                         
{'loss': 1.5408, 'learning_rate': 2.2805191624831636e-07, 'epoch': 3471.6}
+
 86%|████████▌ | 56240/65536 [9:41:40<1:36:11,  1.61it/s]
 86%|████████▌ | 56241/65536 [9:41:41<1:39:44,  1.55it/s]
 86%|████████▌ | 56242/65536 [9:41:41<1:36:39,  1.60it/s]
 86%|████████▌ | 56243/65536 [9:41:42<1:36:28,  1.61it/s]
 86%|████████▌ | 56244/65536 [9:41:43<1:35:45,  1.62it/s]
 86%|████████▌ | 56245/65536 [9:41:43<1:35:35,  1.62it/s]
 86%|████████▌ | 56246/65536 [9:41:44<1:35:43,  1.62it/s]
 86%|████████▌ | 56247/65536 [9:41:45<1:43:28,  1.50it/s]
 86%|████████▌ | 56248/65536 [9:41:45<1:41:32,  1.52it/s]
 86%|████████▌ | 56249/65536 [9:41:46<1:40:15,  1.54it/s]
 86%|████████▌ | 56250/65536 [9:41:47<1:41:15,  1.53it/s]
 86%|████████▌ | 56251/65536 [9:41:47<1:39:33,  1.55it/s]
 86%|████████▌ | 56252/65536 [9:41:48<1:38:27,  1.57it/s]
 86%|████████▌ | 56253/65536 [9:41:48<1:35:16,  1.62it/s]
 86%|████████▌ | 56254/65536 [9:41:49<1:37:54,  1.58it/s]
 86%|████████▌ | 56255/65536 [9:41:50<1:37:30,  1.59it/s]
 86%|████████▌ | 56256/65536 [9:41:50<1:38:15,  1.57it/s]
 86%|████████▌ | 56257/65536 [9:41:51<1:37:58,  1.58it/s]
 86%|████████▌ | 56258/65536 [9:41:52<1:36:39,  1.60it/s]
 86%|████████▌ | 56259/65536 [9:41:52<1:36:50,  1.60it/s]
 86%|████████▌ | 56260/65536 [9:41:53<1:37:19,  1.59it/s]
                                                         
{'loss': 1.5314, 'learning_rate': 2.2777641728909022e-07, 'epoch': 3472.84}
+
 86%|████████▌ | 56260/65536 [9:41:53<1:37:19,  1.59it/s]
 86%|████████▌ | 56261/65536 [9:41:54<1:38:06,  1.58it/s]
 86%|████████▌ | 56262/65536 [9:41:54<1:36:43,  1.60it/s]
 86%|████████▌ | 56263/65536 [9:41:55<1:39:00,  1.56it/s]
 86%|████████▌ | 56264/65536 [9:41:55<1:38:25,  1.57it/s]
 86%|████████▌ | 56265/65536 [9:41:56<1:39:51,  1.55it/s]
 86%|████████▌ | 56266/65536 [9:41:57<1:37:44,  1.58it/s]
 86%|████████▌ | 56267/65536 [9:41:57<1:40:21,  1.54it/s]
 86%|████████▌ | 56268/65536 [9:41:58<1:38:19,  1.57it/s]
 86%|████████▌ | 56269/65536 [9:41:59<1:37:09,  1.59it/s]
 86%|████████▌ | 56270/65536 [9:41:59<1:37:35,  1.58it/s]
 86%|████████▌ | 56271/65536 [9:42:00<1:37:34,  1.58it/s]
 86%|████████▌ | 56272/65536 [9:42:01<1:36:25,  1.60it/s]
 86%|████████▌ | 56273/65536 [9:42:01<1:36:13,  1.60it/s]
 86%|████████▌ | 56274/65536 [9:42:02<1:36:10,  1.60it/s]
 86%|████████▌ | 56275/65536 [9:42:02<1:35:16,  1.62it/s]
 86%|████████▌ | 56276/65536 [9:42:03<1:35:29,  1.62it/s]
 86%|████████▌ | 56277/65536 [9:42:04<1:37:28,  1.58it/s]
 86%|████████▌ | 56278/65536 [9:42:04<1:34:41,  1.63it/s]
 86%|████████▌ | 56279/65536 [9:42:05<1:39:19,  1.55it/s]
 86%|████████▌ | 56280/65536 [9:42:06<1:37:22,  1.58it/s]
                                                         
{'loss': 1.5999, 'learning_rate': 2.2750091832986408e-07, 'epoch': 3474.07}
+
 86%|████████▌ | 56280/65536 [9:42:06<1:37:22,  1.58it/s]
 86%|████████▌ | 56281/65536 [9:42:06<1:38:19,  1.57it/s]
 86%|████████▌ | 56282/65536 [9:42:07<1:39:15,  1.55it/s]
 86%|████████▌ | 56283/65536 [9:42:07<1:37:21,  1.58it/s]
 86%|████████▌ | 56284/65536 [9:42:08<1:36:05,  1.60it/s]
 86%|████████▌ | 56285/65536 [9:42:09<1:35:07,  1.62it/s]
 86%|████████▌ | 56286/65536 [9:42:09<1:33:31,  1.65it/s]
 86%|████████▌ | 56287/65536 [9:42:10<1:33:20,  1.65it/s]
 86%|████████▌ | 56288/65536 [9:42:10<1:34:53,  1.62it/s]
 86%|████████▌ | 56289/65536 [9:42:11<1:34:45,  1.63it/s]
 86%|████████▌ | 56290/65536 [9:42:12<1:35:30,  1.61it/s]
 86%|████████▌ | 56291/65536 [9:42:12<1:38:48,  1.56it/s]
 86%|████████▌ | 56292/65536 [9:42:13<1:37:13,  1.58it/s]
 86%|████████▌ | 56293/65536 [9:42:14<1:38:28,  1.56it/s]
 86%|████████▌ | 56294/65536 [9:42:14<1:38:54,  1.56it/s]
 86%|████████▌ | 56295/65536 [9:42:15<1:38:36,  1.56it/s]
 86%|████████▌ | 56296/65536 [9:42:16<1:41:08,  1.52it/s]
 86%|████████▌ | 56297/65536 [9:42:16<1:39:38,  1.55it/s]
 86%|████████▌ | 56298/65536 [9:42:17<1:42:58,  1.50it/s]
 86%|████████▌ | 56299/65536 [9:42:18<1:39:43,  1.54it/s]
 86%|████████▌ | 56300/65536 [9:42:18<1:38:26,  1.56it/s]
                                                         
{'loss': 1.5545, 'learning_rate': 2.2722541937063794e-07, 'epoch': 3475.31}
+
 86%|████████▌ | 56300/65536 [9:42:18<1:38:26,  1.56it/s]
 86%|████████▌ | 56301/65536 [9:42:19<1:37:50,  1.57it/s]
 86%|████████▌ | 56302/65536 [9:42:19<1:36:45,  1.59it/s]
 86%|████████▌ | 56303/65536 [9:42:20<1:35:47,  1.61it/s]
 86%|████████▌ | 56304/65536 [9:42:21<1:37:40,  1.58it/s]
 86%|████████▌ | 56305/65536 [9:42:21<1:38:29,  1.56it/s]
 86%|████████▌ | 56306/65536 [9:42:22<1:39:02,  1.55it/s]
 86%|████████▌ | 56307/65536 [9:42:23<1:36:44,  1.59it/s]
 86%|████████▌ | 56308/65536 [9:42:23<1:34:48,  1.62it/s]
 86%|████████▌ | 56309/65536 [9:42:24<1:34:47,  1.62it/s]
 86%|████████▌ | 56310/65536 [9:42:24<1:35:38,  1.61it/s]
 86%|████████▌ | 56311/65536 [9:42:25<1:35:15,  1.61it/s]
 86%|████████▌ | 56312/65536 [9:42:26<1:38:01,  1.57it/s]
 86%|████████▌ | 56313/65536 [9:42:26<1:39:00,  1.55it/s]
 86%|████████▌ | 56314/65536 [9:42:27<1:36:45,  1.59it/s]
 86%|████████▌ | 56315/65536 [9:42:28<1:35:57,  1.60it/s]
 86%|████████▌ | 56316/65536 [9:42:28<1:36:03,  1.60it/s]
 86%|████████▌ | 56317/65536 [9:42:29<1:35:27,  1.61it/s]
 86%|████████▌ | 56318/65536 [9:42:29<1:36:03,  1.60it/s]
 86%|████████▌ | 56319/65536 [9:42:30<1:36:22,  1.59it/s]
 86%|████████▌ | 56320/65536 [9:42:31<1:37:38,  1.57it/s]
                                                         
{'loss': 1.5732, 'learning_rate': 2.2694992041141182e-07, 'epoch': 3476.54}
+
 86%|████████▌ | 56320/65536 [9:42:31<1:37:38,  1.57it/s]
 86%|████████▌ | 56321/65536 [9:42:31<1:37:41,  1.57it/s]
 86%|████████▌ | 56322/65536 [9:42:32<1:35:44,  1.60it/s]
 86%|████████▌ | 56323/65536 [9:42:33<1:34:06,  1.63it/s]
 86%|████████▌ | 56324/65536 [9:42:33<1:35:12,  1.61it/s]
 86%|████████▌ | 56325/65536 [9:42:34<1:35:24,  1.61it/s]
 86%|████████▌ | 56326/65536 [9:42:35<1:36:45,  1.59it/s]
 86%|████████▌ | 56327/65536 [9:42:35<1:40:01,  1.53it/s]
 86%|████████▌ | 56328/65536 [9:42:36<1:43:14,  1.49it/s]
 86%|████████▌ | 56329/65536 [9:42:37<1:42:23,  1.50it/s]
 86%|████████▌ | 56330/65536 [9:42:37<1:39:57,  1.53it/s]
 86%|████████▌ | 56331/65536 [9:42:38<1:38:39,  1.56it/s]
 86%|████████▌ | 56332/65536 [9:42:38<1:37:46,  1.57it/s]
 86%|████████▌ | 56333/65536 [9:42:39<1:41:04,  1.52it/s]
 86%|████████▌ | 56334/65536 [9:42:40<1:38:19,  1.56it/s]
 86%|████████▌ | 56335/65536 [9:42:40<1:35:37,  1.60it/s]
 86%|████████▌ | 56336/65536 [9:42:41<1:35:36,  1.60it/s]
 86%|████████▌ | 56337/65536 [9:42:42<1:34:41,  1.62it/s]
 86%|████████▌ | 56338/65536 [9:42:42<1:35:24,  1.61it/s]
 86%|████████▌ | 56339/65536 [9:42:43<1:35:47,  1.60it/s]
 86%|████████▌ | 56340/65536 [9:42:43<1:35:46,  1.60it/s]
                                                         
{'loss': 1.5746, 'learning_rate': 2.2667442145218557e-07, 'epoch': 3477.78}
+
 86%|████████▌ | 56340/65536 [9:42:43<1:35:46,  1.60it/s]
 86%|████████▌ | 56341/65536 [9:42:44<1:35:33,  1.60it/s]
 86%|████████▌ | 56342/65536 [9:42:45<1:34:02,  1.63it/s]
 86%|████████▌ | 56343/65536 [9:42:45<1:34:05,  1.63it/s]
 86%|████████▌ | 56344/65536 [9:42:46<1:38:00,  1.56it/s]
 86%|████████▌ | 56345/65536 [9:42:47<1:36:25,  1.59it/s]
 86%|████████▌ | 56346/65536 [9:42:47<1:35:17,  1.61it/s]
 86%|████████▌ | 56347/65536 [9:42:48<1:33:38,  1.64it/s]
 86%|████████▌ | 56348/65536 [9:42:48<1:37:11,  1.58it/s]
 86%|████████▌ | 56349/65536 [9:42:49<1:34:53,  1.61it/s]
 86%|████████▌ | 56350/65536 [9:42:50<1:35:12,  1.61it/s]
 86%|████████▌ | 56351/65536 [9:42:50<1:38:55,  1.55it/s]
 86%|████████▌ | 56352/65536 [9:42:51<1:37:45,  1.57it/s]
 86%|████████▌ | 56353/65536 [9:42:52<1:35:25,  1.60it/s]
 86%|████████▌ | 56354/65536 [9:42:52<1:33:59,  1.63it/s]
 86%|████████▌ | 56355/65536 [9:42:53<1:33:40,  1.63it/s]
 86%|████████▌ | 56356/65536 [9:42:53<1:32:53,  1.65it/s]
 86%|████████▌ | 56357/65536 [9:42:54<1:32:38,  1.65it/s]
 86%|████████▌ | 56358/65536 [9:42:55<1:32:25,  1.66it/s]
 86%|████████▌ | 56359/65536 [9:42:55<1:34:38,  1.62it/s]
 86%|████████▌ | 56360/65536 [9:42:56<1:41:02,  1.51it/s]
                                                         
{'loss': 1.563, 'learning_rate': 2.2639892249295943e-07, 'epoch': 3479.01}
+
 86%|████████▌ | 56360/65536 [9:42:56<1:41:02,  1.51it/s]
 86%|████████▌ | 56361/65536 [9:42:57<1:37:40,  1.57it/s]
 86%|████████▌ | 56362/65536 [9:42:57<1:40:21,  1.52it/s]
 86%|████████▌ | 56363/65536 [9:42:58<1:36:47,  1.58it/s]
 86%|████████▌ | 56364/65536 [9:42:58<1:36:20,  1.59it/s]
 86%|████████▌ | 56365/65536 [9:42:59<1:35:21,  1.60it/s]
 86%|████████▌ | 56366/65536 [9:43:00<1:38:47,  1.55it/s]
 86%|████████▌ | 56367/65536 [9:43:00<1:37:51,  1.56it/s]
 86%|████████▌ | 56368/65536 [9:43:01<1:36:46,  1.58it/s]
 86%|████████▌ | 56369/65536 [9:43:02<1:37:31,  1.57it/s]
 86%|████████▌ | 56370/65536 [9:43:02<1:37:24,  1.57it/s]
 86%|████████▌ | 56371/65536 [9:43:03<1:37:33,  1.57it/s]
 86%|████████▌ | 56372/65536 [9:43:04<1:38:04,  1.56it/s]
 86%|████████▌ | 56373/65536 [9:43:04<1:35:26,  1.60it/s]
 86%|████████▌ | 56374/65536 [9:43:05<1:33:40,  1.63it/s]
 86%|████████▌ | 56375/65536 [9:43:05<1:34:18,  1.62it/s]
 86%|████████▌ | 56376/65536 [9:43:06<1:34:12,  1.62it/s]
 86%|████████▌ | 56377/65536 [9:43:07<1:40:27,  1.52it/s]
 86%|████████▌ | 56378/65536 [9:43:07<1:37:37,  1.56it/s]
 86%|████████▌ | 56379/65536 [9:43:08<1:40:27,  1.52it/s]
 86%|████████▌ | 56380/65536 [9:43:09<1:39:12,  1.54it/s]
                                                         
{'loss': 1.5645, 'learning_rate': 2.261234235337333e-07, 'epoch': 3480.25}
+
 86%|████████▌ | 56380/65536 [9:43:09<1:39:12,  1.54it/s]
 86%|████████▌ | 56381/65536 [9:43:09<1:36:45,  1.58it/s]
 86%|████████▌ | 56382/65536 [9:43:10<1:34:56,  1.61it/s]
 86%|████████▌ | 56383/65536 [9:43:11<1:36:22,  1.58it/s]
 86%|████████▌ | 56384/65536 [9:43:11<1:34:55,  1.61it/s]
 86%|████████▌ | 56385/65536 [9:43:12<1:34:20,  1.62it/s]
 86%|████████▌ | 56386/65536 [9:43:12<1:33:00,  1.64it/s]
 86%|████████▌ | 56387/65536 [9:43:13<1:33:41,  1.63it/s]
 86%|████████▌ | 56388/65536 [9:43:14<1:35:05,  1.60it/s]
 86%|████████▌ | 56389/65536 [9:43:14<1:33:00,  1.64it/s]
 86%|████████▌ | 56390/65536 [9:43:15<1:32:57,  1.64it/s]
 86%|████████▌ | 56391/65536 [9:43:15<1:35:24,  1.60it/s]
 86%|████████▌ | 56392/65536 [9:43:16<1:34:07,  1.62it/s]
 86%|████████▌ | 56393/65536 [9:43:17<1:36:16,  1.58it/s]
 86%|████████▌ | 56394/65536 [9:43:17<1:35:56,  1.59it/s]
 86%|████████▌ | 56395/65536 [9:43:18<1:34:29,  1.61it/s]
 86%|████████▌ | 56396/65536 [9:43:19<1:34:32,  1.61it/s]
 86%|████████▌ | 56397/65536 [9:43:19<1:33:55,  1.62it/s]
 86%|████████▌ | 56398/65536 [9:43:20<1:36:49,  1.57it/s]
 86%|████████▌ | 56399/65536 [9:43:20<1:35:13,  1.60it/s]
 86%|████████▌ | 56400/65536 [9:43:21<1:35:21,  1.60it/s]
                                                         
{'loss': 1.5557, 'learning_rate': 2.2584792457450715e-07, 'epoch': 3481.48}
+
 86%|████████▌ | 56400/65536 [9:43:21<1:35:21,  1.60it/s]
 86%|████████▌ | 56401/65536 [9:43:22<1:36:25,  1.58it/s]
 86%|████████▌ | 56402/65536 [9:43:22<1:37:01,  1.57it/s]
 86%|████████▌ | 56403/65536 [9:43:23<1:37:49,  1.56it/s]
 86%|████████▌ | 56404/65536 [9:43:24<1:35:58,  1.59it/s]
 86%|████████▌ | 56405/65536 [9:43:24<1:38:07,  1.55it/s]
 86%|████████▌ | 56406/65536 [9:43:25<1:35:28,  1.59it/s]
 86%|████████▌ | 56407/65536 [9:43:26<1:34:51,  1.60it/s]
 86%|████████▌ | 56408/65536 [9:43:26<1:34:28,  1.61it/s]
 86%|████████▌ | 56409/65536 [9:43:27<1:36:50,  1.57it/s]
 86%|████████▌ | 56410/65536 [9:43:27<1:36:25,  1.58it/s]
 86%|████████▌ | 56411/65536 [9:43:28<1:33:16,  1.63it/s]
 86%|████████▌ | 56412/65536 [9:43:29<1:33:36,  1.62it/s]
 86%|████████▌ | 56413/65536 [9:43:29<1:34:06,  1.62it/s]
 86%|████████▌ | 56414/65536 [9:43:30<1:32:35,  1.64it/s]
 86%|████████▌ | 56415/65536 [9:43:31<1:35:20,  1.59it/s]
 86%|████████▌ | 56416/65536 [9:43:31<1:34:30,  1.61it/s]
 86%|████████▌ | 56417/65536 [9:43:32<1:34:58,  1.60it/s]
 86%|████████▌ | 56418/65536 [9:43:32<1:35:19,  1.59it/s]
 86%|████████▌ | 56419/65536 [9:43:33<1:34:29,  1.61it/s]
 86%|████████▌ | 56420/65536 [9:43:34<1:37:55,  1.55it/s]
                                                         
{'loss': 1.6164, 'learning_rate': 2.25572425615281e-07, 'epoch': 3482.72}
+
 86%|████████▌ | 56420/65536 [9:43:34<1:37:55,  1.55it/s]
 86%|████████▌ | 56421/65536 [9:43:34<1:41:02,  1.50it/s]
 86%|████████▌ | 56422/65536 [9:43:35<1:42:24,  1.48it/s]
 86%|████████▌ | 56423/65536 [9:43:36<1:39:47,  1.52it/s]
 86%|████████▌ | 56424/65536 [9:43:36<1:38:10,  1.55it/s]
 86%|████████▌ | 56425/65536 [9:43:37<1:40:21,  1.51it/s]
 86%|████████▌ | 56426/65536 [9:43:38<1:39:43,  1.52it/s]
 86%|████████▌ | 56427/65536 [9:43:38<1:39:04,  1.53it/s]
 86%|████████▌ | 56428/65536 [9:43:39<1:36:31,  1.57it/s]
 86%|████████▌ | 56429/65536 [9:43:40<1:35:04,  1.60it/s]
 86%|████████▌ | 56430/65536 [9:43:40<1:34:16,  1.61it/s]
 86%|████████▌ | 56431/65536 [9:43:41<1:32:52,  1.63it/s]
 86%|████████▌ | 56432/65536 [9:43:41<1:32:06,  1.65it/s]
 86%|████████▌ | 56433/65536 [9:43:42<1:33:25,  1.62it/s]
 86%|████████▌ | 56434/65536 [9:43:43<1:34:46,  1.60it/s]
 86%|████████▌ | 56435/65536 [9:43:43<1:35:48,  1.58it/s]
 86%|████████▌ | 56436/65536 [9:43:44<1:37:34,  1.55it/s]
 86%|████████▌ | 56437/65536 [9:43:45<1:35:00,  1.60it/s]
 86%|████████▌ | 56438/65536 [9:43:45<1:33:56,  1.61it/s]
 86%|████████▌ | 56439/65536 [9:43:46<1:36:59,  1.56it/s]
 86%|████████▌ | 56440/65536 [9:43:46<1:34:46,  1.60it/s]
                                                         
{'loss': 1.558, 'learning_rate': 2.2529692665605487e-07, 'epoch': 3483.95}
+
 86%|████████▌ | 56440/65536 [9:43:46<1:34:46,  1.60it/s]
 86%|████████▌ | 56441/65536 [9:43:47<1:36:48,  1.57it/s]
 86%|████████▌ | 56442/65536 [9:43:48<1:36:00,  1.58it/s]
 86%|████████▌ | 56443/65536 [9:43:48<1:33:55,  1.61it/s]
 86%|████████▌ | 56444/65536 [9:43:49<1:33:11,  1.63it/s]
 86%|████████▌ | 56445/65536 [9:43:50<1:33:35,  1.62it/s]
 86%|████████▌ | 56446/65536 [9:43:50<1:35:37,  1.58it/s]
 86%|████████▌ | 56447/65536 [9:43:51<1:37:41,  1.55it/s]
 86%|████████▌ | 56448/65536 [9:43:51<1:36:23,  1.57it/s]
 86%|████████▌ | 56449/65536 [9:43:52<1:34:32,  1.60it/s]
 86%|████████▌ | 56450/65536 [9:43:53<1:36:29,  1.57it/s]
 86%|████████▌ | 56451/65536 [9:43:53<1:35:00,  1.59it/s]
 86%|████████▌ | 56452/65536 [9:43:54<1:34:22,  1.60it/s]
 86%|████████▌ | 56453/65536 [9:43:55<1:34:17,  1.61it/s]
 86%|████████▌ | 56454/65536 [9:43:55<1:32:28,  1.64it/s]
 86%|████████▌ | 56455/65536 [9:43:56<1:37:08,  1.56it/s]
 86%|████████▌ | 56456/65536 [9:43:56<1:35:31,  1.58it/s]
 86%|████████▌ | 56457/65536 [9:43:57<1:33:49,  1.61it/s]
 86%|████████▌ | 56458/65536 [9:43:58<1:38:03,  1.54it/s]
 86%|████████▌ | 56459/65536 [9:43:58<1:38:24,  1.54it/s]
 86%|████████▌ | 56460/65536 [9:43:59<1:35:15,  1.59it/s]
                                                         
{'loss': 1.5719, 'learning_rate': 2.2502142769682873e-07, 'epoch': 3485.19}
+
 86%|████████▌ | 56460/65536 [9:43:59<1:35:15,  1.59it/s]
 86%|████████▌ | 56461/65536 [9:44:00<1:35:22,  1.59it/s]
 86%|████████▌ | 56462/65536 [9:44:00<1:33:54,  1.61it/s]
 86%|████████▌ | 56463/65536 [9:44:01<1:35:52,  1.58it/s]
 86%|████████▌ | 56464/65536 [9:44:02<1:34:31,  1.60it/s]
 86%|████████▌ | 56465/65536 [9:44:02<1:35:00,  1.59it/s]
 86%|████████▌ | 56466/65536 [9:44:03<1:34:30,  1.60it/s]
 86%|████████▌ | 56467/65536 [9:44:03<1:33:41,  1.61it/s]
 86%|████████▌ | 56468/65536 [9:44:04<1:32:48,  1.63it/s]
 86%|████████▌ | 56469/65536 [9:44:05<1:33:35,  1.61it/s]
 86%|████████▌ | 56470/65536 [9:44:05<1:35:58,  1.57it/s]
 86%|████████▌ | 56471/65536 [9:44:06<1:38:20,  1.54it/s]
 86%|████████▌ | 56472/65536 [9:44:07<1:39:42,  1.52it/s]
 86%|████████▌ | 56473/65536 [9:44:07<1:38:06,  1.54it/s]
 86%|████████▌ | 56474/65536 [9:44:08<1:41:50,  1.48it/s]
 86%|████████▌ | 56475/65536 [9:44:09<1:39:03,  1.52it/s]
 86%|████████▌ | 56476/65536 [9:44:09<1:36:36,  1.56it/s]
 86%|████████▌ | 56477/65536 [9:44:10<1:36:02,  1.57it/s]
 86%|████████▌ | 56478/65536 [9:44:11<1:38:11,  1.54it/s]
 86%|████████▌ | 56479/65536 [9:44:11<1:36:24,  1.57it/s]
 86%|████████▌ | 56480/65536 [9:44:12<1:36:49,  1.56it/s]
                                                         
{'loss': 1.5495, 'learning_rate': 2.247459287376025e-07, 'epoch': 3486.42}
+
 86%|████████▌ | 56480/65536 [9:44:12<1:36:49,  1.56it/s]
 86%|████████▌ | 56481/65536 [9:44:12<1:36:18,  1.57it/s]
 86%|████████▌ | 56482/65536 [9:44:13<1:35:39,  1.58it/s]
 86%|████████▌ | 56483/65536 [9:44:14<1:33:47,  1.61it/s]
 86%|████████▌ | 56484/65536 [9:44:14<1:33:52,  1.61it/s]
 86%|████████▌ | 56485/65536 [9:44:15<1:33:59,  1.61it/s]
 86%|████████▌ | 56486/65536 [9:44:16<1:35:11,  1.58it/s]
 86%|████████▌ | 56487/65536 [9:44:16<1:33:01,  1.62it/s]
 86%|████████▌ | 56488/65536 [9:44:17<1:33:12,  1.62it/s]
 86%|████████▌ | 56489/65536 [9:44:17<1:34:55,  1.59it/s]
 86%|████████▌ | 56490/65536 [9:44:18<1:36:31,  1.56it/s]
 86%|████████▌ | 56491/65536 [9:44:19<1:36:09,  1.57it/s]
 86%|████████▌ | 56492/65536 [9:44:19<1:35:05,  1.59it/s]
 86%|████████▌ | 56493/65536 [9:44:20<1:34:17,  1.60it/s]
 86%|████████▌ | 56494/65536 [9:44:21<1:33:00,  1.62it/s]
 86%|████████▌ | 56495/65536 [9:44:21<1:34:05,  1.60it/s]
 86%|████████▌ | 56496/65536 [9:44:22<1:32:49,  1.62it/s]
 86%|████████▌ | 56497/65536 [9:44:22<1:31:45,  1.64it/s]
 86%|████████▌ | 56498/65536 [9:44:23<1:32:10,  1.63it/s]
 86%|████████▌ | 56499/65536 [9:44:24<1:35:24,  1.58it/s]
 86%|████████▌ | 56500/65536 [9:44:24<1:34:30,  1.59it/s]
                                                         
{'loss': 1.5482, 'learning_rate': 2.244704297783764e-07, 'epoch': 3487.65}
+
 86%|████████▌ | 56500/65536 [9:44:24<1:34:30,  1.59it/s]
 86%|████████▌ | 56501/65536 [9:44:25<1:33:26,  1.61it/s]
 86%|████████▌ | 56502/65536 [9:44:25<1:32:38,  1.63it/s]
 86%|████████▌ | 56503/65536 [9:44:26<1:33:54,  1.60it/s]
 86%|████████▌ | 56504/65536 [9:44:27<1:37:20,  1.55it/s]
 86%|████████▌ | 56505/65536 [9:44:27<1:36:35,  1.56it/s]
 86%|████████▌ | 56506/65536 [9:44:28<1:38:04,  1.53it/s]
 86%|████████▌ | 56507/65536 [9:44:29<1:37:11,  1.55it/s]
 86%|████████▌ | 56508/65536 [9:44:29<1:34:15,  1.60it/s]
 86%|████████▌ | 56509/65536 [9:44:30<1:32:44,  1.62it/s]
 86%|████████▌ | 56510/65536 [9:44:31<1:33:04,  1.62it/s]
 86%|████████▌ | 56511/65536 [9:44:31<1:36:34,  1.56it/s]
 86%|████████▌ | 56512/65536 [9:44:32<1:37:50,  1.54it/s]
 86%|████████▌ | 56513/65536 [9:44:33<1:37:10,  1.55it/s]
 86%|████████▌ | 56514/65536 [9:44:33<1:35:09,  1.58it/s]
 86%|████████▌ | 56515/65536 [9:44:34<1:37:05,  1.55it/s]
 86%|████████▌ | 56516/65536 [9:44:35<1:38:25,  1.53it/s]
 86%|████████▌ | 56517/65536 [9:44:35<1:35:51,  1.57it/s]
 86%|████████▌ | 56518/65536 [9:44:36<1:35:06,  1.58it/s]
 86%|████████▌ | 56519/65536 [9:44:36<1:35:10,  1.58it/s]
 86%|████████▌ | 56520/65536 [9:44:37<1:34:31,  1.59it/s]
                                                         
{'loss': 1.5571, 'learning_rate': 2.2419493081915023e-07, 'epoch': 3488.89}
+
 86%|████████▌ | 56520/65536 [9:44:37<1:34:31,  1.59it/s]
 86%|████████▌ | 56521/65536 [9:44:38<1:32:59,  1.62it/s]
 86%|████████▌ | 56522/65536 [9:44:38<1:35:15,  1.58it/s]
 86%|████████▌ | 56523/65536 [9:44:39<1:33:41,  1.60it/s]
 86%|████████▌ | 56524/65536 [9:44:39<1:32:09,  1.63it/s]
 86%|████████▋ | 56525/65536 [9:44:40<1:32:13,  1.63it/s]
 86%|████████▋ | 56526/65536 [9:44:41<1:32:58,  1.62it/s]
 86%|████████▋ | 56527/65536 [9:44:41<1:32:47,  1.62it/s]
 86%|████████▋ | 56528/65536 [9:44:42<1:32:05,  1.63it/s]
 86%|████████▋ | 56529/65536 [9:44:43<1:34:11,  1.59it/s]
 86%|████████▋ | 56530/65536 [9:44:43<1:34:08,  1.59it/s]
 86%|████████▋ | 56531/65536 [9:44:44<1:38:34,  1.52it/s]
 86%|████████▋ | 56532/65536 [9:44:45<1:37:52,  1.53it/s]
 86%|████████▋ | 56533/65536 [9:44:45<1:36:59,  1.55it/s]
 86%|████████▋ | 56534/65536 [9:44:46<1:37:27,  1.54it/s]
 86%|████████▋ | 56535/65536 [9:44:46<1:36:46,  1.55it/s]
 86%|████████▋ | 56536/65536 [9:44:47<1:35:22,  1.57it/s]
 86%|████████▋ | 56537/65536 [9:44:48<1:36:31,  1.55it/s]
 86%|████████▋ | 56538/65536 [9:44:48<1:36:05,  1.56it/s]
 86%|████████▋ | 56539/65536 [9:44:49<1:39:23,  1.51it/s]
 86%|████████▋ | 56540/65536 [9:44:50<1:37:28,  1.54it/s]
                                                         
{'loss': 1.5504, 'learning_rate': 2.2391943185992408e-07, 'epoch': 3490.12}
+
 86%|████████▋ | 56540/65536 [9:44:50<1:37:28,  1.54it/s]
 86%|████████▋ | 56541/65536 [9:44:50<1:35:56,  1.56it/s]
 86%|████████▋ | 56542/65536 [9:44:51<1:34:35,  1.58it/s]
 86%|████████▋ | 56543/65536 [9:44:52<1:33:09,  1.61it/s]
 86%|████████▋ | 56544/65536 [9:44:52<1:33:37,  1.60it/s]
 86%|████████▋ | 56545/65536 [9:44:53<1:37:36,  1.54it/s]
 86%|████████▋ | 56546/65536 [9:44:54<1:36:15,  1.56it/s]
 86%|████████▋ | 56547/65536 [9:44:54<1:34:49,  1.58it/s]
 86%|████████▋ | 56548/65536 [9:44:55<1:32:44,  1.62it/s]
 86%|████████▋ | 56549/65536 [9:44:55<1:34:15,  1.59it/s]
 86%|████████▋ | 56550/65536 [9:44:56<1:33:45,  1.60it/s]
 86%|████████▋ | 56551/65536 [9:44:57<1:32:38,  1.62it/s]
 86%|████████▋ | 56552/65536 [9:44:57<1:33:54,  1.59it/s]
 86%|████████▋ | 56553/65536 [9:44:58<1:32:54,  1.61it/s]
 86%|████████▋ | 56554/65536 [9:44:59<1:35:05,  1.57it/s]
 86%|████████▋ | 56555/65536 [9:44:59<1:37:22,  1.54it/s]
 86%|████████▋ | 56556/65536 [9:45:00<1:35:12,  1.57it/s]
 86%|████████▋ | 56557/65536 [9:45:00<1:33:28,  1.60it/s]
 86%|████████▋ | 56558/65536 [9:45:01<1:35:12,  1.57it/s]
 86%|████████▋ | 56559/65536 [9:45:02<1:33:18,  1.60it/s]
 86%|████████▋ | 56560/65536 [9:45:02<1:31:26,  1.64it/s]
                                                         
{'loss': 1.5562, 'learning_rate': 2.2364393290069794e-07, 'epoch': 3491.36}
+
 86%|████████▋ | 56560/65536 [9:45:02<1:31:26,  1.64it/s]
 86%|████████▋ | 56561/65536 [9:45:03<1:31:44,  1.63it/s]
 86%|████████▋ | 56562/65536 [9:45:03<1:31:57,  1.63it/s]
 86%|████████▋ | 56563/65536 [9:45:04<1:31:24,  1.64it/s]
 86%|████████▋ | 56564/65536 [9:45:05<1:31:28,  1.63it/s]
 86%|████████▋ | 56565/65536 [9:45:05<1:31:28,  1.63it/s]
 86%|████████▋ | 56566/65536 [9:45:06<1:32:12,  1.62it/s]
 86%|████████▋ | 56567/65536 [9:45:07<1:34:25,  1.58it/s]
 86%|████████▋ | 56568/65536 [9:45:07<1:37:03,  1.54it/s]
 86%|████████▋ | 56569/65536 [9:45:08<1:39:10,  1.51it/s]
 86%|████████▋ | 56570/65536 [9:45:09<1:36:14,  1.55it/s]
 86%|████████▋ | 56571/65536 [9:45:09<1:38:32,  1.52it/s]
 86%|████████▋ | 56572/65536 [9:45:10<1:36:01,  1.56it/s]
 86%|████████▋ | 56573/65536 [9:45:10<1:34:04,  1.59it/s]
 86%|████████▋ | 56574/65536 [9:45:11<1:36:09,  1.55it/s]
 86%|████████▋ | 56575/65536 [9:45:12<1:35:09,  1.57it/s]
 86%|████████▋ | 56576/65536 [9:45:12<1:34:54,  1.57it/s]
 86%|████████▋ | 56577/65536 [9:45:13<1:34:44,  1.58it/s]
 86%|████████▋ | 56578/65536 [9:45:14<1:36:00,  1.56it/s]
 86%|████████▋ | 56579/65536 [9:45:14<1:34:18,  1.58it/s]
 86%|████████▋ | 56580/65536 [9:45:15<1:32:57,  1.61it/s]
                                                         
{'loss': 1.5411, 'learning_rate': 2.233684339414718e-07, 'epoch': 3492.59}
+
 86%|████████▋ | 56580/65536 [9:45:15<1:32:57,  1.61it/s]
 86%|████████▋ | 56581/65536 [9:45:16<1:34:15,  1.58it/s]
 86%|████████▋ | 56582/65536 [9:45:16<1:33:11,  1.60it/s]
 86%|████████▋ | 56583/65536 [9:45:17<1:32:28,  1.61it/s]
 86%|████████▋ | 56584/65536 [9:45:17<1:34:10,  1.58it/s]
 86%|████████▋ | 56585/65536 [9:45:18<1:34:42,  1.58it/s]
 86%|████████▋ | 56586/65536 [9:45:19<1:34:54,  1.57it/s]
 86%|████████▋ | 56587/65536 [9:45:19<1:38:41,  1.51it/s]
 86%|████████▋ | 56588/65536 [9:45:20<1:36:51,  1.54it/s]
 86%|████████▋ | 56589/65536 [9:45:21<1:37:53,  1.52it/s]
 86%|████████▋ | 56590/65536 [9:45:21<1:38:49,  1.51it/s]
 86%|████████▋ | 56591/65536 [9:45:22<1:35:20,  1.56it/s]
 86%|████████▋ | 56592/65536 [9:45:23<1:33:31,  1.59it/s]
 86%|████████▋ | 56593/65536 [9:45:23<1:33:08,  1.60it/s]
 86%|████████▋ | 56594/65536 [9:45:24<1:33:11,  1.60it/s]
 86%|████████▋ | 56595/65536 [9:45:24<1:33:12,  1.60it/s]
 86%|████████▋ | 56596/65536 [9:45:25<1:32:39,  1.61it/s]
 86%|████████▋ | 56597/65536 [9:45:26<1:31:28,  1.63it/s]
 86%|████████▋ | 56598/65536 [9:45:26<1:31:03,  1.64it/s]
 86%|████████▋ | 56599/65536 [9:45:27<1:32:49,  1.60it/s]
 86%|████████▋ | 56600/65536 [9:45:28<1:31:59,  1.62it/s]
                                                         
{'loss': 1.5854, 'learning_rate': 2.2309293498224566e-07, 'epoch': 3493.83}
+
 86%|████████▋ | 56600/65536 [9:45:28<1:31:59,  1.62it/s]
 86%|████████▋ | 56601/65536 [9:45:28<1:31:34,  1.63it/s]
 86%|████████▋ | 56602/65536 [9:45:29<1:31:34,  1.63it/s]
 86%|████████▋ | 56603/65536 [9:45:30<1:37:05,  1.53it/s]
 86%|████████▋ | 56604/65536 [9:45:30<1:36:16,  1.55it/s]
 86%|████████▋ | 56605/65536 [9:45:31<1:33:57,  1.58it/s]
 86%|████████▋ | 56606/65536 [9:45:31<1:35:20,  1.56it/s]
 86%|████████▋ | 56607/65536 [9:45:32<1:34:03,  1.58it/s]
 86%|████████▋ | 56608/65536 [9:45:33<1:32:29,  1.61it/s]
 86%|████████▋ | 56609/65536 [9:45:33<1:31:10,  1.63it/s]
 86%|████████▋ | 56610/65536 [9:45:34<1:31:40,  1.62it/s]
 86%|████████▋ | 56611/65536 [9:45:34<1:30:59,  1.63it/s]
 86%|████████▋ | 56612/65536 [9:45:35<1:31:04,  1.63it/s]
 86%|████████▋ | 56613/65536 [9:45:36<1:32:28,  1.61it/s]
 86%|████████▋ | 56614/65536 [9:45:36<1:31:39,  1.62it/s]
 86%|████████▋ | 56615/65536 [9:45:37<1:31:43,  1.62it/s]
 86%|████████▋ | 56616/65536 [9:45:38<1:33:43,  1.59it/s]
 86%|████████▋ | 56617/65536 [9:45:38<1:37:13,  1.53it/s]
 86%|████████▋ | 56618/65536 [9:45:39<1:36:08,  1.55it/s]
 86%|████████▋ | 56619/65536 [9:45:40<1:34:49,  1.57it/s]
 86%|████████▋ | 56620/65536 [9:45:40<1:35:40,  1.55it/s]
                                                         
{'loss': 1.5206, 'learning_rate': 2.2281743602301941e-07, 'epoch': 3495.06}
+
 86%|████████▋ | 56620/65536 [9:45:40<1:35:40,  1.55it/s]
 86%|████████▋ | 56621/65536 [9:45:41<1:35:00,  1.56it/s]
 86%|████████▋ | 56622/65536 [9:45:41<1:35:14,  1.56it/s]
 86%|████████▋ | 56623/65536 [9:45:42<1:33:18,  1.59it/s]
 86%|████████▋ | 56624/65536 [9:45:43<1:34:51,  1.57it/s]
 86%|████████▋ | 56625/65536 [9:45:43<1:35:27,  1.56it/s]
 86%|████████▋ | 56626/65536 [9:45:44<1:36:59,  1.53it/s]
 86%|████████▋ | 56627/65536 [9:45:45<1:36:37,  1.54it/s]
 86%|████████▋ | 56628/65536 [9:45:45<1:33:41,  1.58it/s]
 86%|████████▋ | 56629/65536 [9:45:46<1:33:27,  1.59it/s]
 86%|████████▋ | 56630/65536 [9:45:47<1:33:37,  1.59it/s]
 86%|████████▋ | 56631/65536 [9:45:47<1:31:45,  1.62it/s]
 86%|████████▋ | 56632/65536 [9:45:48<1:31:25,  1.62it/s]
 86%|████████▋ | 56633/65536 [9:45:48<1:30:18,  1.64it/s]
 86%|████████▋ | 56634/65536 [9:45:49<1:29:32,  1.66it/s]
 86%|████████▋ | 56635/65536 [9:45:50<1:33:30,  1.59it/s]
 86%|████████▋ | 56636/65536 [9:45:50<1:35:34,  1.55it/s]
 86%|████████▋ | 56637/65536 [9:45:51<1:33:44,  1.58it/s]
 86%|████████▋ | 56638/65536 [9:45:51<1:31:21,  1.62it/s]
 86%|████████▋ | 56639/65536 [9:45:52<1:30:40,  1.64it/s]
 86%|████████▋ | 56640/65536 [9:45:53<1:32:11,  1.61it/s]
                                                         
{'loss': 1.5742, 'learning_rate': 2.2254193706379327e-07, 'epoch': 3496.3}
+
 86%|████████▋ | 56640/65536 [9:45:53<1:32:11,  1.61it/s]
 86%|████████▋ | 56641/65536 [9:45:53<1:33:05,  1.59it/s]
 86%|████████▋ | 56642/65536 [9:45:54<1:33:00,  1.59it/s]
 86%|████████▋ | 56643/65536 [9:45:55<1:36:20,  1.54it/s]
 86%|████████▋ | 56644/65536 [9:45:55<1:35:50,  1.55it/s]
 86%|████████▋ | 56645/65536 [9:45:56<1:39:38,  1.49it/s]
 86%|████████▋ | 56646/65536 [9:45:57<1:36:51,  1.53it/s]
 86%|████████▋ | 56647/65536 [9:45:57<1:34:13,  1.57it/s]
 86%|████████▋ | 56648/65536 [9:45:58<1:33:43,  1.58it/s]
 86%|████████▋ | 56649/65536 [9:45:58<1:31:50,  1.61it/s]
 86%|████████▋ | 56650/65536 [9:45:59<1:31:01,  1.63it/s]
 86%|████████▋ | 56651/65536 [9:46:00<1:29:20,  1.66it/s]
 86%|████████▋ | 56652/65536 [9:46:00<1:31:48,  1.61it/s]
 86%|████████▋ | 56653/65536 [9:46:01<1:31:21,  1.62it/s]
 86%|████████▋ | 56654/65536 [9:46:02<1:34:25,  1.57it/s]
 86%|████████▋ | 56655/65536 [9:46:02<1:33:45,  1.58it/s]
 86%|████████▋ | 56656/65536 [9:46:03<1:33:38,  1.58it/s]
 86%|████████▋ | 56657/65536 [9:46:03<1:32:30,  1.60it/s]
 86%|████████▋ | 56658/65536 [9:46:04<1:34:07,  1.57it/s]
 86%|████████▋ | 56659/65536 [9:46:05<1:33:33,  1.58it/s]
 86%|████████▋ | 56660/65536 [9:46:05<1:31:45,  1.61it/s]
                                                         
{'loss': 1.5247, 'learning_rate': 2.2226643810456716e-07, 'epoch': 3497.53}
+
 86%|████████▋ | 56660/65536 [9:46:05<1:31:45,  1.61it/s]
 86%|████████▋ | 56661/65536 [9:46:06<1:32:04,  1.61it/s]
 86%|████████▋ | 56662/65536 [9:46:07<1:32:23,  1.60it/s]
 86%|████████▋ | 56663/65536 [9:46:07<1:31:05,  1.62it/s]
 86%|████████▋ | 56664/65536 [9:46:08<1:33:49,  1.58it/s]
 86%|████████▋ | 56665/65536 [9:46:08<1:31:27,  1.62it/s]
 86%|████████▋ | 56666/65536 [9:46:09<1:31:06,  1.62it/s]
 86%|████████▋ | 56667/65536 [9:46:10<1:32:07,  1.60it/s]
 86%|████████▋ | 56668/65536 [9:46:10<1:34:45,  1.56it/s]
 86%|████████▋ | 56669/65536 [9:46:11<1:32:03,  1.61it/s]
 86%|████████▋ | 56670/65536 [9:46:12<1:30:14,  1.64it/s]
 86%|████████▋ | 56671/65536 [9:46:12<1:32:14,  1.60it/s]
 86%|████████▋ | 56672/65536 [9:46:13<1:30:34,  1.63it/s]
 86%|████████▋ | 56673/65536 [9:46:13<1:31:42,  1.61it/s]
 86%|████████▋ | 56674/65536 [9:46:14<1:30:15,  1.64it/s]
 86%|████████▋ | 56675/65536 [9:46:15<1:29:08,  1.66it/s]
 86%|████████▋ | 56676/65536 [9:46:15<1:29:15,  1.65it/s]
 86%|████████▋ | 56677/65536 [9:46:16<1:32:15,  1.60it/s]
 86%|████████▋ | 56678/65536 [9:46:17<1:31:52,  1.61it/s]
 86%|████████▋ | 56679/65536 [9:46:17<1:31:44,  1.61it/s]
 86%|████████▋ | 56680/65536 [9:46:18<1:32:16,  1.60it/s]
                                                         
{'loss': 1.578, 'learning_rate': 2.2199093914534102e-07, 'epoch': 3498.77}
+
 86%|████████▋ | 56680/65536 [9:46:18<1:32:16,  1.60it/s]
 86%|████████▋ | 56681/65536 [9:46:18<1:32:09,  1.60it/s]
 86%|████████▋ | 56682/65536 [9:46:19<1:35:59,  1.54it/s]
 86%|████████▋ | 56683/65536 [9:46:20<1:35:10,  1.55it/s]
 86%|████████▋ | 56684/65536 [9:46:20<1:35:59,  1.54it/s]
 86%|████████▋ | 56685/65536 [9:46:21<1:34:26,  1.56it/s]
 86%|████████▋ | 56686/65536 [9:46:22<1:32:24,  1.60it/s]
 86%|████████▋ | 56687/65536 [9:46:22<1:30:27,  1.63it/s]
 86%|████████▋ | 56688/65536 [9:46:23<1:30:10,  1.64it/s]
 87%|████████▋ | 56689/65536 [9:46:23<1:28:45,  1.66it/s]
 87%|████████▋ | 56690/65536 [9:46:24<1:29:14,  1.65it/s]
 87%|████████▋ | 56691/65536 [9:46:25<1:29:41,  1.64it/s]
 87%|████████▋ | 56692/65536 [9:46:25<1:29:22,  1.65it/s]
 87%|████████▋ | 56693/65536 [9:46:26<1:31:03,  1.62it/s]
 87%|████████▋ | 56694/65536 [9:46:26<1:30:38,  1.63it/s]
 87%|████████▋ | 56695/65536 [9:46:27<1:31:06,  1.62it/s]
 87%|████████▋ | 56696/65536 [9:46:28<1:31:25,  1.61it/s]
 87%|████████▋ | 56697/65536 [9:46:28<1:33:50,  1.57it/s]
 87%|████████▋ | 56698/65536 [9:46:29<1:34:09,  1.56it/s]
 87%|████████▋ | 56699/65536 [9:46:30<1:33:11,  1.58it/s]
 87%|████████▋ | 56700/65536 [9:46:30<1:34:40,  1.56it/s]
                                                         
{'loss': 1.5607, 'learning_rate': 2.2171544018611488e-07, 'epoch': 3500.0}
+
 87%|████████▋ | 56700/65536 [9:46:30<1:34:40,  1.56it/s]
 87%|████████▋ | 56701/65536 [9:46:31<1:41:21,  1.45it/s]
 87%|████████▋ | 56702/65536 [9:46:32<1:37:34,  1.51it/s]
 87%|████████▋ | 56703/65536 [9:46:32<1:36:03,  1.53it/s]
 87%|████████▋ | 56704/65536 [9:46:33<1:34:25,  1.56it/s]
 87%|████████▋ | 56705/65536 [9:46:34<1:31:27,  1.61it/s]
 87%|████████▋ | 56706/65536 [9:46:34<1:32:19,  1.59it/s]
 87%|████████▋ | 56707/65536 [9:46:35<1:31:08,  1.61it/s]
 87%|████████▋ | 56708/65536 [9:46:35<1:31:27,  1.61it/s]
 87%|████████▋ | 56709/65536 [9:46:36<1:31:28,  1.61it/s]
 87%|████████▋ | 56710/65536 [9:46:37<1:32:58,  1.58it/s]
 87%|████████▋ | 56711/65536 [9:46:37<1:31:41,  1.60it/s]
 87%|████████▋ | 56712/65536 [9:46:38<1:30:44,  1.62it/s]
 87%|████████▋ | 56713/65536 [9:46:39<1:30:44,  1.62it/s]
 87%|████████▋ | 56714/65536 [9:46:39<1:30:33,  1.62it/s]
 87%|████████▋ | 56715/65536 [9:46:40<1:30:05,  1.63it/s]
 87%|████████▋ | 56716/65536 [9:46:40<1:30:46,  1.62it/s]
 87%|████████▋ | 56717/65536 [9:46:41<1:34:37,  1.55it/s]
 87%|████████▋ | 56718/65536 [9:46:42<1:33:33,  1.57it/s]
 87%|████████▋ | 56719/65536 [9:46:42<1:33:04,  1.58it/s]
 87%|████████▋ | 56720/65536 [9:46:43<1:31:50,  1.60it/s]
                                                         
{'loss': 1.5706, 'learning_rate': 2.2143994122688873e-07, 'epoch': 3501.23}
+
 87%|████████▋ | 56720/65536 [9:46:43<1:31:50,  1.60it/s]
 87%|████████▋ | 56721/65536 [9:46:44<1:31:03,  1.61it/s]
 87%|████████▋ | 56722/65536 [9:46:44<1:31:56,  1.60it/s]
 87%|████████▋ | 56723/65536 [9:46:45<1:31:13,  1.61it/s]
 87%|████████▋ | 56724/65536 [9:46:45<1:30:55,  1.62it/s]
 87%|████████▋ | 56725/65536 [9:46:46<1:31:24,  1.61it/s]
 87%|████████▋ | 56726/65536 [9:46:47<1:34:24,  1.56it/s]
 87%|████████▋ | 56727/65536 [9:46:47<1:33:55,  1.56it/s]
 87%|████████▋ | 56728/65536 [9:46:48<1:31:25,  1.61it/s]
 87%|████████▋ | 56729/65536 [9:46:49<1:30:44,  1.62it/s]
 87%|████████▋ | 56730/65536 [9:46:49<1:32:01,  1.59it/s]
 87%|████████▋ | 56731/65536 [9:46:50<1:31:40,  1.60it/s]
 87%|████████▋ | 56732/65536 [9:46:50<1:29:41,  1.64it/s]
 87%|████████▋ | 56733/65536 [9:46:51<1:32:34,  1.58it/s]
 87%|████████▋ | 56734/65536 [9:46:52<1:31:16,  1.61it/s]
 87%|████████▋ | 56735/65536 [9:46:52<1:30:31,  1.62it/s]
 87%|████████▋ | 56736/65536 [9:46:53<1:29:33,  1.64it/s]
 87%|████████▋ | 56737/65536 [9:46:54<1:31:33,  1.60it/s]
 87%|████████▋ | 56738/65536 [9:46:54<1:33:15,  1.57it/s]
 87%|████████▋ | 56739/65536 [9:46:55<1:30:59,  1.61it/s]
 87%|████████▋ | 56740/65536 [9:46:55<1:31:32,  1.60it/s]
                                                         
{'loss': 1.5374, 'learning_rate': 2.211644422676625e-07, 'epoch': 3502.47}
+
 87%|████████▋ | 56740/65536 [9:46:55<1:31:32,  1.60it/s]
 87%|████████▋ | 56741/65536 [9:46:56<1:30:37,  1.62it/s]
 87%|████████▋ | 56742/65536 [9:46:57<1:28:24,  1.66it/s]
 87%|████████▋ | 56743/65536 [9:46:57<1:29:14,  1.64it/s]
 87%|████████▋ | 56744/65536 [9:46:58<1:28:28,  1.66it/s]
 87%|████████▋ | 56745/65536 [9:46:58<1:28:16,  1.66it/s]
 87%|████████▋ | 56746/65536 [9:46:59<1:29:31,  1.64it/s]
 87%|████████▋ | 56747/65536 [9:47:00<1:32:28,  1.58it/s]
 87%|████████▋ | 56748/65536 [9:47:00<1:31:11,  1.61it/s]
 87%|████████▋ | 56749/65536 [9:47:01<1:32:25,  1.58it/s]
 87%|████████▋ | 56750/65536 [9:47:02<1:30:35,  1.62it/s]
 87%|████████▋ | 56751/65536 [9:47:02<1:31:54,  1.59it/s]
 87%|████████▋ | 56752/65536 [9:47:03<1:32:16,  1.59it/s]
 87%|████████▋ | 56753/65536 [9:47:03<1:32:57,  1.57it/s]
 87%|████████▋ | 56754/65536 [9:47:04<1:35:08,  1.54it/s]
 87%|████████▋ | 56755/65536 [9:47:05<1:36:02,  1.52it/s]
 87%|████████▋ | 56756/65536 [9:47:05<1:35:39,  1.53it/s]
 87%|████████▋ | 56757/65536 [9:47:06<1:34:35,  1.55it/s]
 87%|████████▋ | 56758/65536 [9:47:07<1:31:19,  1.60it/s]
 87%|████████▋ | 56759/65536 [9:47:07<1:29:52,  1.63it/s]
 87%|████████▋ | 56760/65536 [9:47:08<1:29:14,  1.64it/s]
                                                         
{'loss': 1.5452, 'learning_rate': 2.2088894330843635e-07, 'epoch': 3503.7}
+
 87%|████████▋ | 56760/65536 [9:47:08<1:29:14,  1.64it/s]
 87%|████████▋ | 56761/65536 [9:47:08<1:30:16,  1.62it/s]
 87%|████████▋ | 56762/65536 [9:47:09<1:31:02,  1.61it/s]
 87%|████████▋ | 56763/65536 [9:47:10<1:31:56,  1.59it/s]
 87%|████████▋ | 56764/65536 [9:47:10<1:33:45,  1.56it/s]
 87%|████████▋ | 56765/65536 [9:47:11<1:35:19,  1.53it/s]
 87%|████████▋ | 56766/65536 [9:47:12<1:37:22,  1.50it/s]
 87%|████████▋ | 56767/65536 [9:47:12<1:37:02,  1.51it/s]
 87%|████████▋ | 56768/65536 [9:47:13<1:35:18,  1.53it/s]
 87%|████████▋ | 56769/65536 [9:47:14<1:34:35,  1.54it/s]
 87%|████████▋ | 56770/65536 [9:47:14<1:30:56,  1.61it/s]
 87%|████████▋ | 56771/65536 [9:47:15<1:30:38,  1.61it/s]
 87%|████████▋ | 56772/65536 [9:47:16<1:29:46,  1.63it/s]
 87%|████████▋ | 56773/65536 [9:47:16<1:28:15,  1.65it/s]
 87%|████████▋ | 56774/65536 [9:47:17<1:28:05,  1.66it/s]
 87%|████████▋ | 56775/65536 [9:47:17<1:30:15,  1.62it/s]
 87%|████████▋ | 56776/65536 [9:47:18<1:31:37,  1.59it/s]
 87%|████████▋ | 56777/65536 [9:47:19<1:30:02,  1.62it/s]
 87%|████████▋ | 56778/65536 [9:47:19<1:29:47,  1.63it/s]
 87%|████████▋ | 56779/65536 [9:47:20<1:29:15,  1.64it/s]
 87%|████████▋ | 56780/65536 [9:47:20<1:29:28,  1.63it/s]
                                                         
{'loss': 1.5477, 'learning_rate': 2.206134443492102e-07, 'epoch': 3504.94}
+
 87%|████████▋ | 56780/65536 [9:47:20<1:29:28,  1.63it/s]
 87%|████████▋ | 56781/65536 [9:47:21<1:31:32,  1.59it/s]
 87%|████████▋ | 56782/65536 [9:47:22<1:36:13,  1.52it/s]
 87%|████████▋ | 56783/65536 [9:47:22<1:37:03,  1.50it/s]
 87%|████████▋ | 56784/65536 [9:47:23<1:37:19,  1.50it/s]
 87%|████████▋ | 56785/65536 [9:47:24<1:33:01,  1.57it/s]
 87%|████████▋ | 56786/65536 [9:47:24<1:31:06,  1.60it/s]
 87%|████████▋ | 56787/65536 [9:47:25<1:30:41,  1.61it/s]
 87%|████████▋ | 56788/65536 [9:47:26<1:31:41,  1.59it/s]
 87%|████████▋ | 56789/65536 [9:47:26<1:30:50,  1.60it/s]
 87%|████████▋ | 56790/65536 [9:47:27<1:31:09,  1.60it/s]
 87%|████████▋ | 56791/65536 [9:47:27<1:29:41,  1.62it/s]
 87%|████████▋ | 56792/65536 [9:47:28<1:29:13,  1.63it/s]
 87%|████████▋ | 56793/65536 [9:47:29<1:29:26,  1.63it/s]
 87%|████████▋ | 56794/65536 [9:47:29<1:27:31,  1.66it/s]
 87%|████████▋ | 56795/65536 [9:47:30<1:28:33,  1.65it/s]
 87%|████████▋ | 56796/65536 [9:47:30<1:29:09,  1.63it/s]
 87%|████████▋ | 56797/65536 [9:47:31<1:26:40,  1.68it/s]
 87%|████████▋ | 56798/65536 [9:47:32<1:30:10,  1.62it/s]
 87%|████████▋ | 56799/65536 [9:47:32<1:28:34,  1.64it/s]
 87%|████████▋ | 56800/65536 [9:47:33<1:27:24,  1.67it/s]
                                                         
{'loss': 1.5657, 'learning_rate': 2.2033794538998406e-07, 'epoch': 3506.17}
+
 87%|████████▋ | 56800/65536 [9:47:33<1:27:24,  1.67it/s]
 87%|████████▋ | 56801/65536 [9:47:33<1:27:20,  1.67it/s]
 87%|████████▋ | 56802/65536 [9:47:34<1:30:18,  1.61it/s]
 87%|████████▋ | 56803/65536 [9:47:35<1:28:15,  1.65it/s]
 87%|████████▋ | 56804/65536 [9:47:35<1:28:17,  1.65it/s]
 87%|████████▋ | 56805/65536 [9:47:36<1:29:47,  1.62it/s]
 87%|████████▋ | 56806/65536 [9:47:37<1:30:38,  1.61it/s]
 87%|████████▋ | 56807/65536 [9:47:37<1:30:32,  1.61it/s]
 87%|████████▋ | 56808/65536 [9:47:38<1:31:59,  1.58it/s]
 87%|████████▋ | 56809/65536 [9:47:38<1:30:16,  1.61it/s]
 87%|████████▋ | 56810/65536 [9:47:39<1:31:28,  1.59it/s]
 87%|████████▋ | 56811/65536 [9:47:40<1:30:57,  1.60it/s]
 87%|████████▋ | 56812/65536 [9:47:40<1:32:02,  1.58it/s]
 87%|████████▋ | 56813/65536 [9:47:41<1:33:00,  1.56it/s]
 87%|████████▋ | 56814/65536 [9:47:42<1:35:43,  1.52it/s]
 87%|████████▋ | 56815/65536 [9:47:42<1:35:23,  1.52it/s]
 87%|████████▋ | 56816/65536 [9:47:43<1:33:30,  1.55it/s]
 87%|████████▋ | 56817/65536 [9:47:44<1:35:22,  1.52it/s]
 87%|████████▋ | 56818/65536 [9:47:44<1:36:13,  1.51it/s]
 87%|████████▋ | 56819/65536 [9:47:45<1:34:06,  1.54it/s]
 87%|████████▋ | 56820/65536 [9:47:46<1:34:16,  1.54it/s]
                                                         
{'loss': 1.5085, 'learning_rate': 2.2006244643075792e-07, 'epoch': 3507.41}
+
 87%|████████▋ | 56820/65536 [9:47:46<1:34:16,  1.54it/s]
 87%|████████▋ | 56821/65536 [9:47:46<1:32:56,  1.56it/s]
 87%|████████▋ | 56822/65536 [9:47:47<1:31:26,  1.59it/s]
 87%|████████▋ | 56823/65536 [9:47:47<1:31:25,  1.59it/s]
 87%|████████▋ | 56824/65536 [9:47:48<1:33:20,  1.56it/s]
 87%|████████▋ | 56825/65536 [9:47:49<1:30:51,  1.60it/s]
 87%|████████▋ | 56826/65536 [9:47:49<1:30:05,  1.61it/s]
 87%|████████▋ | 56827/65536 [9:47:50<1:30:22,  1.61it/s]
 87%|████████▋ | 56828/65536 [9:47:51<1:30:00,  1.61it/s]
 87%|████████▋ | 56829/65536 [9:47:51<1:29:05,  1.63it/s]
 87%|████████▋ | 56830/65536 [9:47:52<1:32:40,  1.57it/s]
 87%|████████▋ | 56831/65536 [9:47:52<1:30:13,  1.61it/s]
 87%|████████▋ | 56832/65536 [9:47:53<1:29:33,  1.62it/s]
 87%|████████▋ | 56833/65536 [9:47:54<1:28:35,  1.64it/s]
 87%|████████▋ | 56834/65536 [9:47:54<1:28:05,  1.65it/s]
 87%|████████▋ | 56835/65536 [9:47:55<1:29:18,  1.62it/s]
 87%|████████▋ | 56836/65536 [9:47:56<1:31:29,  1.58it/s]
 87%|████████▋ | 56837/65536 [9:47:56<1:32:40,  1.56it/s]
 87%|████████▋ | 56838/65536 [9:47:57<1:35:33,  1.52it/s]
 87%|████████▋ | 56839/65536 [9:47:58<1:33:44,  1.55it/s]
 87%|████████▋ | 56840/65536 [9:47:58<1:33:35,  1.55it/s]
                                                         
{'loss': 1.5564, 'learning_rate': 2.1978694747153178e-07, 'epoch': 3508.64}
+
 87%|████████▋ | 56840/65536 [9:47:58<1:33:35,  1.55it/s]
 87%|████████▋ | 56841/65536 [9:47:59<1:31:14,  1.59it/s]
 87%|████████▋ | 56842/65536 [9:47:59<1:31:02,  1.59it/s]
 87%|████████▋ | 56843/65536 [9:48:00<1:31:07,  1.59it/s]
 87%|████████▋ | 56844/65536 [9:48:01<1:30:07,  1.61it/s]
 87%|████████▋ | 56845/65536 [9:48:01<1:31:06,  1.59it/s]
 87%|████████▋ | 56846/65536 [9:48:02<1:35:39,  1.51it/s]
 87%|████████▋ | 56847/65536 [9:48:03<1:33:10,  1.55it/s]
 87%|████████▋ | 56848/65536 [9:48:03<1:32:16,  1.57it/s]
 87%|████████▋ | 56849/65536 [9:48:04<1:32:12,  1.57it/s]
 87%|████████▋ | 56850/65536 [9:48:04<1:30:26,  1.60it/s]
 87%|████████▋ | 56851/65536 [9:48:05<1:29:58,  1.61it/s]
 87%|████████▋ | 56852/65536 [9:48:06<1:30:28,  1.60it/s]
 87%|████████▋ | 56853/65536 [9:48:06<1:29:02,  1.63it/s]
 87%|████████▋ | 56854/65536 [9:48:07<1:28:33,  1.63it/s]
 87%|████████▋ | 56855/65536 [9:48:08<1:30:17,  1.60it/s]
 87%|████████▋ | 56856/65536 [9:48:08<1:29:09,  1.62it/s]
 87%|████████▋ | 56857/65536 [9:48:09<1:30:37,  1.60it/s]
 87%|████████▋ | 56858/65536 [9:48:09<1:29:41,  1.61it/s]
 87%|████████▋ | 56859/65536 [9:48:10<1:29:44,  1.61it/s]
 87%|████████▋ | 56860/65536 [9:48:11<1:27:52,  1.65it/s]
                                                         
{'loss': 1.564, 'learning_rate': 2.1951144851230564e-07, 'epoch': 3509.88}
+
 87%|████████▋ | 56860/65536 [9:48:11<1:27:52,  1.65it/s]
 87%|████████▋ | 56861/65536 [9:48:11<1:31:24,  1.58it/s]
 87%|████████▋ | 56862/65536 [9:48:12<1:30:30,  1.60it/s]
 87%|████████▋ | 56863/65536 [9:48:13<1:32:43,  1.56it/s]
 87%|████████▋ | 56864/65536 [9:48:13<1:33:48,  1.54it/s]
 87%|████████▋ | 56865/65536 [9:48:14<1:33:44,  1.54it/s]
 87%|████████▋ | 56866/65536 [9:48:15<1:39:45,  1.45it/s]
 87%|████████▋ | 56867/65536 [9:48:15<1:38:53,  1.46it/s]
 87%|████████▋ | 56868/65536 [9:48:16<1:38:10,  1.47it/s]
 87%|████████▋ | 56869/65536 [9:48:17<1:36:02,  1.50it/s]
 87%|████████▋ | 56870/65536 [9:48:17<1:36:44,  1.49it/s]
 87%|████████▋ | 56871/65536 [9:48:18<1:34:22,  1.53it/s]
 87%|████████▋ | 56872/65536 [9:48:19<1:32:34,  1.56it/s]
 87%|████████▋ | 56873/65536 [9:48:19<1:29:55,  1.61it/s]
 87%|████████▋ | 56874/65536 [9:48:20<1:29:17,  1.62it/s]
 87%|████████▋ | 56875/65536 [9:48:20<1:30:32,  1.59it/s]
 87%|████████▋ | 56876/65536 [9:48:21<1:29:43,  1.61it/s]
 87%|████████▋ | 56877/65536 [9:48:22<1:28:13,  1.64it/s]
 87%|████████▋ | 56878/65536 [9:48:22<1:28:27,  1.63it/s]
 87%|████████▋ | 56879/65536 [9:48:23<1:33:55,  1.54it/s]
 87%|████████▋ | 56880/65536 [9:48:24<1:33:05,  1.55it/s]
                                                         
{'loss': 1.5655, 'learning_rate': 2.1923594955307942e-07, 'epoch': 3511.11}
+
 87%|████████▋ | 56880/65536 [9:48:24<1:33:05,  1.55it/s]
 87%|████████▋ | 56881/65536 [9:48:24<1:32:07,  1.57it/s]
 87%|████████▋ | 56882/65536 [9:48:25<1:32:02,  1.57it/s]
 87%|████████▋ | 56883/65536 [9:48:25<1:30:24,  1.60it/s]
 87%|████████▋ | 56884/65536 [9:48:26<1:29:27,  1.61it/s]
 87%|████████▋ | 56885/65536 [9:48:27<1:28:53,  1.62it/s]
 87%|████████▋ | 56886/65536 [9:48:27<1:27:58,  1.64it/s]
 87%|████████▋ | 56887/65536 [9:48:28<1:26:54,  1.66it/s]
 87%|████████▋ | 56888/65536 [9:48:28<1:27:44,  1.64it/s]
 87%|████████▋ | 56889/65536 [9:48:29<1:29:29,  1.61it/s]
 87%|████████▋ | 56890/65536 [9:48:30<1:31:49,  1.57it/s]
 87%|████████▋ | 56891/65536 [9:48:30<1:31:49,  1.57it/s]
 87%|████████▋ | 56892/65536 [9:48:31<1:30:43,  1.59it/s]
 87%|████████▋ | 56893/65536 [9:48:32<1:29:54,  1.60it/s]
 87%|████████▋ | 56894/65536 [9:48:32<1:29:51,  1.60it/s]
 87%|████████▋ | 56895/65536 [9:48:33<1:34:01,  1.53it/s]
 87%|████████▋ | 56896/65536 [9:48:34<1:31:26,  1.57it/s]
 87%|████████▋ | 56897/65536 [9:48:34<1:30:50,  1.58it/s]
 87%|████████▋ | 56898/65536 [9:48:35<1:29:04,  1.62it/s]
 87%|████████▋ | 56899/65536 [9:48:35<1:30:58,  1.58it/s]
 87%|████████▋ | 56900/65536 [9:48:36<1:31:29,  1.57it/s]
                                                         
{'loss': 1.5342, 'learning_rate': 2.1896045059385328e-07, 'epoch': 3512.35}
+
 87%|████████▋ | 56900/65536 [9:48:36<1:31:29,  1.57it/s]
 87%|████████▋ | 56901/65536 [9:48:37<1:30:13,  1.60it/s]
 87%|████████▋ | 56902/65536 [9:48:37<1:30:28,  1.59it/s]
 87%|████████▋ | 56903/65536 [9:48:38<1:29:06,  1.61it/s]
 87%|████████▋ | 56904/65536 [9:48:39<1:27:32,  1.64it/s]
 87%|████████▋ | 56905/65536 [9:48:39<1:30:53,  1.58it/s]
 87%|████████▋ | 56906/65536 [9:48:40<1:31:06,  1.58it/s]
 87%|████████▋ | 56907/65536 [9:48:40<1:29:54,  1.60it/s]
 87%|████████▋ | 56908/65536 [9:48:41<1:28:46,  1.62it/s]
 87%|████████▋ | 56909/65536 [9:48:42<1:29:41,  1.60it/s]
 87%|████████▋ | 56910/65536 [9:48:42<1:29:17,  1.61it/s]
 87%|████████▋ | 56911/65536 [9:48:43<1:33:05,  1.54it/s]
 87%|████████▋ | 56912/65536 [9:48:44<1:31:42,  1.57it/s]
 87%|████████▋ | 56913/65536 [9:48:44<1:29:59,  1.60it/s]
 87%|████████▋ | 56914/65536 [9:48:45<1:33:10,  1.54it/s]
 87%|████████▋ | 56915/65536 [9:48:46<1:34:50,  1.51it/s]
 87%|████████▋ | 56916/65536 [9:48:46<1:33:08,  1.54it/s]
 87%|████████▋ | 56917/65536 [9:48:47<1:32:44,  1.55it/s]
 87%|████████▋ | 56918/65536 [9:48:48<1:31:31,  1.57it/s]
 87%|████████▋ | 56919/65536 [9:48:48<1:32:51,  1.55it/s]
 87%|████████▋ | 56920/65536 [9:48:49<1:32:22,  1.55it/s]
                                                         
{'loss': 1.5541, 'learning_rate': 2.1868495163462714e-07, 'epoch': 3513.58}
+
 87%|████████▋ | 56920/65536 [9:48:49<1:32:22,  1.55it/s]
 87%|████████▋ | 56921/65536 [9:48:49<1:32:20,  1.55it/s]
 87%|████████▋ | 56922/65536 [9:48:50<1:33:35,  1.53it/s]
 87%|████████▋ | 56923/65536 [9:48:51<1:34:31,  1.52it/s]
 87%|████████▋ | 56924/65536 [9:48:52<1:36:00,  1.50it/s]
 87%|████████▋ | 56925/65536 [9:48:52<1:33:56,  1.53it/s]
 87%|████████▋ | 56926/65536 [9:48:53<1:30:49,  1.58it/s]
 87%|████████▋ | 56927/65536 [9:48:53<1:32:54,  1.54it/s]
 87%|████████▋ | 56928/65536 [9:48:54<1:30:49,  1.58it/s]
 87%|████████▋ | 56929/65536 [9:48:55<1:30:02,  1.59it/s]
 87%|████████▋ | 56930/65536 [9:48:55<1:29:33,  1.60it/s]
 87%|████████▋ | 56931/65536 [9:48:56<1:30:25,  1.59it/s]
 87%|████████▋ | 56932/65536 [9:48:57<1:30:53,  1.58it/s]
 87%|████████▋ | 56933/65536 [9:48:57<1:31:07,  1.57it/s]
 87%|████████▋ | 56934/65536 [9:48:58<1:32:49,  1.54it/s]
 87%|████████▋ | 56935/65536 [9:48:58<1:29:30,  1.60it/s]
 87%|████████▋ | 56936/65536 [9:48:59<1:28:46,  1.61it/s]
 87%|████████▋ | 56937/65536 [9:49:00<1:29:37,  1.60it/s]
 87%|████████▋ | 56938/65536 [9:49:00<1:29:25,  1.60it/s]
 87%|████████▋ | 56939/65536 [9:49:01<1:28:42,  1.62it/s]
 87%|████████▋ | 56940/65536 [9:49:01<1:26:32,  1.66it/s]
                                                         
{'loss': 1.5518, 'learning_rate': 2.18409452675401e-07, 'epoch': 3514.81}
+
 87%|████████▋ | 56940/65536 [9:49:01<1:26:32,  1.66it/s]
 87%|████████▋ | 56941/65536 [9:49:02<1:25:57,  1.67it/s]
 87%|████████▋ | 56942/65536 [9:49:03<1:28:26,  1.62it/s]
 87%|████████▋ | 56943/65536 [9:49:03<1:28:05,  1.63it/s]
 87%|████████▋ | 56944/65536 [9:49:04<1:31:29,  1.57it/s]
 87%|████████▋ | 56945/65536 [9:49:05<1:29:54,  1.59it/s]
 87%|████████▋ | 56946/65536 [9:49:05<1:30:18,  1.59it/s]
 87%|████████▋ | 56947/65536 [9:49:06<1:29:34,  1.60it/s]
 87%|████████▋ | 56948/65536 [9:49:07<1:31:56,  1.56it/s]
 87%|████████▋ | 56949/65536 [9:49:07<1:29:17,  1.60it/s]
 87%|████████▋ | 56950/65536 [9:49:08<1:28:47,  1.61it/s]
 87%|████████▋ | 56951/65536 [9:49:08<1:28:21,  1.62it/s]
 87%|████████▋ | 56952/65536 [9:49:09<1:27:09,  1.64it/s]
 87%|████████▋ | 56953/65536 [9:49:10<1:29:41,  1.59it/s]
 87%|████████▋ | 56954/65536 [9:49:10<1:29:24,  1.60it/s]
 87%|████████▋ | 56955/65536 [9:49:11<1:29:38,  1.60it/s]
 87%|████████▋ | 56956/65536 [9:49:11<1:28:58,  1.61it/s]
 87%|████████▋ | 56957/65536 [9:49:12<1:30:42,  1.58it/s]
 87%|████████▋ | 56958/65536 [9:49:13<1:28:24,  1.62it/s]
 87%|████████▋ | 56959/65536 [9:49:13<1:27:38,  1.63it/s]
 87%|████████▋ | 56960/65536 [9:49:14<1:33:42,  1.53it/s]
                                                         
{'loss': 1.5251, 'learning_rate': 2.1813395371617485e-07, 'epoch': 3516.05}
+
 87%|████████▋ | 56960/65536 [9:49:14<1:33:42,  1.53it/s]
 87%|████████▋ | 56961/65536 [9:49:15<1:33:24,  1.53it/s]
 87%|████████▋ | 56962/65536 [9:49:15<1:31:16,  1.57it/s]
 87%|████████▋ | 56963/65536 [9:49:16<1:29:18,  1.60it/s]
 87%|████████▋ | 56964/65536 [9:49:16<1:27:50,  1.63it/s]
 87%|████████▋ | 56965/65536 [9:49:17<1:28:10,  1.62it/s]
 87%|████████▋ | 56966/65536 [9:49:18<1:28:57,  1.61it/s]
 87%|████████▋ | 56967/65536 [9:49:18<1:27:33,  1.63it/s]
 87%|████████▋ | 56968/65536 [9:49:19<1:27:15,  1.64it/s]
 87%|████████▋ | 56969/65536 [9:49:20<1:27:41,  1.63it/s]
 87%|████████▋ | 56970/65536 [9:49:20<1:28:21,  1.62it/s]
 87%|████████▋ | 56971/65536 [9:49:21<1:27:13,  1.64it/s]
 87%|████████▋ | 56972/65536 [9:49:21<1:27:25,  1.63it/s]
 87%|████████▋ | 56973/65536 [9:49:22<1:27:42,  1.63it/s]
 87%|████████▋ | 56974/65536 [9:49:23<1:28:34,  1.61it/s]
 87%|████████▋ | 56975/65536 [9:49:23<1:28:07,  1.62it/s]
 87%|████████▋ | 56976/65536 [9:49:24<1:31:44,  1.56it/s]
 87%|████████▋ | 56977/65536 [9:49:25<1:30:45,  1.57it/s]
 87%|████████▋ | 56978/65536 [9:49:25<1:31:46,  1.55it/s]
 87%|████████▋ | 56979/65536 [9:49:26<1:34:07,  1.52it/s]
 87%|████████▋ | 56980/65536 [9:49:27<1:35:00,  1.50it/s]
                                                         
{'loss': 1.5355, 'learning_rate': 2.1785845475694871e-07, 'epoch': 3517.28}
+
 87%|████████▋ | 56980/65536 [9:49:27<1:35:00,  1.50it/s]
 87%|████████▋ | 56981/65536 [9:49:27<1:31:57,  1.55it/s]
 87%|████████▋ | 56982/65536 [9:49:28<1:30:38,  1.57it/s]
 87%|████████▋ | 56983/65536 [9:49:28<1:30:05,  1.58it/s]
 87%|████████▋ | 56984/65536 [9:49:29<1:30:31,  1.57it/s]
 87%|████████▋ | 56985/65536 [9:49:30<1:30:17,  1.58it/s]
 87%|████████▋ | 56986/65536 [9:49:30<1:29:26,  1.59it/s]
 87%|████████▋ | 56987/65536 [9:49:31<1:27:06,  1.64it/s]
 87%|████████▋ | 56988/65536 [9:49:32<1:27:30,  1.63it/s]
 87%|████████▋ | 56989/65536 [9:49:32<1:26:21,  1.65it/s]
 87%|████████▋ | 56990/65536 [9:49:33<1:26:54,  1.64it/s]
 87%|████████▋ | 56991/65536 [9:49:33<1:25:53,  1.66it/s]
 87%|████████▋ | 56992/65536 [9:49:34<1:27:24,  1.63it/s]
 87%|████████▋ | 56993/65536 [9:49:35<1:26:56,  1.64it/s]
 87%|████████▋ | 56994/65536 [9:49:35<1:25:48,  1.66it/s]
 87%|████████▋ | 56995/65536 [9:49:36<1:29:06,  1.60it/s]
 87%|████████▋ | 56996/65536 [9:49:36<1:28:00,  1.62it/s]
 87%|████████▋ | 56997/65536 [9:49:37<1:28:37,  1.61it/s]
 87%|████████▋ | 56998/65536 [9:49:38<1:25:50,  1.66it/s]
 87%|████████▋ | 56999/65536 [9:49:38<1:25:10,  1.67it/s]
 87%|████████▋ | 57000/65536 [9:49:39<1:28:47,  1.60it/s]
                                                         
{'loss': 1.5846, 'learning_rate': 2.1758295579772257e-07, 'epoch': 3518.52}
+
 87%|████████▋ | 57000/65536 [9:49:39<1:28:47,  1.60it/s]
 87%|████████▋ | 57001/65536 [9:49:40<1:27:45,  1.62it/s]
 87%|████████▋ | 57002/65536 [9:49:40<1:26:59,  1.63it/s]
 87%|████████▋ | 57003/65536 [9:49:41<1:26:47,  1.64it/s]
 87%|████████▋ | 57004/65536 [9:49:41<1:30:10,  1.58it/s]
 87%|████████▋ | 57005/65536 [9:49:42<1:29:26,  1.59it/s]
 87%|████████▋ | 57006/65536 [9:49:43<1:29:02,  1.60it/s]
 87%|████████▋ | 57007/65536 [9:49:43<1:28:01,  1.62it/s]
 87%|████████▋ | 57008/65536 [9:49:44<1:28:55,  1.60it/s]
 87%|████████▋ | 57009/65536 [9:49:45<1:31:12,  1.56it/s]
 87%|████████▋ | 57010/65536 [9:49:45<1:31:16,  1.56it/s]
 87%|████████▋ | 57011/65536 [9:49:46<1:30:49,  1.56it/s]
 87%|████████▋ | 57012/65536 [9:49:46<1:29:05,  1.59it/s]
 87%|████████▋ | 57013/65536 [9:49:47<1:28:38,  1.60it/s]
 87%|████████▋ | 57014/65536 [9:49:48<1:28:06,  1.61it/s]
 87%|████████▋ | 57015/65536 [9:49:48<1:28:08,  1.61it/s]
 87%|████████▋ | 57016/65536 [9:49:49<1:26:30,  1.64it/s]
 87%|████████▋ | 57017/65536 [9:49:49<1:25:02,  1.67it/s]
 87%|████████▋ | 57018/65536 [9:49:50<1:25:33,  1.66it/s]
 87%|████████▋ | 57019/65536 [9:49:51<1:26:04,  1.65it/s]
 87%|████████▋ | 57020/65536 [9:49:51<1:27:44,  1.62it/s]
                                                         
{'loss': 1.5342, 'learning_rate': 2.1730745683849635e-07, 'epoch': 3519.75}
+
 87%|████████▋ | 57020/65536 [9:49:51<1:27:44,  1.62it/s]
 87%|████████▋ | 57021/65536 [9:49:52<1:29:08,  1.59it/s]
 87%|████████▋ | 57022/65536 [9:49:53<1:27:46,  1.62it/s]
 87%|████████▋ | 57023/65536 [9:49:53<1:26:45,  1.64it/s]
 87%|████████▋ | 57024/65536 [9:49:54<1:28:19,  1.61it/s]
 87%|████████▋ | 57025/65536 [9:49:54<1:30:11,  1.57it/s]
 87%|████████▋ | 57026/65536 [9:49:55<1:31:00,  1.56it/s]
 87%|████████▋ | 57027/65536 [9:49:56<1:29:06,  1.59it/s]
 87%|████████▋ | 57028/65536 [9:49:56<1:26:59,  1.63it/s]
 87%|████████▋ | 57029/65536 [9:49:57<1:25:16,  1.66it/s]
 87%|████████▋ | 57030/65536 [9:49:57<1:25:01,  1.67it/s]
 87%|████████▋ | 57031/65536 [9:49:58<1:27:10,  1.63it/s]
 87%|████████▋ | 57032/65536 [9:49:59<1:26:01,  1.65it/s]
 87%|████████▋ | 57033/65536 [9:49:59<1:26:29,  1.64it/s]
 87%|████████▋ | 57034/65536 [9:50:00<1:25:51,  1.65it/s]
 87%|████████▋ | 57035/65536 [9:50:01<1:27:57,  1.61it/s]
 87%|████████▋ | 57036/65536 [9:50:01<1:28:47,  1.60it/s]
 87%|████████▋ | 57037/65536 [9:50:02<1:30:33,  1.56it/s]
 87%|████████▋ | 57038/65536 [9:50:03<1:28:53,  1.59it/s]
 87%|████████▋ | 57039/65536 [9:50:03<1:30:11,  1.57it/s]
 87%|████████▋ | 57040/65536 [9:50:04<1:29:17,  1.59it/s]
                                                         
{'loss': 1.5399, 'learning_rate': 2.170319578792702e-07, 'epoch': 3520.99}
+
 87%|████████▋ | 57040/65536 [9:50:04<1:29:17,  1.59it/s]
 87%|████████▋ | 57041/65536 [9:50:04<1:32:14,  1.53it/s]
 87%|████████▋ | 57042/65536 [9:50:05<1:30:25,  1.57it/s]
 87%|████████▋ | 57043/65536 [9:50:06<1:31:00,  1.56it/s]
 87%|████████▋ | 57044/65536 [9:50:06<1:28:43,  1.60it/s]
 87%|████████▋ | 57045/65536 [9:50:07<1:27:41,  1.61it/s]
 87%|████████▋ | 57046/65536 [9:50:08<1:26:41,  1.63it/s]
 87%|████████▋ | 57047/65536 [9:50:08<1:27:34,  1.62it/s]
 87%|████████▋ | 57048/65536 [9:50:09<1:28:20,  1.60it/s]
 87%|████████▋ | 57049/65536 [9:50:09<1:26:50,  1.63it/s]
 87%|████████▋ | 57050/65536 [9:50:10<1:30:29,  1.56it/s]
 87%|████████▋ | 57051/65536 [9:50:11<1:27:49,  1.61it/s]
 87%|████████▋ | 57052/65536 [9:50:11<1:27:31,  1.62it/s]
 87%|████████▋ | 57053/65536 [9:50:12<1:26:58,  1.63it/s]
 87%|████████▋ | 57054/65536 [9:50:12<1:26:32,  1.63it/s]
 87%|████████▋ | 57055/65536 [9:50:13<1:26:40,  1.63it/s]
 87%|████████▋ | 57056/65536 [9:50:14<1:27:28,  1.62it/s]
 87%|████████▋ | 57057/65536 [9:50:14<1:31:31,  1.54it/s]
 87%|████████▋ | 57058/65536 [9:50:15<1:30:19,  1.56it/s]
 87%|████████▋ | 57059/65536 [9:50:16<1:28:54,  1.59it/s]
 87%|████████▋ | 57060/65536 [9:50:16<1:28:10,  1.60it/s]
                                                         
{'loss': 1.575, 'learning_rate': 2.1675645892004407e-07, 'epoch': 3522.22}
+
 87%|████████▋ | 57060/65536 [9:50:16<1:28:10,  1.60it/s]
 87%|████████▋ | 57061/65536 [9:50:17<1:27:17,  1.62it/s]
 87%|████████▋ | 57062/65536 [9:50:17<1:26:32,  1.63it/s]
 87%|████████▋ | 57063/65536 [9:50:18<1:26:52,  1.63it/s]
 87%|████████▋ | 57064/65536 [9:50:19<1:25:38,  1.65it/s]
 87%|████████▋ | 57065/65536 [9:50:19<1:28:48,  1.59it/s]
 87%|████████▋ | 57066/65536 [9:50:20<1:27:32,  1.61it/s]
 87%|████████▋ | 57067/65536 [9:50:21<1:27:04,  1.62it/s]
 87%|████████▋ | 57068/65536 [9:50:21<1:25:07,  1.66it/s]
 87%|████████▋ | 57069/65536 [9:50:22<1:24:50,  1.66it/s]
 87%|████████▋ | 57070/65536 [9:50:22<1:24:45,  1.66it/s]
 87%|████████▋ | 57071/65536 [9:50:23<1:28:28,  1.59it/s]
 87%|████████▋ | 57072/65536 [9:50:24<1:30:02,  1.57it/s]
 87%|████████▋ | 57073/65536 [9:50:24<1:31:29,  1.54it/s]
 87%|████████▋ | 57074/65536 [9:50:25<1:30:36,  1.56it/s]
 87%|████████▋ | 57075/65536 [9:50:26<1:30:02,  1.57it/s]
 87%|████████▋ | 57076/65536 [9:50:26<1:28:31,  1.59it/s]
 87%|████████▋ | 57077/65536 [9:50:27<1:26:02,  1.64it/s]
 87%|████████▋ | 57078/65536 [9:50:27<1:26:41,  1.63it/s]
 87%|████████▋ | 57079/65536 [9:50:28<1:29:52,  1.57it/s]
 87%|████████▋ | 57080/65536 [9:50:29<1:29:34,  1.57it/s]
                                                         
{'loss': 1.522, 'learning_rate': 2.1648095996081793e-07, 'epoch': 3523.46}
+
 87%|████████▋ | 57080/65536 [9:50:29<1:29:34,  1.57it/s]
 87%|████████▋ | 57081/65536 [9:50:29<1:30:20,  1.56it/s]
 87%|████████▋ | 57082/65536 [9:50:30<1:27:46,  1.61it/s]
 87%|████████▋ | 57083/65536 [9:50:31<1:25:55,  1.64it/s]
 87%|████████▋ | 57084/65536 [9:50:31<1:25:59,  1.64it/s]
 87%|████████▋ | 57085/65536 [9:50:32<1:26:30,  1.63it/s]
 87%|████████▋ | 57086/65536 [9:50:32<1:25:05,  1.66it/s]
 87%|████████▋ | 57087/65536 [9:50:33<1:23:35,  1.68it/s]
 87%|████████▋ | 57088/65536 [9:50:34<1:25:04,  1.65it/s]
 87%|████████▋ | 57089/65536 [9:50:34<1:28:30,  1.59it/s]
 87%|████████▋ | 57090/65536 [9:50:35<1:26:19,  1.63it/s]
 87%|████████▋ | 57091/65536 [9:50:35<1:27:25,  1.61it/s]
 87%|████████▋ | 57092/65536 [9:50:36<1:27:26,  1.61it/s]
 87%|████████▋ | 57093/65536 [9:50:37<1:27:25,  1.61it/s]
 87%|████████▋ | 57094/65536 [9:50:37<1:27:31,  1.61it/s]
 87%|████████▋ | 57095/65536 [9:50:38<1:26:51,  1.62it/s]
 87%|████████▋ | 57096/65536 [9:50:39<1:26:38,  1.62it/s]
 87%|████████▋ | 57097/65536 [9:50:39<1:28:25,  1.59it/s]
 87%|████████▋ | 57098/65536 [9:50:40<1:26:28,  1.63it/s]
 87%|████████▋ | 57099/65536 [9:50:40<1:29:10,  1.58it/s]
 87%|████████▋ | 57100/65536 [9:50:41<1:27:29,  1.61it/s]
                                                         
{'loss': 1.5412, 'learning_rate': 2.162054610015918e-07, 'epoch': 3524.69}
+
 87%|████████▋ | 57100/65536 [9:50:41<1:27:29,  1.61it/s]
 87%|████████▋ | 57101/65536 [9:50:42<1:26:41,  1.62it/s]
 87%|████████▋ | 57102/65536 [9:50:42<1:27:08,  1.61it/s]
 87%|████████▋ | 57103/65536 [9:50:43<1:28:08,  1.59it/s]
 87%|████████▋ | 57104/65536 [9:50:44<1:27:44,  1.60it/s]
 87%|████████▋ | 57105/65536 [9:50:44<1:26:16,  1.63it/s]
 87%|████████▋ | 57106/65536 [9:50:45<1:30:14,  1.56it/s]
 87%|████████▋ | 57107/65536 [9:50:46<1:30:42,  1.55it/s]
 87%|████████▋ | 57108/65536 [9:50:46<1:28:08,  1.59it/s]
 87%|████████▋ | 57109/65536 [9:50:47<1:29:07,  1.58it/s]
 87%|████████▋ | 57110/65536 [9:50:47<1:27:52,  1.60it/s]
 87%|████████▋ | 57111/65536 [9:50:48<1:28:09,  1.59it/s]
 87%|████████▋ | 57112/65536 [9:50:49<1:26:16,  1.63it/s]
 87%|████████▋ | 57113/65536 [9:50:49<1:24:51,  1.65it/s]
 87%|████████▋ | 57114/65536 [9:50:50<1:24:38,  1.66it/s]
 87%|████████▋ | 57115/65536 [9:50:50<1:25:14,  1.65it/s]
 87%|████████▋ | 57116/65536 [9:50:51<1:25:16,  1.65it/s]
 87%|████████▋ | 57117/65536 [9:50:52<1:24:58,  1.65it/s]
 87%|████████▋ | 57118/65536 [9:50:52<1:24:59,  1.65it/s]
 87%|████████▋ | 57119/65536 [9:50:53<1:26:14,  1.63it/s]
 87%|████████▋ | 57120/65536 [9:50:53<1:26:30,  1.62it/s]
                                                         
{'loss': 1.5715, 'learning_rate': 2.1592996204236567e-07, 'epoch': 3525.93}
+
 87%|████████▋ | 57120/65536 [9:50:53<1:26:30,  1.62it/s]
 87%|████████▋ | 57121/65536 [9:50:54<1:27:48,  1.60it/s]
 87%|████████▋ | 57122/65536 [9:50:55<1:29:15,  1.57it/s]
 87%|████████▋ | 57123/65536 [9:50:55<1:28:08,  1.59it/s]
 87%|████████▋ | 57124/65536 [9:50:56<1:29:20,  1.57it/s]
 87%|████████▋ | 57125/65536 [9:50:57<1:27:20,  1.60it/s]
 87%|████████▋ | 57126/65536 [9:50:57<1:26:21,  1.62it/s]
 87%|████████▋ | 57127/65536 [9:50:58<1:25:43,  1.63it/s]
 87%|████████▋ | 57128/65536 [9:50:58<1:26:00,  1.63it/s]
 87%|████████▋ | 57129/65536 [9:50:59<1:27:22,  1.60it/s]
 87%|████████▋ | 57130/65536 [9:51:00<1:25:44,  1.63it/s]
 87%|████████▋ | 57131/65536 [9:51:00<1:26:14,  1.62it/s]
 87%|████████▋ | 57132/65536 [9:51:01<1:28:15,  1.59it/s]
 87%|████████▋ | 57133/65536 [9:51:02<1:28:02,  1.59it/s]
 87%|████████▋ | 57134/65536 [9:51:02<1:25:32,  1.64it/s]
 87%|████████▋ | 57135/65536 [9:51:03<1:26:32,  1.62it/s]
 87%|████████▋ | 57136/65536 [9:51:03<1:27:56,  1.59it/s]
 87%|████████▋ | 57137/65536 [9:51:04<1:26:37,  1.62it/s]
 87%|████████▋ | 57138/65536 [9:51:05<1:28:21,  1.58it/s]
 87%|████████▋ | 57139/65536 [9:51:05<1:27:42,  1.60it/s]
 87%|████████▋ | 57140/65536 [9:51:06<1:26:25,  1.62it/s]
                                                         
{'loss': 1.5668, 'learning_rate': 2.156544630831394e-07, 'epoch': 3527.16}
+
 87%|████████▋ | 57140/65536 [9:51:06<1:26:25,  1.62it/s]
 87%|████████▋ | 57141/65536 [9:51:06<1:24:30,  1.66it/s]
 87%|████████▋ | 57142/65536 [9:51:07<1:25:00,  1.65it/s]
 87%|████████▋ | 57143/65536 [9:51:08<1:28:20,  1.58it/s]
 87%|████████▋ | 57144/65536 [9:51:08<1:26:06,  1.62it/s]
 87%|████████▋ | 57145/65536 [9:51:09<1:24:26,  1.66it/s]
 87%|████████▋ | 57146/65536 [9:51:10<1:25:15,  1.64it/s]
 87%|████████▋ | 57147/65536 [9:51:10<1:25:51,  1.63it/s]
 87%|████████▋ | 57148/65536 [9:51:11<1:27:03,  1.61it/s]
 87%|████████▋ | 57149/65536 [9:51:11<1:25:48,  1.63it/s]
 87%|████████▋ | 57150/65536 [9:51:12<1:25:36,  1.63it/s]
 87%|████████▋ | 57151/65536 [9:51:13<1:24:47,  1.65it/s]
 87%|████████▋ | 57152/65536 [9:51:13<1:25:23,  1.64it/s]
 87%|████████▋ | 57153/65536 [9:51:14<1:24:31,  1.65it/s]
 87%|████████▋ | 57154/65536 [9:51:15<1:27:31,  1.60it/s]
 87%|████████▋ | 57155/65536 [9:51:15<1:26:07,  1.62it/s]
 87%|████████▋ | 57156/65536 [9:51:16<1:27:05,  1.60it/s]
 87%|████████▋ | 57157/65536 [9:51:16<1:25:27,  1.63it/s]
 87%|████████▋ | 57158/65536 [9:51:17<1:25:00,  1.64it/s]
 87%|████████▋ | 57159/65536 [9:51:18<1:24:22,  1.65it/s]
 87%|████████▋ | 57160/65536 [9:51:18<1:24:37,  1.65it/s]
                                                         
{'loss': 1.569, 'learning_rate': 2.1537896412391326e-07, 'epoch': 3528.4}
+
 87%|████████▋ | 57160/65536 [9:51:18<1:24:37,  1.65it/s]
 87%|████████▋ | 57161/65536 [9:51:19<1:25:55,  1.62it/s]
 87%|████████▋ | 57162/65536 [9:51:19<1:25:40,  1.63it/s]
 87%|████████▋ | 57163/65536 [9:51:20<1:24:20,  1.65it/s]
 87%|████████▋ | 57164/65536 [9:51:21<1:24:55,  1.64it/s]
 87%|████████▋ | 57165/65536 [9:51:21<1:27:58,  1.59it/s]
 87%|████████▋ | 57166/65536 [9:51:22<1:27:46,  1.59it/s]
 87%|████████▋ | 57167/65536 [9:51:23<1:26:25,  1.61it/s]
 87%|████████▋ | 57168/65536 [9:51:23<1:27:15,  1.60it/s]
 87%|████████▋ | 57169/65536 [9:51:24<1:27:43,  1.59it/s]
 87%|████████▋ | 57170/65536 [9:51:24<1:28:33,  1.57it/s]
 87%|████████▋ | 57171/65536 [9:51:25<1:28:05,  1.58it/s]
 87%|████████▋ | 57172/65536 [9:51:26<1:27:42,  1.59it/s]
 87%|████████▋ | 57173/65536 [9:51:26<1:28:18,  1.58it/s]
 87%|████████▋ | 57174/65536 [9:51:27<1:29:30,  1.56it/s]
 87%|████████▋ | 57175/65536 [9:51:28<1:28:48,  1.57it/s]
 87%|████████▋ | 57176/65536 [9:51:28<1:27:58,  1.58it/s]
 87%|████████▋ | 57177/65536 [9:51:29<1:25:30,  1.63it/s]
 87%|████████▋ | 57178/65536 [9:51:29<1:23:54,  1.66it/s]
 87%|████████▋ | 57179/65536 [9:51:30<1:23:00,  1.68it/s]
 87%|████████▋ | 57180/65536 [9:51:31<1:22:43,  1.68it/s]
                                                         
{'loss': 1.5271, 'learning_rate': 2.1510346516468712e-07, 'epoch': 3529.63}
+
 87%|████████▋ | 57180/65536 [9:51:31<1:22:43,  1.68it/s]
 87%|████████▋ | 57181/65536 [9:51:31<1:24:31,  1.65it/s]
 87%|████████▋ | 57182/65536 [9:51:32<1:23:06,  1.68it/s]
 87%|████████▋ | 57183/65536 [9:51:32<1:23:36,  1.67it/s]
 87%|████████▋ | 57184/65536 [9:51:33<1:26:25,  1.61it/s]
 87%|████████▋ | 57185/65536 [9:51:34<1:26:18,  1.61it/s]
 87%|████████▋ | 57186/65536 [9:51:34<1:26:01,  1.62it/s]
 87%|████████▋ | 57187/65536 [9:51:35<1:29:19,  1.56it/s]
 87%|████████▋ | 57188/65536 [9:51:36<1:29:16,  1.56it/s]
 87%|████████▋ | 57189/65536 [9:51:36<1:29:38,  1.55it/s]
 87%|████████▋ | 57190/65536 [9:51:37<1:27:06,  1.60it/s]
 87%|████████▋ | 57191/65536 [9:51:37<1:27:34,  1.59it/s]
 87%|████████▋ | 57192/65536 [9:51:38<1:26:27,  1.61it/s]
 87%|████████▋ | 57193/65536 [9:51:39<1:25:01,  1.64it/s]
 87%|████████▋ | 57194/65536 [9:51:39<1:25:12,  1.63it/s]
 87%|████████▋ | 57195/65536 [9:51:40<1:24:46,  1.64it/s]
 87%|████████▋ | 57196/65536 [9:51:41<1:26:02,  1.62it/s]
 87%|████████▋ | 57197/65536 [9:51:41<1:24:45,  1.64it/s]
 87%|████████▋ | 57198/65536 [9:51:42<1:24:01,  1.65it/s]
 87%|████████▋ | 57199/65536 [9:51:42<1:24:01,  1.65it/s]
 87%|████████▋ | 57200/65536 [9:51:43<1:23:01,  1.67it/s]
                                                         
{'loss': 1.5394, 'learning_rate': 2.1482796620546098e-07, 'epoch': 3530.86}
+
 87%|████████▋ | 57200/65536 [9:51:43<1:23:01,  1.67it/s]
 87%|████████▋ | 57201/65536 [9:51:44<1:25:45,  1.62it/s]
 87%|████████▋ | 57202/65536 [9:51:44<1:25:15,  1.63it/s]
 87%|████████▋ | 57203/65536 [9:51:45<1:27:01,  1.60it/s]
 87%|████████▋ | 57204/65536 [9:51:45<1:26:07,  1.61it/s]
 87%|████████▋ | 57205/65536 [9:51:46<1:26:54,  1.60it/s]
 87%|████████▋ | 57206/65536 [9:51:47<1:30:03,  1.54it/s]
 87%|████████▋ | 57207/65536 [9:51:47<1:28:20,  1.57it/s]
 87%|████████▋ | 57208/65536 [9:51:48<1:26:29,  1.60it/s]
 87%|████████▋ | 57209/65536 [9:51:49<1:27:07,  1.59it/s]
 87%|████████▋ | 57210/65536 [9:51:49<1:26:59,  1.60it/s]
 87%|████████▋ | 57211/65536 [9:51:50<1:26:42,  1.60it/s]
 87%|████████▋ | 57212/65536 [9:51:50<1:26:09,  1.61it/s]
 87%|████████▋ | 57213/65536 [9:51:51<1:25:06,  1.63it/s]
 87%|████████▋ | 57214/65536 [9:51:52<1:26:28,  1.60it/s]
 87%|████████▋ | 57215/65536 [9:51:52<1:24:46,  1.64it/s]
 87%|████████▋ | 57216/65536 [9:51:53<1:26:22,  1.61it/s]
 87%|████████▋ | 57217/65536 [9:51:54<1:26:44,  1.60it/s]
 87%|████████▋ | 57218/65536 [9:51:54<1:24:01,  1.65it/s]
 87%|████████▋ | 57219/65536 [9:51:55<1:26:52,  1.60it/s]
 87%|████████▋ | 57220/65536 [9:51:55<1:28:20,  1.57it/s]
                                                         
{'loss': 1.5728, 'learning_rate': 2.1455246724623483e-07, 'epoch': 3532.1}
+
 87%|████████▋ | 57220/65536 [9:51:55<1:28:20,  1.57it/s]
 87%|████████▋ | 57221/65536 [9:51:56<1:28:09,  1.57it/s]
 87%|████████▋ | 57222/65536 [9:51:57<1:27:08,  1.59it/s]
 87%|████████▋ | 57223/65536 [9:51:57<1:27:19,  1.59it/s]
 87%|████████▋ | 57224/65536 [9:51:58<1:24:57,  1.63it/s]
 87%|████████▋ | 57225/65536 [9:51:59<1:26:56,  1.59it/s]
 87%|████████▋ | 57226/65536 [9:51:59<1:25:43,  1.62it/s]
 87%|████████▋ | 57227/65536 [9:52:00<1:23:35,  1.66it/s]
 87%|████████▋ | 57228/65536 [9:52:00<1:23:05,  1.67it/s]
 87%|████████▋ | 57229/65536 [9:52:01<1:23:38,  1.66it/s]
 87%|████████▋ | 57230/65536 [9:52:02<1:24:10,  1.64it/s]
 87%|████████▋ | 57231/65536 [9:52:02<1:22:12,  1.68it/s]
 87%|████████▋ | 57232/65536 [9:52:03<1:24:02,  1.65it/s]
 87%|████████▋ | 57233/65536 [9:52:03<1:24:05,  1.65it/s]
 87%|████████▋ | 57234/65536 [9:52:04<1:23:32,  1.66it/s]
 87%|████████▋ | 57235/65536 [9:52:05<1:26:27,  1.60it/s]
 87%|████████▋ | 57236/65536 [9:52:05<1:25:34,  1.62it/s]
 87%|████████▋ | 57237/65536 [9:52:06<1:24:56,  1.63it/s]
 87%|████████▋ | 57238/65536 [9:52:06<1:24:13,  1.64it/s]
 87%|████████▋ | 57239/65536 [9:52:07<1:24:21,  1.64it/s]
 87%|████████▋ | 57240/65536 [9:52:08<1:23:35,  1.65it/s]
                                                         
{'loss': 1.6056, 'learning_rate': 2.142769682870087e-07, 'epoch': 3533.33}
+
 87%|████████▋ | 57240/65536 [9:52:08<1:23:35,  1.65it/s]
 87%|████████▋ | 57241/65536 [9:52:08<1:24:55,  1.63it/s]
 87%|████████▋ | 57242/65536 [9:52:09<1:25:10,  1.62it/s]
 87%|████████▋ | 57243/65536 [9:52:10<1:25:27,  1.62it/s]
 87%|████████▋ | 57244/65536 [9:52:10<1:24:01,  1.64it/s]
 87%|████████▋ | 57245/65536 [9:52:11<1:24:06,  1.64it/s]
 87%|████████▋ | 57246/65536 [9:52:11<1:22:02,  1.68it/s]
 87%|████████▋ | 57247/65536 [9:52:12<1:22:45,  1.67it/s]
 87%|████████▋ | 57248/65536 [9:52:13<1:24:05,  1.64it/s]
 87%|████████▋ | 57249/65536 [9:52:13<1:26:08,  1.60it/s]
 87%|████████▋ | 57250/65536 [9:52:14<1:25:09,  1.62it/s]
 87%|████████▋ | 57251/65536 [9:52:14<1:27:27,  1.58it/s]
 87%|████████▋ | 57252/65536 [9:52:15<1:30:13,  1.53it/s]
 87%|████████▋ | 57253/65536 [9:52:16<1:27:49,  1.57it/s]
 87%|████████▋ | 57254/65536 [9:52:16<1:26:16,  1.60it/s]
 87%|████████▋ | 57255/65536 [9:52:17<1:25:58,  1.61it/s]
 87%|████████▋ | 57256/65536 [9:52:18<1:25:38,  1.61it/s]
 87%|████████▋ | 57257/65536 [9:52:18<1:24:36,  1.63it/s]
 87%|████████▋ | 57258/65536 [9:52:19<1:27:09,  1.58it/s]
 87%|████████▋ | 57259/65536 [9:52:19<1:25:55,  1.61it/s]
 87%|████████▋ | 57260/65536 [9:52:20<1:26:17,  1.60it/s]
                                                         
{'loss': 1.5627, 'learning_rate': 2.1400146932778255e-07, 'epoch': 3534.57}
+
 87%|████████▋ | 57260/65536 [9:52:20<1:26:17,  1.60it/s]
 87%|████████▋ | 57261/65536 [9:52:21<1:24:57,  1.62it/s]
 87%|████████▋ | 57262/65536 [9:52:21<1:23:48,  1.65it/s]
 87%|████████▋ | 57263/65536 [9:52:22<1:23:06,  1.66it/s]
 87%|████████▋ | 57264/65536 [9:52:22<1:23:53,  1.64it/s]
 87%|████████▋ | 57265/65536 [9:52:23<1:22:28,  1.67it/s]
 87%|████████▋ | 57266/65536 [9:52:24<1:24:07,  1.64it/s]
 87%|████████▋ | 57267/65536 [9:52:24<1:30:26,  1.52it/s]
 87%|████████▋ | 57268/65536 [9:52:25<1:30:51,  1.52it/s]
 87%|████████▋ | 57269/65536 [9:52:26<1:31:40,  1.50it/s]
 87%|████████▋ | 57270/65536 [9:52:26<1:29:20,  1.54it/s]
 87%|████████▋ | 57271/65536 [9:52:27<1:30:04,  1.53it/s]
 87%|████████▋ | 57272/65536 [9:52:28<1:26:59,  1.58it/s]
 87%|████████▋ | 57273/65536 [9:52:28<1:28:33,  1.56it/s]
 87%|████████▋ | 57274/65536 [9:52:29<1:27:23,  1.58it/s]
 87%|████████▋ | 57275/65536 [9:52:30<1:26:01,  1.60it/s]
 87%|████████▋ | 57276/65536 [9:52:30<1:24:17,  1.63it/s]
 87%|████████▋ | 57277/65536 [9:52:31<1:24:46,  1.62it/s]
 87%|████████▋ | 57278/65536 [9:52:31<1:26:44,  1.59it/s]
 87%|████████▋ | 57279/65536 [9:52:32<1:25:49,  1.60it/s]
 87%|████████▋ | 57280/65536 [9:52:33<1:26:13,  1.60it/s]
                                                         
{'loss': 1.5417, 'learning_rate': 2.1372597036855633e-07, 'epoch': 3535.8}
+
 87%|████████▋ | 57280/65536 [9:52:33<1:26:13,  1.60it/s]
 87%|████████▋ | 57281/65536 [9:52:33<1:24:45,  1.62it/s]
 87%|████████▋ | 57282/65536 [9:52:34<1:24:55,  1.62it/s]
 87%|████████▋ | 57283/65536 [9:52:34<1:23:35,  1.65it/s]
 87%|████████▋ | 57284/65536 [9:52:35<1:26:06,  1.60it/s]
 87%|████████▋ | 57285/65536 [9:52:36<1:28:05,  1.56it/s]
 87%|████████▋ | 57286/65536 [9:52:36<1:26:53,  1.58it/s]
 87%|████████▋ | 57287/65536 [9:52:37<1:26:05,  1.60it/s]
 87%|████████▋ | 57288/65536 [9:52:38<1:23:53,  1.64it/s]
 87%|████████▋ | 57289/65536 [9:52:38<1:23:31,  1.65it/s]
 87%|████████▋ | 57290/65536 [9:52:39<1:24:15,  1.63it/s]
 87%|████████▋ | 57291/65536 [9:52:39<1:25:00,  1.62it/s]
 87%|████████▋ | 57292/65536 [9:52:40<1:26:56,  1.58it/s]
 87%|████████▋ | 57293/65536 [9:52:41<1:25:36,  1.60it/s]
 87%|████████▋ | 57294/65536 [9:52:41<1:24:31,  1.63it/s]
 87%|████████▋ | 57295/65536 [9:52:42<1:24:30,  1.63it/s]
 87%|████████▋ | 57296/65536 [9:52:43<1:25:25,  1.61it/s]
 87%|████████▋ | 57297/65536 [9:52:43<1:24:59,  1.62it/s]
 87%|████████▋ | 57298/65536 [9:52:44<1:23:24,  1.65it/s]
 87%|████████▋ | 57299/65536 [9:52:44<1:22:04,  1.67it/s]
 87%|████████▋ | 57300/65536 [9:52:45<1:24:49,  1.62it/s]
                                                         
{'loss': 1.5812, 'learning_rate': 2.134504714093302e-07, 'epoch': 3537.04}
+
 87%|████████▋ | 57300/65536 [9:52:45<1:24:49,  1.62it/s]
 87%|████████▋ | 57301/65536 [9:52:46<1:25:57,  1.60it/s]
 87%|████████▋ | 57302/65536 [9:52:46<1:24:57,  1.62it/s]
 87%|████████▋ | 57303/65536 [9:52:47<1:26:01,  1.59it/s]
 87%|████████▋ | 57304/65536 [9:52:47<1:23:57,  1.63it/s]
 87%|████████▋ | 57305/65536 [9:52:48<1:23:26,  1.64it/s]
 87%|████████▋ | 57306/65536 [9:52:49<1:24:54,  1.62it/s]
 87%|████████▋ | 57307/65536 [9:52:49<1:23:52,  1.64it/s]
 87%|████████▋ | 57308/65536 [9:52:50<1:26:22,  1.59it/s]
 87%|████████▋ | 57309/65536 [9:52:51<1:25:48,  1.60it/s]
 87%|████████▋ | 57310/65536 [9:52:51<1:26:11,  1.59it/s]
 87%|████████▋ | 57311/65536 [9:52:52<1:27:22,  1.57it/s]
 87%|████████▋ | 57312/65536 [9:52:53<1:25:43,  1.60it/s]
 87%|████████▋ | 57313/65536 [9:52:53<1:25:59,  1.59it/s]
 87%|████████▋ | 57314/65536 [9:52:54<1:25:27,  1.60it/s]
 87%|████████▋ | 57315/65536 [9:52:54<1:24:34,  1.62it/s]
 87%|████████▋ | 57316/65536 [9:52:55<1:29:55,  1.52it/s]
 87%|████████▋ | 57317/65536 [9:52:56<1:28:53,  1.54it/s]
 87%|████████▋ | 57318/65536 [9:52:56<1:26:13,  1.59it/s]
 87%|████████▋ | 57319/65536 [9:52:57<1:28:04,  1.56it/s]
 87%|████████▋ | 57320/65536 [9:52:58<1:26:12,  1.59it/s]
                                                         
{'loss': 1.5156, 'learning_rate': 2.1317497245010405e-07, 'epoch': 3538.27}
+
 87%|████████▋ | 57320/65536 [9:52:58<1:26:12,  1.59it/s]
 87%|████████▋ | 57321/65536 [9:52:58<1:25:56,  1.59it/s]
 87%|████████▋ | 57322/65536 [9:52:59<1:23:29,  1.64it/s]
 87%|████████▋ | 57323/65536 [9:52:59<1:25:51,  1.59it/s]
 87%|████████▋ | 57324/65536 [9:53:00<1:27:09,  1.57it/s]
 87%|████████▋ | 57325/65536 [9:53:01<1:26:38,  1.58it/s]
 87%|████████▋ | 57326/65536 [9:53:01<1:25:41,  1.60it/s]
 87%|████████▋ | 57327/65536 [9:53:02<1:24:05,  1.63it/s]
 87%|████████▋ | 57328/65536 [9:53:03<1:23:43,  1.63it/s]
 87%|████████▋ | 57329/65536 [9:53:03<1:23:26,  1.64it/s]
 87%|████████▋ | 57330/65536 [9:53:04<1:24:36,  1.62it/s]
 87%|████████▋ | 57331/65536 [9:53:04<1:24:17,  1.62it/s]
 87%|████████▋ | 57332/65536 [9:53:05<1:25:40,  1.60it/s]
 87%|████████▋ | 57333/65536 [9:53:06<1:28:03,  1.55it/s]
 87%|████████▋ | 57334/65536 [9:53:06<1:27:04,  1.57it/s]
 87%|████████▋ | 57335/65536 [9:53:07<1:27:38,  1.56it/s]
 87%|████████▋ | 57336/65536 [9:53:08<1:24:45,  1.61it/s]
 87%|████████▋ | 57337/65536 [9:53:08<1:24:26,  1.62it/s]
 87%|████████▋ | 57338/65536 [9:53:09<1:26:20,  1.58it/s]
 87%|████████▋ | 57339/65536 [9:53:09<1:24:43,  1.61it/s]
 87%|████████▋ | 57340/65536 [9:53:10<1:24:28,  1.62it/s]
                                                         
{'loss': 1.521, 'learning_rate': 2.128994734908779e-07, 'epoch': 3539.51}
+
 87%|████████▋ | 57340/65536 [9:53:10<1:24:28,  1.62it/s]
 87%|████████▋ | 57341/65536 [9:53:11<1:25:14,  1.60it/s]
 87%|████████▋ | 57342/65536 [9:53:11<1:24:43,  1.61it/s]
 87%|████████▋ | 57343/65536 [9:53:12<1:24:13,  1.62it/s]
 88%|████████▊ | 57344/65536 [9:53:12<1:22:37,  1.65it/s]
 88%|████████▊ | 57345/65536 [9:53:13<1:23:25,  1.64it/s]
 88%|████████▊ | 57346/65536 [9:53:14<1:22:39,  1.65it/s]
 88%|████████▊ | 57347/65536 [9:53:14<1:22:17,  1.66it/s]
 88%|████████▊ | 57348/65536 [9:53:15<1:22:33,  1.65it/s]
 88%|████████▊ | 57349/65536 [9:53:16<1:25:58,  1.59it/s]
 88%|████████▊ | 57350/65536 [9:53:16<1:23:56,  1.63it/s]
 88%|████████▊ | 57351/65536 [9:53:17<1:23:22,  1.64it/s]
 88%|████████▊ | 57352/65536 [9:53:17<1:22:40,  1.65it/s]
 88%|████████▊ | 57353/65536 [9:53:18<1:23:13,  1.64it/s]
 88%|████████▊ | 57354/65536 [9:53:19<1:24:05,  1.62it/s]
 88%|████████▊ | 57355/65536 [9:53:19<1:25:14,  1.60it/s]
 88%|████████▊ | 57356/65536 [9:53:20<1:24:22,  1.62it/s]
 88%|████████▊ | 57357/65536 [9:53:21<1:26:11,  1.58it/s]
 88%|████████▊ | 57358/65536 [9:53:21<1:26:21,  1.58it/s]
 88%|████████▊ | 57359/65536 [9:53:22<1:23:22,  1.63it/s]
 88%|████████▊ | 57360/65536 [9:53:22<1:22:16,  1.66it/s]
                                                         
{'loss': 1.5698, 'learning_rate': 2.1262397453165177e-07, 'epoch': 3540.74}
+
 88%|████████▊ | 57360/65536 [9:53:22<1:22:16,  1.66it/s]
 88%|████████▊ | 57361/65536 [9:53:23<1:21:04,  1.68it/s]
 88%|████████▊ | 57362/65536 [9:53:24<1:22:18,  1.66it/s]
 88%|████████▊ | 57363/65536 [9:53:24<1:24:45,  1.61it/s]
 88%|████████▊ | 57364/65536 [9:53:25<1:24:38,  1.61it/s]
 88%|████████▊ | 57365/65536 [9:53:25<1:26:16,  1.58it/s]
 88%|████████▊ | 57366/65536 [9:53:26<1:26:25,  1.58it/s]
 88%|████████▊ | 57367/65536 [9:53:27<1:25:51,  1.59it/s]
 88%|████████▊ | 57368/65536 [9:53:27<1:25:14,  1.60it/s]
 88%|████████▊ | 57369/65536 [9:53:28<1:24:23,  1.61it/s]
 88%|████████▊ | 57370/65536 [9:53:29<1:25:48,  1.59it/s]
 88%|████████▊ | 57371/65536 [9:53:29<1:24:45,  1.61it/s]
 88%|████████▊ | 57372/65536 [9:53:30<1:22:49,  1.64it/s]
 88%|████████▊ | 57373/65536 [9:53:30<1:25:06,  1.60it/s]
 88%|████████▊ | 57374/65536 [9:53:31<1:23:43,  1.62it/s]
 88%|████████▊ | 57375/65536 [9:53:32<1:22:35,  1.65it/s]
 88%|████████▊ | 57376/65536 [9:53:32<1:22:26,  1.65it/s]
 88%|████████▊ | 57377/65536 [9:53:33<1:22:33,  1.65it/s]
 88%|████████▊ | 57378/65536 [9:53:33<1:21:53,  1.66it/s]
 88%|████████▊ | 57379/65536 [9:53:34<1:23:55,  1.62it/s]
 88%|████████▊ | 57380/65536 [9:53:35<1:26:32,  1.57it/s]
                                                         
{'loss': 1.5729, 'learning_rate': 2.1234847557242565e-07, 'epoch': 3541.98}
+
 88%|████████▊ | 57380/65536 [9:53:35<1:26:32,  1.57it/s]
 88%|████████▊ | 57381/65536 [9:53:36<1:31:48,  1.48it/s]
 88%|████████▊ | 57382/65536 [9:53:36<1:28:46,  1.53it/s]
 88%|████████▊ | 57383/65536 [9:53:37<1:26:06,  1.58it/s]
 88%|████████▊ | 57384/65536 [9:53:37<1:25:28,  1.59it/s]
 88%|████████▊ | 57385/65536 [9:53:38<1:25:00,  1.60it/s]
 88%|████████▊ | 57386/65536 [9:53:39<1:24:09,  1.61it/s]
 88%|████████▊ | 57387/65536 [9:53:39<1:22:38,  1.64it/s]
 88%|████████▊ | 57388/65536 [9:53:40<1:25:39,  1.59it/s]
 88%|████████▊ | 57389/65536 [9:53:40<1:24:14,  1.61it/s]
 88%|████████▊ | 57390/65536 [9:53:41<1:23:32,  1.63it/s]
 88%|████████▊ | 57391/65536 [9:53:42<1:22:43,  1.64it/s]
 88%|████████▊ | 57392/65536 [9:53:42<1:23:20,  1.63it/s]
 88%|████████▊ | 57393/65536 [9:53:43<1:21:54,  1.66it/s]
 88%|████████▊ | 57394/65536 [9:53:43<1:22:00,  1.65it/s]
 88%|████████▊ | 57395/65536 [9:53:44<1:22:25,  1.65it/s]
 88%|████████▊ | 57396/65536 [9:53:45<1:22:34,  1.64it/s]
 88%|████████▊ | 57397/65536 [9:53:45<1:25:51,  1.58it/s]
 88%|████████▊ | 57398/65536 [9:53:46<1:25:17,  1.59it/s]
 88%|████████▊ | 57399/65536 [9:53:47<1:25:38,  1.58it/s]
 88%|████████▊ | 57400/65536 [9:53:47<1:23:20,  1.63it/s]
                                                         
{'loss': 1.5858, 'learning_rate': 2.120729766131995e-07, 'epoch': 3543.21}
+
 88%|████████▊ | 57400/65536 [9:53:47<1:23:20,  1.63it/s]
 88%|████████▊ | 57401/65536 [9:53:48<1:24:21,  1.61it/s]
 88%|████████▊ | 57402/65536 [9:53:48<1:22:33,  1.64it/s]
 88%|████████▊ | 57403/65536 [9:53:49<1:23:39,  1.62it/s]
 88%|████████▊ | 57404/65536 [9:53:50<1:23:11,  1.63it/s]
 88%|████████▊ | 57405/65536 [9:53:50<1:26:44,  1.56it/s]
 88%|████████▊ | 57406/65536 [9:53:51<1:25:01,  1.59it/s]
 88%|████████▊ | 57407/65536 [9:53:52<1:23:50,  1.62it/s]
 88%|████████▊ | 57408/65536 [9:53:52<1:23:58,  1.61it/s]
 88%|████████▊ | 57409/65536 [9:53:53<1:22:44,  1.64it/s]
 88%|████████▊ | 57410/65536 [9:53:53<1:23:42,  1.62it/s]
 88%|████████▊ | 57411/65536 [9:53:54<1:22:06,  1.65it/s]
 88%|████████▊ | 57412/65536 [9:53:55<1:23:40,  1.62it/s]
 88%|████████▊ | 57413/65536 [9:53:55<1:24:41,  1.60it/s]
 88%|████████▊ | 57414/65536 [9:53:56<1:24:20,  1.60it/s]
 88%|████████▊ | 57415/65536 [9:53:56<1:23:07,  1.63it/s]
 88%|████████▊ | 57416/65536 [9:53:57<1:22:03,  1.65it/s]
 88%|████████▊ | 57417/65536 [9:53:58<1:22:13,  1.65it/s]
 88%|████████▊ | 57418/65536 [9:53:58<1:23:37,  1.62it/s]
 88%|████████▊ | 57419/65536 [9:53:59<1:23:17,  1.62it/s]
 88%|████████▊ | 57420/65536 [9:54:00<1:25:48,  1.58it/s]
                                                         
{'loss': 1.5581, 'learning_rate': 2.1179747765397326e-07, 'epoch': 3544.44}
+
 88%|████████▊ | 57420/65536 [9:54:00<1:25:48,  1.58it/s]
 88%|████████▊ | 57421/65536 [9:54:00<1:23:36,  1.62it/s]
 88%|████████▊ | 57422/65536 [9:54:01<1:22:55,  1.63it/s]
 88%|████████▊ | 57423/65536 [9:54:01<1:22:07,  1.65it/s]
 88%|████████▊ | 57424/65536 [9:54:02<1:23:47,  1.61it/s]
 88%|████████▊ | 57425/65536 [9:54:03<1:24:19,  1.60it/s]
 88%|████████▊ | 57426/65536 [9:54:03<1:22:47,  1.63it/s]
 88%|████████▊ | 57427/65536 [9:54:04<1:23:36,  1.62it/s]
 88%|████████▊ | 57428/65536 [9:54:05<1:25:21,  1.58it/s]
 88%|████████▊ | 57429/65536 [9:54:05<1:25:07,  1.59it/s]
 88%|████████▊ | 57430/65536 [9:54:06<1:28:01,  1.53it/s]
 88%|████████▊ | 57431/65536 [9:54:06<1:26:14,  1.57it/s]
 88%|████████▊ | 57432/65536 [9:54:07<1:23:29,  1.62it/s]
 88%|████████▊ | 57433/65536 [9:54:08<1:22:46,  1.63it/s]
 88%|████████▊ | 57434/65536 [9:54:08<1:21:52,  1.65it/s]
 88%|████████▊ | 57435/65536 [9:54:09<1:24:09,  1.60it/s]
 88%|████████▊ | 57436/65536 [9:54:10<1:23:25,  1.62it/s]
 88%|████████▊ | 57437/65536 [9:54:10<1:26:33,  1.56it/s]
 88%|████████▊ | 57438/65536 [9:54:11<1:25:42,  1.57it/s]
 88%|████████▊ | 57439/65536 [9:54:11<1:25:01,  1.59it/s]
 88%|████████▊ | 57440/65536 [9:54:12<1:22:54,  1.63it/s]
                                                         
{'loss': 1.5241, 'learning_rate': 2.1152197869474712e-07, 'epoch': 3545.68}
+
 88%|████████▊ | 57440/65536 [9:54:12<1:22:54,  1.63it/s]
 88%|████████▊ | 57441/65536 [9:54:13<1:20:43,  1.67it/s]
 88%|████████▊ | 57442/65536 [9:54:13<1:21:26,  1.66it/s]
 88%|████████▊ | 57443/65536 [9:54:14<1:20:21,  1.68it/s]
 88%|████████▊ | 57444/65536 [9:54:14<1:21:20,  1.66it/s]
 88%|████████▊ | 57445/65536 [9:54:15<1:23:05,  1.62it/s]
 88%|████████▊ | 57446/65536 [9:54:16<1:28:05,  1.53it/s]
 88%|████████▊ | 57447/65536 [9:54:16<1:26:19,  1.56it/s]
 88%|████████▊ | 57448/65536 [9:54:17<1:26:15,  1.56it/s]
 88%|████████▊ | 57449/65536 [9:54:18<1:24:31,  1.59it/s]
 88%|████████▊ | 57450/65536 [9:54:18<1:22:51,  1.63it/s]
 88%|████████▊ | 57451/65536 [9:54:19<1:22:53,  1.63it/s]
 88%|████████▊ | 57452/65536 [9:54:19<1:23:16,  1.62it/s]
 88%|████████▊ | 57453/65536 [9:54:20<1:21:43,  1.65it/s]
 88%|████████▊ | 57454/65536 [9:54:21<1:20:43,  1.67it/s]
 88%|████████▊ | 57455/65536 [9:54:21<1:21:50,  1.65it/s]
 88%|████████▊ | 57456/65536 [9:54:22<1:21:51,  1.65it/s]
 88%|████████▊ | 57457/65536 [9:54:22<1:23:11,  1.62it/s]
 88%|████████▊ | 57458/65536 [9:54:23<1:22:14,  1.64it/s]
 88%|████████▊ | 57459/65536 [9:54:24<1:20:37,  1.67it/s]
 88%|████████▊ | 57460/65536 [9:54:24<1:21:11,  1.66it/s]
                                                         
{'loss': 1.593, 'learning_rate': 2.1124647973552098e-07, 'epoch': 3546.91}
+
 88%|████████▊ | 57460/65536 [9:54:24<1:21:11,  1.66it/s]
 88%|████████▊ | 57461/65536 [9:54:25<1:24:09,  1.60it/s]
 88%|████████▊ | 57462/65536 [9:54:26<1:27:06,  1.54it/s]
 88%|████████▊ | 57463/65536 [9:54:26<1:26:00,  1.56it/s]
 88%|████████▊ | 57464/65536 [9:54:27<1:23:52,  1.60it/s]
 88%|████████▊ | 57465/65536 [9:54:28<1:25:07,  1.58it/s]
 88%|████████▊ | 57466/65536 [9:54:28<1:23:29,  1.61it/s]
 88%|████████▊ | 57467/65536 [9:54:29<1:24:30,  1.59it/s]
 88%|████████▊ | 57468/65536 [9:54:29<1:23:12,  1.62it/s]
 88%|████████▊ | 57469/65536 [9:54:30<1:22:55,  1.62it/s]
 88%|████████▊ | 57470/65536 [9:54:31<1:24:37,  1.59it/s]
 88%|████████▊ | 57471/65536 [9:54:31<1:23:27,  1.61it/s]
 88%|████████▊ | 57472/65536 [9:54:32<1:26:05,  1.56it/s]
 88%|████████▊ | 57473/65536 [9:54:32<1:24:13,  1.60it/s]
 88%|████████▊ | 57474/65536 [9:54:33<1:23:40,  1.61it/s]
 88%|████████▊ | 57475/65536 [9:54:34<1:22:43,  1.62it/s]
 88%|████████▊ | 57476/65536 [9:54:34<1:21:26,  1.65it/s]
 88%|████████▊ | 57477/65536 [9:54:35<1:21:45,  1.64it/s]
 88%|████████▊ | 57478/65536 [9:54:36<1:26:00,  1.56it/s]
 88%|████████▊ | 57479/65536 [9:54:36<1:23:50,  1.60it/s]
 88%|████████▊ | 57480/65536 [9:54:37<1:24:33,  1.59it/s]
                                                         
{'loss': 1.5178, 'learning_rate': 2.1097098077629487e-07, 'epoch': 3548.15}
+
 88%|████████▊ | 57480/65536 [9:54:37<1:24:33,  1.59it/s]
 88%|████████▊ | 57481/65536 [9:54:37<1:23:24,  1.61it/s]
 88%|████████▊ | 57482/65536 [9:54:38<1:24:12,  1.59it/s]
 88%|████████▊ | 57483/65536 [9:54:39<1:22:42,  1.62it/s]
 88%|████████▊ | 57484/65536 [9:54:39<1:22:05,  1.63it/s]
 88%|████████▊ | 57485/65536 [9:54:40<1:21:18,  1.65it/s]
 88%|████████▊ | 57486/65536 [9:54:41<1:23:59,  1.60it/s]
 88%|████████▊ | 57487/65536 [9:54:41<1:23:45,  1.60it/s]
 88%|████████▊ | 57488/65536 [9:54:42<1:24:37,  1.59it/s]
 88%|████████▊ | 57489/65536 [9:54:42<1:24:05,  1.60it/s]
 88%|████████▊ | 57490/65536 [9:54:43<1:22:44,  1.62it/s]
 88%|████████▊ | 57491/65536 [9:54:44<1:23:38,  1.60it/s]
 88%|████████▊ | 57492/65536 [9:54:44<1:22:38,  1.62it/s]
 88%|████████▊ | 57493/65536 [9:54:45<1:23:11,  1.61it/s]
 88%|████████▊ | 57494/65536 [9:54:46<1:28:18,  1.52it/s]
 88%|████████▊ | 57495/65536 [9:54:46<1:27:28,  1.53it/s]
 88%|████████▊ | 57496/65536 [9:54:47<1:26:41,  1.55it/s]
 88%|████████▊ | 57497/65536 [9:54:48<1:24:52,  1.58it/s]
 88%|████████▊ | 57498/65536 [9:54:48<1:24:11,  1.59it/s]
 88%|████████▊ | 57499/65536 [9:54:49<1:23:29,  1.60it/s]
 88%|████████▊ | 57500/65536 [9:54:49<1:24:40,  1.58it/s]
                                                         
{'loss': 1.528, 'learning_rate': 2.106954818170687e-07, 'epoch': 3549.38}
+
 88%|████████▊ | 57500/65536 [9:54:49<1:24:40,  1.58it/s]
 88%|████████▊ | 57501/65536 [9:54:50<1:22:25,  1.62it/s]
 88%|████████▊ | 57502/65536 [9:54:51<1:23:45,  1.60it/s]
 88%|████████▊ | 57503/65536 [9:54:51<1:24:43,  1.58it/s]
 88%|████████▊ | 57504/65536 [9:54:52<1:24:50,  1.58it/s]
 88%|████████▊ | 57505/65536 [9:54:53<1:23:38,  1.60it/s]
 88%|████████▊ | 57506/65536 [9:54:53<1:23:48,  1.60it/s]
 88%|████████▊ | 57507/65536 [9:54:54<1:22:01,  1.63it/s]
 88%|████████▊ | 57508/65536 [9:54:54<1:21:43,  1.64it/s]
 88%|████████▊ | 57509/65536 [9:54:55<1:22:24,  1.62it/s]
 88%|████████▊ | 57510/65536 [9:54:56<1:22:20,  1.62it/s]
 88%|████████▊ | 57511/65536 [9:54:56<1:24:12,  1.59it/s]
 88%|████████▊ | 57512/65536 [9:54:57<1:23:19,  1.60it/s]
 88%|████████▊ | 57513/65536 [9:54:57<1:23:07,  1.61it/s]
 88%|████████▊ | 57514/65536 [9:54:58<1:24:17,  1.59it/s]
 88%|████████▊ | 57515/65536 [9:54:59<1:23:02,  1.61it/s]
 88%|████████▊ | 57516/65536 [9:54:59<1:23:55,  1.59it/s]
 88%|████████▊ | 57517/65536 [9:55:00<1:23:07,  1.61it/s]
 88%|████████▊ | 57518/65536 [9:55:01<1:21:44,  1.63it/s]
 88%|████████▊ | 57519/65536 [9:55:01<1:22:28,  1.62it/s]
 88%|████████▊ | 57520/65536 [9:55:02<1:24:04,  1.59it/s]
                                                         
{'loss': 1.5427, 'learning_rate': 2.1041998285784256e-07, 'epoch': 3550.62}
+
 88%|████████▊ | 57520/65536 [9:55:02<1:24:04,  1.59it/s]
 88%|████████▊ | 57521/65536 [9:55:02<1:25:03,  1.57it/s]
 88%|████████▊ | 57522/65536 [9:55:03<1:24:33,  1.58it/s]
 88%|████████▊ | 57523/65536 [9:55:04<1:22:00,  1.63it/s]
 88%|████████▊ | 57524/65536 [9:55:04<1:21:23,  1.64it/s]
 88%|████████▊ | 57525/65536 [9:55:05<1:20:50,  1.65it/s]
 88%|████████▊ | 57526/65536 [9:55:06<1:21:22,  1.64it/s]
 88%|████████▊ | 57527/65536 [9:55:06<1:26:32,  1.54it/s]
 88%|████████▊ | 57528/65536 [9:55:07<1:24:21,  1.58it/s]
 88%|████████▊ | 57529/65536 [9:55:07<1:22:48,  1.61it/s]
 88%|████████▊ | 57530/65536 [9:55:08<1:22:32,  1.62it/s]
 88%|████████▊ | 57531/65536 [9:55:09<1:22:11,  1.62it/s]
 88%|████████▊ | 57532/65536 [9:55:09<1:20:47,  1.65it/s]
 88%|████████▊ | 57533/65536 [9:55:10<1:20:51,  1.65it/s]
 88%|████████▊ | 57534/65536 [9:55:10<1:21:22,  1.64it/s]
 88%|████████▊ | 57535/65536 [9:55:11<1:22:21,  1.62it/s]
 88%|████████▊ | 57536/65536 [9:55:12<1:20:59,  1.65it/s]
 88%|████████▊ | 57537/65536 [9:55:12<1:23:50,  1.59it/s]
 88%|████████▊ | 57538/65536 [9:55:13<1:22:09,  1.62it/s]
 88%|████████▊ | 57539/65536 [9:55:14<1:23:34,  1.59it/s]
 88%|████████▊ | 57540/65536 [9:55:14<1:21:43,  1.63it/s]
                                                         
{'loss': 1.5824, 'learning_rate': 2.1014448389861642e-07, 'epoch': 3551.85}
+
 88%|████████▊ | 57540/65536 [9:55:14<1:21:43,  1.63it/s]
 88%|████████▊ | 57541/65536 [9:55:15<1:22:13,  1.62it/s]
 88%|████████▊ | 57542/65536 [9:55:15<1:22:30,  1.61it/s]
 88%|████████▊ | 57543/65536 [9:55:16<1:26:09,  1.55it/s]
 88%|████████▊ | 57544/65536 [9:55:17<1:24:45,  1.57it/s]
 88%|████████▊ | 57545/65536 [9:55:17<1:22:28,  1.61it/s]
 88%|████████▊ | 57546/65536 [9:55:18<1:22:33,  1.61it/s]
 88%|████████▊ | 57547/65536 [9:55:19<1:22:10,  1.62it/s]
 88%|████████▊ | 57548/65536 [9:55:19<1:22:20,  1.62it/s]
 88%|████████▊ | 57549/65536 [9:55:20<1:22:13,  1.62it/s]
 88%|████████▊ | 57550/65536 [9:55:20<1:23:07,  1.60it/s]
 88%|████████▊ | 57551/65536 [9:55:21<1:22:05,  1.62it/s]
 88%|████████▊ | 57552/65536 [9:55:22<1:22:02,  1.62it/s]
 88%|████████▊ | 57553/65536 [9:55:22<1:21:32,  1.63it/s]
 88%|████████▊ | 57554/65536 [9:55:23<1:21:55,  1.62it/s]
 88%|████████▊ | 57555/65536 [9:55:24<1:23:57,  1.58it/s]
 88%|████████▊ | 57556/65536 [9:55:24<1:21:45,  1.63it/s]
 88%|████████▊ | 57557/65536 [9:55:25<1:21:42,  1.63it/s]
 88%|████████▊ | 57558/65536 [9:55:25<1:23:24,  1.59it/s]
 88%|████████▊ | 57559/65536 [9:55:26<1:26:11,  1.54it/s]
 88%|████████▊ | 57560/65536 [9:55:27<1:24:04,  1.58it/s]
                                                         
{'loss': 1.5518, 'learning_rate': 2.0986898493939017e-07, 'epoch': 3553.09}
+
 88%|████████▊ | 57560/65536 [9:55:27<1:24:04,  1.58it/s]
 88%|████████▊ | 57561/65536 [9:55:27<1:23:40,  1.59it/s]
 88%|████████▊ | 57562/65536 [9:55:28<1:22:17,  1.61it/s]
 88%|████████▊ | 57563/65536 [9:55:29<1:21:14,  1.64it/s]
 88%|████████▊ | 57564/65536 [9:55:29<1:23:48,  1.59it/s]
 88%|████████▊ | 57565/65536 [9:55:30<1:24:26,  1.57it/s]
 88%|████████▊ | 57566/65536 [9:55:30<1:24:45,  1.57it/s]
 88%|████████▊ | 57567/65536 [9:55:31<1:25:23,  1.56it/s]
 88%|████████▊ | 57568/65536 [9:55:32<1:24:38,  1.57it/s]
 88%|████████▊ | 57569/65536 [9:55:32<1:23:10,  1.60it/s]
 88%|████████▊ | 57570/65536 [9:55:33<1:21:38,  1.63it/s]
 88%|████████▊ | 57571/65536 [9:55:34<1:21:46,  1.62it/s]
 88%|████████▊ | 57572/65536 [9:55:34<1:21:20,  1.63it/s]
 88%|████████▊ | 57573/65536 [9:55:35<1:20:56,  1.64it/s]
 88%|████████▊ | 57574/65536 [9:55:35<1:22:52,  1.60it/s]
 88%|████████▊ | 57575/65536 [9:55:36<1:25:10,  1.56it/s]
 88%|████████▊ | 57576/65536 [9:55:37<1:24:08,  1.58it/s]
 88%|████████▊ | 57577/65536 [9:55:37<1:21:04,  1.64it/s]
 88%|████████▊ | 57578/65536 [9:55:38<1:21:35,  1.63it/s]
 88%|████████▊ | 57579/65536 [9:55:39<1:20:55,  1.64it/s]
 88%|████████▊ | 57580/65536 [9:55:39<1:24:29,  1.57it/s]
                                                         
{'loss': 1.5727, 'learning_rate': 2.0959348598016403e-07, 'epoch': 3554.32}
+
 88%|████████▊ | 57580/65536 [9:55:39<1:24:29,  1.57it/s]
 88%|████████▊ | 57581/65536 [9:55:40<1:22:51,  1.60it/s]
 88%|████████▊ | 57582/65536 [9:55:40<1:21:27,  1.63it/s]
 88%|████████▊ | 57583/65536 [9:55:41<1:22:10,  1.61it/s]
 88%|████████▊ | 57584/65536 [9:55:42<1:21:11,  1.63it/s]
 88%|████████▊ | 57585/65536 [9:55:42<1:22:56,  1.60it/s]
 88%|████████▊ | 57586/65536 [9:55:43<1:22:01,  1.62it/s]
 88%|████████▊ | 57587/65536 [9:55:43<1:21:36,  1.62it/s]
 88%|████████▊ | 57588/65536 [9:55:44<1:22:20,  1.61it/s]
 88%|████████▊ | 57589/65536 [9:55:45<1:21:32,  1.62it/s]
 88%|████████▊ | 57590/65536 [9:55:45<1:20:45,  1.64it/s]
 88%|████████▊ | 57591/65536 [9:55:46<1:20:24,  1.65it/s]
 88%|████████▊ | 57592/65536 [9:55:47<1:21:56,  1.62it/s]
 88%|████████▊ | 57593/65536 [9:55:47<1:23:15,  1.59it/s]
 88%|████████▊ | 57594/65536 [9:55:48<1:23:42,  1.58it/s]
 88%|████████▊ | 57595/65536 [9:55:48<1:22:59,  1.59it/s]
 88%|████████▊ | 57596/65536 [9:55:49<1:23:14,  1.59it/s]
 88%|████████▊ | 57597/65536 [9:55:50<1:21:53,  1.62it/s]
 88%|████████▊ | 57598/65536 [9:55:50<1:24:13,  1.57it/s]
 88%|████████▊ | 57599/65536 [9:55:51<1:21:33,  1.62it/s]
 88%|████████▊ | 57600/65536 [9:55:52<1:21:12,  1.63it/s]
                                                         
{'loss': 1.5564, 'learning_rate': 2.093179870209379e-07, 'epoch': 3555.56}
+
 88%|████████▊ | 57600/65536 [9:55:52<1:21:12,  1.63it/s]
 88%|████████▊ | 57601/65536 [9:55:52<1:22:11,  1.61it/s]
 88%|████████▊ | 57602/65536 [9:55:53<1:21:43,  1.62it/s]
 88%|████████▊ | 57603/65536 [9:55:53<1:21:40,  1.62it/s]
 88%|████████▊ | 57604/65536 [9:55:54<1:22:06,  1.61it/s]
 88%|████████▊ | 57605/65536 [9:55:55<1:22:55,  1.59it/s]
 88%|████████▊ | 57606/65536 [9:55:55<1:21:28,  1.62it/s]
 88%|████████▊ | 57607/65536 [9:55:56<1:20:38,  1.64it/s]
 88%|████████▊ | 57608/65536 [9:55:57<1:25:29,  1.55it/s]
 88%|████████▊ | 57609/65536 [9:55:57<1:24:54,  1.56it/s]
 88%|████████▊ | 57610/65536 [9:55:58<1:24:50,  1.56it/s]
 88%|████████▊ | 57611/65536 [9:55:59<1:24:12,  1.57it/s]
 88%|████████▊ | 57612/65536 [9:55:59<1:22:22,  1.60it/s]
 88%|████████▊ | 57613/65536 [9:56:00<1:21:52,  1.61it/s]
 88%|████████▊ | 57614/65536 [9:56:00<1:20:09,  1.65it/s]
 88%|████████▊ | 57615/65536 [9:56:01<1:22:18,  1.60it/s]
 88%|████████▊ | 57616/65536 [9:56:02<1:21:07,  1.63it/s]
 88%|████████▊ | 57617/65536 [9:56:02<1:23:03,  1.59it/s]
 88%|████████▊ | 57618/65536 [9:56:03<1:21:56,  1.61it/s]
 88%|████████▊ | 57619/65536 [9:56:03<1:21:25,  1.62it/s]
 88%|████████▊ | 57620/65536 [9:56:04<1:24:25,  1.56it/s]
                                                         
{'loss': 1.5272, 'learning_rate': 2.0904248806171175e-07, 'epoch': 3556.79}
+
 88%|████████▊ | 57620/65536 [9:56:04<1:24:25,  1.56it/s]
 88%|████████▊ | 57621/65536 [9:56:05<1:22:38,  1.60it/s]
 88%|████████▊ | 57622/65536 [9:56:05<1:20:34,  1.64it/s]
 88%|████████▊ | 57623/65536 [9:56:06<1:20:14,  1.64it/s]
 88%|████████▊ | 57624/65536 [9:56:07<1:23:25,  1.58it/s]
 88%|████████▊ | 57625/65536 [9:56:07<1:25:03,  1.55it/s]
 88%|████████▊ | 57626/65536 [9:56:08<1:23:14,  1.58it/s]
 88%|████████▊ | 57627/65536 [9:56:08<1:20:20,  1.64it/s]
 88%|████████▊ | 57628/65536 [9:56:09<1:19:18,  1.66it/s]
 88%|████████▊ | 57629/65536 [9:56:10<1:20:39,  1.63it/s]
 88%|████████▊ | 57630/65536 [9:56:10<1:20:51,  1.63it/s]
 88%|████████▊ | 57631/65536 [9:56:11<1:19:25,  1.66it/s]
 88%|████████▊ | 57632/65536 [9:56:11<1:20:04,  1.65it/s]
 88%|████████▊ | 57633/65536 [9:56:12<1:19:44,  1.65it/s]
 88%|████████▊ | 57634/65536 [9:56:13<1:18:51,  1.67it/s]
 88%|████████▊ | 57635/65536 [9:56:13<1:19:38,  1.65it/s]
 88%|████████▊ | 57636/65536 [9:56:14<1:21:50,  1.61it/s]
 88%|████████▊ | 57637/65536 [9:56:14<1:20:39,  1.63it/s]
 88%|████████▊ | 57638/65536 [9:56:15<1:20:32,  1.63it/s]
 88%|████████▊ | 57639/65536 [9:56:16<1:20:42,  1.63it/s]
 88%|████████▊ | 57640/65536 [9:56:16<1:25:12,  1.54it/s]
                                                         
{'loss': 1.5444, 'learning_rate': 2.087669891024856e-07, 'epoch': 3558.02}
+
 88%|████████▊ | 57640/65536 [9:56:16<1:25:12,  1.54it/s]
 88%|████████▊ | 57641/65536 [9:56:17<1:23:45,  1.57it/s]
 88%|████████▊ | 57642/65536 [9:56:18<1:21:18,  1.62it/s]
 88%|████████▊ | 57643/65536 [9:56:18<1:20:46,  1.63it/s]
 88%|████████▊ | 57644/65536 [9:56:19<1:22:58,  1.59it/s]
 88%|████████▊ | 57645/65536 [9:56:20<1:22:32,  1.59it/s]
 88%|████████▊ | 57646/65536 [9:56:20<1:21:18,  1.62it/s]
 88%|████████▊ | 57647/65536 [9:56:21<1:21:48,  1.61it/s]
 88%|████████▊ | 57648/65536 [9:56:21<1:24:04,  1.56it/s]
 88%|████████▊ | 57649/65536 [9:56:22<1:22:46,  1.59it/s]
 88%|████████▊ | 57650/65536 [9:56:23<1:23:17,  1.58it/s]
 88%|████████▊ | 57651/65536 [9:56:23<1:23:10,  1.58it/s]
 88%|████████▊ | 57652/65536 [9:56:24<1:23:11,  1.58it/s]
 88%|████████▊ | 57653/65536 [9:56:25<1:21:56,  1.60it/s]
 88%|████████▊ | 57654/65536 [9:56:25<1:19:33,  1.65it/s]
 88%|████████▊ | 57655/65536 [9:56:26<1:20:02,  1.64it/s]
 88%|████████▊ | 57656/65536 [9:56:26<1:21:48,  1.61it/s]
 88%|████████▊ | 57657/65536 [9:56:27<1:21:32,  1.61it/s]
 88%|████████▊ | 57658/65536 [9:56:28<1:20:29,  1.63it/s]
 88%|████████▊ | 57659/65536 [9:56:28<1:19:24,  1.65it/s]
 88%|████████▊ | 57660/65536 [9:56:29<1:22:47,  1.59it/s]
                                                         
{'loss': 1.5539, 'learning_rate': 2.084914901432595e-07, 'epoch': 3559.26}
+
 88%|████████▊ | 57660/65536 [9:56:29<1:22:47,  1.59it/s]
 88%|████████▊ | 57661/65536 [9:56:29<1:21:25,  1.61it/s]
 88%|████████▊ | 57662/65536 [9:56:30<1:21:46,  1.60it/s]
 88%|████████▊ | 57663/65536 [9:56:31<1:21:09,  1.62it/s]
 88%|████████▊ | 57664/65536 [9:56:31<1:20:52,  1.62it/s]
 88%|████████▊ | 57665/65536 [9:56:32<1:19:43,  1.65it/s]
 88%|████████▊ | 57666/65536 [9:56:33<1:20:12,  1.64it/s]
 88%|████████▊ | 57667/65536 [9:56:33<1:20:28,  1.63it/s]
 88%|████████▊ | 57668/65536 [9:56:34<1:23:21,  1.57it/s]
 88%|████████▊ | 57669/65536 [9:56:35<1:25:02,  1.54it/s]
 88%|████████▊ | 57670/65536 [9:56:35<1:22:17,  1.59it/s]
 88%|████████▊ | 57671/65536 [9:56:36<1:23:03,  1.58it/s]
 88%|████████▊ | 57672/65536 [9:56:36<1:20:59,  1.62it/s]
 88%|████████▊ | 57673/65536 [9:56:37<1:23:21,  1.57it/s]
 88%|████████▊ | 57674/65536 [9:56:38<1:22:01,  1.60it/s]
 88%|████████▊ | 57675/65536 [9:56:38<1:21:30,  1.61it/s]
 88%|████████▊ | 57676/65536 [9:56:39<1:22:32,  1.59it/s]
 88%|████████▊ | 57677/65536 [9:56:39<1:22:19,  1.59it/s]
 88%|████████▊ | 57678/65536 [9:56:40<1:22:59,  1.58it/s]
 88%|████████▊ | 57679/65536 [9:56:41<1:22:35,  1.59it/s]
 88%|████████▊ | 57680/65536 [9:56:41<1:21:12,  1.61it/s]
                                                         
{'loss': 1.5628, 'learning_rate': 2.0821599118403324e-07, 'epoch': 3560.49}
+
 88%|████████▊ | 57680/65536 [9:56:41<1:21:12,  1.61it/s]
 88%|████████▊ | 57681/65536 [9:56:42<1:22:01,  1.60it/s]
 88%|████████▊ | 57682/65536 [9:56:43<1:22:01,  1.60it/s]
 88%|████████▊ | 57683/65536 [9:56:43<1:19:52,  1.64it/s]
 88%|████████▊ | 57684/65536 [9:56:44<1:20:15,  1.63it/s]
 88%|████████▊ | 57685/65536 [9:56:44<1:20:49,  1.62it/s]
 88%|████████▊ | 57686/65536 [9:56:45<1:18:53,  1.66it/s]
 88%|████████▊ | 57687/65536 [9:56:46<1:20:51,  1.62it/s]
 88%|████████▊ | 57688/65536 [9:56:46<1:21:36,  1.60it/s]
 88%|████████▊ | 57689/65536 [9:56:47<1:25:21,  1.53it/s]
 88%|████████▊ | 57690/65536 [9:56:48<1:22:46,  1.58it/s]
 88%|████████▊ | 57691/65536 [9:56:48<1:22:32,  1.58it/s]
 88%|████████▊ | 57692/65536 [9:56:49<1:21:57,  1.59it/s]
 88%|████████▊ | 57693/65536 [9:56:49<1:20:52,  1.62it/s]
 88%|████████▊ | 57694/65536 [9:56:50<1:19:17,  1.65it/s]
 88%|████████▊ | 57695/65536 [9:56:51<1:18:59,  1.65it/s]
 88%|████████▊ | 57696/65536 [9:56:51<1:20:12,  1.63it/s]
 88%|████████▊ | 57697/65536 [9:56:52<1:21:37,  1.60it/s]
 88%|████████▊ | 57698/65536 [9:56:53<1:20:58,  1.61it/s]
 88%|████████▊ | 57699/65536 [9:56:53<1:21:42,  1.60it/s]
 88%|████████▊ | 57700/65536 [9:56:54<1:21:57,  1.59it/s]
                                                         
{'loss': 1.543, 'learning_rate': 2.079404922248071e-07, 'epoch': 3561.73}
+
 88%|████████▊ | 57700/65536 [9:56:54<1:21:57,  1.59it/s]
 88%|████████▊ | 57701/65536 [9:56:54<1:20:25,  1.62it/s]
 88%|████████▊ | 57702/65536 [9:56:55<1:21:41,  1.60it/s]
 88%|████████▊ | 57703/65536 [9:56:56<1:23:15,  1.57it/s]
 88%|████████▊ | 57704/65536 [9:56:56<1:19:43,  1.64it/s]
 88%|████████▊ | 57705/65536 [9:56:57<1:23:52,  1.56it/s]
 88%|████████▊ | 57706/65536 [9:56:58<1:22:44,  1.58it/s]
 88%|████████▊ | 57707/65536 [9:56:58<1:20:22,  1.62it/s]
 88%|████████▊ | 57708/65536 [9:56:59<1:19:57,  1.63it/s]
 88%|████████▊ | 57709/65536 [9:56:59<1:20:54,  1.61it/s]
 88%|████████▊ | 57710/65536 [9:57:00<1:21:10,  1.61it/s]
 88%|████████▊ | 57711/65536 [9:57:01<1:20:08,  1.63it/s]
 88%|████████▊ | 57712/65536 [9:57:01<1:18:37,  1.66it/s]
 88%|████████▊ | 57713/65536 [9:57:02<1:20:24,  1.62it/s]
 88%|████████▊ | 57714/65536 [9:57:02<1:20:00,  1.63it/s]
 88%|████████▊ | 57715/65536 [9:57:03<1:18:25,  1.66it/s]
 88%|████████▊ | 57716/65536 [9:57:04<1:18:34,  1.66it/s]
 88%|████████▊ | 57717/65536 [9:57:04<1:18:33,  1.66it/s]
 88%|████████▊ | 57718/65536 [9:57:05<1:24:04,  1.55it/s]
 88%|████████▊ | 57719/65536 [9:57:06<1:26:55,  1.50it/s]
 88%|████████▊ | 57720/65536 [9:57:06<1:24:22,  1.54it/s]
                                                         
{'loss': 1.5733, 'learning_rate': 2.0766499326558096e-07, 'epoch': 3562.96}
+
 88%|████████▊ | 57720/65536 [9:57:06<1:24:22,  1.54it/s]
 88%|████████▊ | 57721/65536 [9:57:07<1:26:56,  1.50it/s]
 88%|████████▊ | 57722/65536 [9:57:08<1:27:28,  1.49it/s]
 88%|████████▊ | 57723/65536 [9:57:08<1:26:01,  1.51it/s]
 88%|████████▊ | 57724/65536 [9:57:09<1:25:10,  1.53it/s]
 88%|████████▊ | 57725/65536 [9:57:10<1:22:49,  1.57it/s]
 88%|████████▊ | 57726/65536 [9:57:10<1:21:16,  1.60it/s]
 88%|████████▊ | 57727/65536 [9:57:11<1:19:51,  1.63it/s]
 88%|████████▊ | 57728/65536 [9:57:11<1:21:48,  1.59it/s]
 88%|████████▊ | 57729/65536 [9:57:12<1:21:09,  1.60it/s]
 88%|████████▊ | 57730/65536 [9:57:13<1:22:20,  1.58it/s]
 88%|████████▊ | 57731/65536 [9:57:13<1:20:41,  1.61it/s]
 88%|████████▊ | 57732/65536 [9:57:14<1:20:45,  1.61it/s]
 88%|████████▊ | 57733/65536 [9:57:14<1:18:00,  1.67it/s]
 88%|████████▊ | 57734/65536 [9:57:15<1:18:46,  1.65it/s]
 88%|████████▊ | 57735/65536 [9:57:16<1:18:54,  1.65it/s]
 88%|████████▊ | 57736/65536 [9:57:16<1:17:33,  1.68it/s]
 88%|████████▊ | 57737/65536 [9:57:17<1:20:06,  1.62it/s]
 88%|████████▊ | 57738/65536 [9:57:18<1:20:44,  1.61it/s]
 88%|████████▊ | 57739/65536 [9:57:18<1:20:52,  1.61it/s]
 88%|████████▊ | 57740/65536 [9:57:19<1:20:42,  1.61it/s]
                                                         
{'loss': 1.5765, 'learning_rate': 2.0738949430635485e-07, 'epoch': 3564.2}
+
 88%|████████▊ | 57740/65536 [9:57:19<1:20:42,  1.61it/s]
 88%|████████▊ | 57741/65536 [9:57:19<1:20:44,  1.61it/s]
 88%|████████▊ | 57742/65536 [9:57:20<1:22:38,  1.57it/s]
 88%|████████▊ | 57743/65536 [9:57:21<1:19:34,  1.63it/s]
 88%|████████▊ | 57744/65536 [9:57:21<1:20:29,  1.61it/s]
 88%|████████▊ | 57745/65536 [9:57:22<1:20:00,  1.62it/s]
 88%|████████▊ | 57746/65536 [9:57:23<1:20:48,  1.61it/s]
 88%|████████▊ | 57747/65536 [9:57:23<1:19:26,  1.63it/s]
 88%|████████▊ | 57748/65536 [9:57:24<1:18:21,  1.66it/s]
 88%|████████▊ | 57749/65536 [9:57:24<1:20:31,  1.61it/s]
 88%|████████▊ | 57750/65536 [9:57:25<1:19:47,  1.63it/s]
 88%|████████▊ | 57751/65536 [9:57:26<1:19:54,  1.62it/s]
 88%|████████▊ | 57752/65536 [9:57:26<1:19:10,  1.64it/s]
 88%|████████▊ | 57753/65536 [9:57:27<1:18:11,  1.66it/s]
 88%|████████▊ | 57754/65536 [9:57:27<1:20:33,  1.61it/s]
 88%|████████▊ | 57755/65536 [9:57:28<1:21:13,  1.60it/s]
 88%|████████▊ | 57756/65536 [9:57:29<1:19:48,  1.62it/s]
 88%|████████▊ | 57757/65536 [9:57:29<1:18:19,  1.66it/s]
 88%|████████▊ | 57758/65536 [9:57:30<1:17:31,  1.67it/s]
 88%|████████▊ | 57759/65536 [9:57:30<1:17:27,  1.67it/s]
 88%|████████▊ | 57760/65536 [9:57:31<1:17:49,  1.67it/s]
                                                         
{'loss': 1.5361, 'learning_rate': 2.071139953471287e-07, 'epoch': 3565.43}
+
 88%|████████▊ | 57760/65536 [9:57:31<1:17:49,  1.67it/s]
 88%|████████▊ | 57761/65536 [9:57:32<1:20:38,  1.61it/s]
 88%|████████▊ | 57762/65536 [9:57:32<1:22:20,  1.57it/s]
 88%|████████▊ | 57763/65536 [9:57:33<1:22:09,  1.58it/s]
 88%|████████▊ | 57764/65536 [9:57:34<1:22:01,  1.58it/s]
 88%|████████▊ | 57765/65536 [9:57:34<1:20:19,  1.61it/s]
 88%|████████▊ | 57766/65536 [9:57:35<1:20:05,  1.62it/s]
 88%|████████▊ | 57767/65536 [9:57:35<1:19:37,  1.63it/s]
 88%|████████▊ | 57768/65536 [9:57:36<1:22:13,  1.57it/s]
 88%|████████▊ | 57769/65536 [9:57:37<1:20:54,  1.60it/s]
 88%|████████▊ | 57770/65536 [9:57:37<1:21:48,  1.58it/s]
 88%|████████▊ | 57771/65536 [9:57:38<1:21:46,  1.58it/s]
 88%|████████▊ | 57772/65536 [9:57:39<1:19:49,  1.62it/s]
 88%|████████▊ | 57773/65536 [9:57:39<1:19:41,  1.62it/s]
 88%|████████▊ | 57774/65536 [9:57:40<1:19:15,  1.63it/s]
 88%|████████▊ | 57775/65536 [9:57:40<1:19:49,  1.62it/s]
 88%|████████▊ | 57776/65536 [9:57:41<1:20:52,  1.60it/s]
 88%|████████▊ | 57777/65536 [9:57:42<1:20:21,  1.61it/s]
 88%|████████▊ | 57778/65536 [9:57:42<1:19:38,  1.62it/s]
 88%|████████▊ | 57779/65536 [9:57:43<1:21:04,  1.59it/s]
 88%|████████▊ | 57780/65536 [9:57:44<1:20:24,  1.61it/s]
                                                         
{'loss': 1.5325, 'learning_rate': 2.0683849638790256e-07, 'epoch': 3566.67}
+
 88%|████████▊ | 57780/65536 [9:57:44<1:20:24,  1.61it/s]
 88%|████████▊ | 57781/65536 [9:57:44<1:21:44,  1.58it/s]
 88%|████████▊ | 57782/65536 [9:57:45<1:21:35,  1.58it/s]
 88%|████████▊ | 57783/65536 [9:57:46<1:23:52,  1.54it/s]
 88%|████████▊ | 57784/65536 [9:57:46<1:22:35,  1.56it/s]
 88%|████████▊ | 57785/65536 [9:57:47<1:20:46,  1.60it/s]
 88%|████████▊ | 57786/65536 [9:57:47<1:22:49,  1.56it/s]
 88%|████████▊ | 57787/65536 [9:57:48<1:22:15,  1.57it/s]
 88%|████████▊ | 57788/65536 [9:57:49<1:20:45,  1.60it/s]
 88%|████████▊ | 57789/65536 [9:57:49<1:22:54,  1.56it/s]
 88%|████████▊ | 57790/65536 [9:57:50<1:20:48,  1.60it/s]
 88%|████████▊ | 57791/65536 [9:57:51<1:20:50,  1.60it/s]
 88%|████████▊ | 57792/65536 [9:57:51<1:19:14,  1.63it/s]
 88%|████████▊ | 57793/65536 [9:57:52<1:20:20,  1.61it/s]
 88%|████████▊ | 57794/65536 [9:57:52<1:20:04,  1.61it/s]
 88%|████████▊ | 57795/65536 [9:57:53<1:18:56,  1.63it/s]
 88%|████████▊ | 57796/65536 [9:57:54<1:19:02,  1.63it/s]
 88%|████████▊ | 57797/65536 [9:57:54<1:18:30,  1.64it/s]
 88%|████████▊ | 57798/65536 [9:57:55<1:18:01,  1.65it/s]
 88%|████████▊ | 57799/65536 [9:57:55<1:17:35,  1.66it/s]
 88%|████████▊ | 57800/65536 [9:57:56<1:20:50,  1.59it/s]
                                                         
{'loss': 1.567, 'learning_rate': 2.0656299742867642e-07, 'epoch': 3567.9}
+
 88%|████████▊ | 57800/65536 [9:57:56<1:20:50,  1.59it/s]
 88%|████████▊ | 57801/65536 [9:57:57<1:19:10,  1.63it/s]
 88%|████████▊ | 57802/65536 [9:57:57<1:24:29,  1.53it/s]
 88%|████████▊ | 57803/65536 [9:57:58<1:21:07,  1.59it/s]
 88%|████████▊ | 57804/65536 [9:57:59<1:19:27,  1.62it/s]
 88%|████████▊ | 57805/65536 [9:57:59<1:20:56,  1.59it/s]
 88%|████████▊ | 57806/65536 [9:58:00<1:19:17,  1.62it/s]
 88%|████████▊ | 57807/65536 [9:58:00<1:19:59,  1.61it/s]
 88%|████████▊ | 57808/65536 [9:58:01<1:20:34,  1.60it/s]
 88%|████████▊ | 57809/65536 [9:58:02<1:18:43,  1.64it/s]
 88%|████████▊ | 57810/65536 [9:58:02<1:19:40,  1.62it/s]
 88%|████████▊ | 57811/65536 [9:58:03<1:20:56,  1.59it/s]
 88%|████████▊ | 57812/65536 [9:58:04<1:19:41,  1.62it/s]
 88%|████████▊ | 57813/65536 [9:58:04<1:18:43,  1.63it/s]
 88%|████████▊ | 57814/65536 [9:58:05<1:18:18,  1.64it/s]
 88%|████████▊ | 57815/65536 [9:58:05<1:19:28,  1.62it/s]
 88%|████████▊ | 57816/65536 [9:58:06<1:17:50,  1.65it/s]
 88%|████████▊ | 57817/65536 [9:58:07<1:17:47,  1.65it/s]
 88%|████████▊ | 57818/65536 [9:58:07<1:23:17,  1.54it/s]
 88%|████████▊ | 57819/65536 [9:58:08<1:20:40,  1.59it/s]
 88%|████████▊ | 57820/65536 [9:58:08<1:21:22,  1.58it/s]
                                                         
{'loss': 1.5774, 'learning_rate': 2.0628749846945018e-07, 'epoch': 3569.14}
+
 88%|████████▊ | 57820/65536 [9:58:08<1:21:22,  1.58it/s]
 88%|████████▊ | 57821/65536 [9:58:09<1:19:32,  1.62it/s]
 88%|████████▊ | 57822/65536 [9:58:10<1:18:15,  1.64it/s]
 88%|████████▊ | 57823/65536 [9:58:10<1:20:05,  1.60it/s]
 88%|████████▊ | 57824/65536 [9:58:11<1:20:20,  1.60it/s]
 88%|████████▊ | 57825/65536 [9:58:12<1:20:27,  1.60it/s]
 88%|████████▊ | 57826/65536 [9:58:12<1:21:18,  1.58it/s]
 88%|████████▊ | 57827/65536 [9:58:13<1:22:01,  1.57it/s]
 88%|████████▊ | 57828/65536 [9:58:14<1:21:48,  1.57it/s]
 88%|████████▊ | 57829/65536 [9:58:14<1:21:55,  1.57it/s]
 88%|████████▊ | 57830/65536 [9:58:15<1:19:55,  1.61it/s]
 88%|████████▊ | 57831/65536 [9:58:15<1:19:24,  1.62it/s]
 88%|████████▊ | 57832/65536 [9:58:16<1:20:27,  1.60it/s]
 88%|████████▊ | 57833/65536 [9:58:17<1:19:13,  1.62it/s]
 88%|████████▊ | 57834/65536 [9:58:17<1:19:19,  1.62it/s]
 88%|████████▊ | 57835/65536 [9:58:18<1:22:56,  1.55it/s]
 88%|████████▊ | 57836/65536 [9:58:19<1:22:11,  1.56it/s]
 88%|████████▊ | 57837/65536 [9:58:19<1:20:56,  1.59it/s]
 88%|████████▊ | 57838/65536 [9:58:20<1:22:27,  1.56it/s]
 88%|████████▊ | 57839/65536 [9:58:20<1:20:24,  1.60it/s]
 88%|████████▊ | 57840/65536 [9:58:21<1:18:29,  1.63it/s]
                                                         
{'loss': 1.5808, 'learning_rate': 2.06011999510224e-07, 'epoch': 3570.37}
+
 88%|████████▊ | 57840/65536 [9:58:21<1:18:29,  1.63it/s]
 88%|████████▊ | 57841/65536 [9:58:22<1:19:19,  1.62it/s]
 88%|████████▊ | 57842/65536 [9:58:22<1:18:28,  1.63it/s]
 88%|████████▊ | 57843/65536 [9:58:23<1:17:58,  1.64it/s]
 88%|████████▊ | 57844/65536 [9:58:23<1:17:07,  1.66it/s]
 88%|████████▊ | 57845/65536 [9:58:24<1:19:22,  1.61it/s]
 88%|████████▊ | 57846/65536 [9:58:25<1:18:07,  1.64it/s]
 88%|████████▊ | 57847/65536 [9:58:25<1:17:25,  1.66it/s]
 88%|████████▊ | 57848/65536 [9:58:26<1:19:15,  1.62it/s]
 88%|████████▊ | 57849/65536 [9:58:26<1:17:53,  1.64it/s]
 88%|████████▊ | 57850/65536 [9:58:27<1:16:49,  1.67it/s]
 88%|████████▊ | 57851/65536 [9:58:28<1:19:22,  1.61it/s]
 88%|████████▊ | 57852/65536 [9:58:28<1:18:04,  1.64it/s]
 88%|████████▊ | 57853/65536 [9:58:29<1:17:52,  1.64it/s]
 88%|████████▊ | 57854/65536 [9:58:30<1:19:03,  1.62it/s]
 88%|████████▊ | 57855/65536 [9:58:30<1:17:30,  1.65it/s]
 88%|████████▊ | 57856/65536 [9:58:31<1:17:06,  1.66it/s]
 88%|████████▊ | 57857/65536 [9:58:31<1:19:30,  1.61it/s]
 88%|████████▊ | 57858/65536 [9:58:32<1:20:30,  1.59it/s]
 88%|████████▊ | 57859/65536 [9:58:33<1:19:56,  1.60it/s]
 88%|████████▊ | 57860/65536 [9:58:33<1:20:10,  1.60it/s]
                                                         
{'loss': 1.543, 'learning_rate': 2.057365005509979e-07, 'epoch': 3571.6}
+
 88%|████████▊ | 57860/65536 [9:58:33<1:20:10,  1.60it/s]
 88%|████████▊ | 57861/65536 [9:58:34<1:19:27,  1.61it/s]
 88%|████████▊ | 57862/65536 [9:58:35<1:19:12,  1.61it/s]
 88%|████████▊ | 57863/65536 [9:58:35<1:18:49,  1.62it/s]
 88%|████████▊ | 57864/65536 [9:58:36<1:20:39,  1.59it/s]
 88%|████████▊ | 57865/65536 [9:58:36<1:21:02,  1.58it/s]
 88%|████████▊ | 57866/65536 [9:58:37<1:19:18,  1.61it/s]
 88%|████████▊ | 57867/65536 [9:58:38<1:20:49,  1.58it/s]
 88%|████████▊ | 57868/65536 [9:58:38<1:23:01,  1.54it/s]
 88%|████████▊ | 57869/65536 [9:58:39<1:21:08,  1.57it/s]
 88%|████████▊ | 57870/65536 [9:58:40<1:18:39,  1.62it/s]
 88%|████████▊ | 57871/65536 [9:58:40<1:18:15,  1.63it/s]
 88%|████████▊ | 57872/65536 [9:58:41<1:17:34,  1.65it/s]
 88%|████████▊ | 57873/65536 [9:58:41<1:18:13,  1.63it/s]
 88%|████████▊ | 57874/65536 [9:58:42<1:19:04,  1.62it/s]
 88%|████████▊ | 57875/65536 [9:58:43<1:18:57,  1.62it/s]
 88%|████████▊ | 57876/65536 [9:58:43<1:22:34,  1.55it/s]
 88%|████████▊ | 57877/65536 [9:58:44<1:22:27,  1.55it/s]
 88%|████████▊ | 57878/65536 [9:58:45<1:20:25,  1.59it/s]
 88%|████████▊ | 57879/65536 [9:58:45<1:19:21,  1.61it/s]
 88%|████████▊ | 57880/65536 [9:58:46<1:19:27,  1.61it/s]
                                                         
{'loss': 1.5267, 'learning_rate': 2.0546100159177175e-07, 'epoch': 3572.84}
+
 88%|████████▊ | 57880/65536 [9:58:46<1:19:27,  1.61it/s]
 88%|████████▊ | 57881/65536 [9:58:46<1:18:41,  1.62it/s]
 88%|████████▊ | 57882/65536 [9:58:47<1:18:38,  1.62it/s]
 88%|████████▊ | 57883/65536 [9:58:48<1:19:38,  1.60it/s]
 88%|████████▊ | 57884/65536 [9:58:48<1:17:44,  1.64it/s]
 88%|████████▊ | 57885/65536 [9:58:49<1:18:00,  1.63it/s]
 88%|████████▊ | 57886/65536 [9:58:49<1:17:12,  1.65it/s]
 88%|████████▊ | 57887/65536 [9:58:50<1:19:11,  1.61it/s]
 88%|████████▊ | 57888/65536 [9:58:51<1:19:06,  1.61it/s]
 88%|████████▊ | 57889/65536 [9:58:51<1:21:39,  1.56it/s]
 88%|████████▊ | 57890/65536 [9:58:52<1:21:22,  1.57it/s]
 88%|████████▊ | 57891/65536 [9:58:53<1:19:35,  1.60it/s]
 88%|████████▊ | 57892/65536 [9:58:53<1:19:36,  1.60it/s]
 88%|████████▊ | 57893/65536 [9:58:54<1:19:11,  1.61it/s]
 88%|████████▊ | 57894/65536 [9:58:55<1:19:40,  1.60it/s]
 88%|████████▊ | 57895/65536 [9:58:55<1:18:19,  1.63it/s]
 88%|████████▊ | 57896/65536 [9:58:56<1:18:12,  1.63it/s]
 88%|████████▊ | 57897/65536 [9:58:56<1:16:50,  1.66it/s]
 88%|████████▊ | 57898/65536 [9:58:57<1:18:51,  1.61it/s]
 88%|████████▊ | 57899/65536 [9:58:58<1:21:15,  1.57it/s]
 88%|████████▊ | 57900/65536 [9:58:58<1:18:54,  1.61it/s]
                                                         
{'loss': 1.5612, 'learning_rate': 2.051855026325456e-07, 'epoch': 3574.07}
+
 88%|████████▊ | 57900/65536 [9:58:58<1:18:54,  1.61it/s]
 88%|████████▊ | 57901/65536 [9:58:59<1:19:00,  1.61it/s]
 88%|████████▊ | 57902/65536 [9:58:59<1:18:51,  1.61it/s]
 88%|████████▊ | 57903/65536 [9:59:00<1:17:48,  1.63it/s]
 88%|████████▊ | 57904/65536 [9:59:01<1:19:04,  1.61it/s]
 88%|████████▊ | 57905/65536 [9:59:01<1:18:52,  1.61it/s]
 88%|████████▊ | 57906/65536 [9:59:02<1:19:12,  1.61it/s]
 88%|████████▊ | 57907/65536 [9:59:03<1:20:25,  1.58it/s]
 88%|████████▊ | 57908/65536 [9:59:03<1:19:45,  1.59it/s]
 88%|████████▊ | 57909/65536 [9:59:04<1:22:23,  1.54it/s]
 88%|████████▊ | 57910/65536 [9:59:04<1:19:25,  1.60it/s]
 88%|████████▊ | 57911/65536 [9:59:05<1:17:12,  1.65it/s]
 88%|████████▊ | 57912/65536 [9:59:06<1:18:21,  1.62it/s]
 88%|████████▊ | 57913/65536 [9:59:06<1:17:02,  1.65it/s]
 88%|████████▊ | 57914/65536 [9:59:07<1:17:13,  1.64it/s]
 88%|████████▊ | 57915/65536 [9:59:07<1:16:39,  1.66it/s]
 88%|████████▊ | 57916/65536 [9:59:08<1:19:15,  1.60it/s]
 88%|████████▊ | 57917/65536 [9:59:09<1:18:14,  1.62it/s]
 88%|████████▊ | 57918/65536 [9:59:09<1:19:35,  1.60it/s]
 88%|████████▊ | 57919/65536 [9:59:10<1:18:24,  1.62it/s]
 88%|████████▊ | 57920/65536 [9:59:11<1:16:34,  1.66it/s]
                                                         
{'loss': 1.5589, 'learning_rate': 2.0491000367331947e-07, 'epoch': 3575.31}
+
 88%|████████▊ | 57920/65536 [9:59:11<1:16:34,  1.66it/s]
 88%|████████▊ | 57921/65536 [9:59:11<1:16:22,  1.66it/s]
 88%|████████▊ | 57922/65536 [9:59:12<1:16:06,  1.67it/s]
 88%|████████▊ | 57923/65536 [9:59:12<1:19:36,  1.59it/s]
 88%|████████▊ | 57924/65536 [9:59:13<1:17:49,  1.63it/s]
 88%|████████▊ | 57925/65536 [9:59:14<1:19:22,  1.60it/s]
 88%|████████▊ | 57926/65536 [9:59:14<1:18:17,  1.62it/s]
 88%|████████▊ | 57927/65536 [9:59:15<1:20:07,  1.58it/s]
 88%|████████▊ | 57928/65536 [9:59:16<1:20:21,  1.58it/s]
 88%|████████▊ | 57929/65536 [9:59:16<1:20:24,  1.58it/s]
 88%|████████▊ | 57930/65536 [9:59:17<1:18:57,  1.61it/s]
 88%|████████▊ | 57931/65536 [9:59:17<1:16:27,  1.66it/s]
 88%|████████▊ | 57932/65536 [9:59:18<1:18:54,  1.61it/s]
 88%|████████▊ | 57933/65536 [9:59:19<1:18:52,  1.61it/s]
 88%|████████▊ | 57934/65536 [9:59:19<1:17:44,  1.63it/s]
 88%|████████▊ | 57935/65536 [9:59:20<1:20:24,  1.58it/s]
 88%|████████▊ | 57936/65536 [9:59:21<1:20:04,  1.58it/s]
 88%|████████▊ | 57937/65536 [9:59:21<1:19:35,  1.59it/s]
 88%|████████▊ | 57938/65536 [9:59:22<1:17:37,  1.63it/s]
 88%|████████▊ | 57939/65536 [9:59:22<1:19:42,  1.59it/s]
 88%|████████▊ | 57940/65536 [9:59:23<1:20:40,  1.57it/s]
                                                         
{'loss': 1.542, 'learning_rate': 2.0463450471409333e-07, 'epoch': 3576.54}
+
 88%|████████▊ | 57940/65536 [9:59:23<1:20:40,  1.57it/s]
 88%|████████▊ | 57941/65536 [9:59:24<1:19:39,  1.59it/s]
 88%|████████▊ | 57942/65536 [9:59:24<1:18:51,  1.60it/s]
 88%|████████▊ | 57943/65536 [9:59:25<1:16:44,  1.65it/s]
 88%|████████▊ | 57944/65536 [9:59:25<1:17:15,  1.64it/s]
 88%|████████▊ | 57945/65536 [9:59:26<1:18:02,  1.62it/s]
 88%|████████▊ | 57946/65536 [9:59:27<1:16:50,  1.65it/s]
 88%|████████▊ | 57947/65536 [9:59:27<1:17:54,  1.62it/s]
 88%|████████▊ | 57948/65536 [9:59:28<1:22:37,  1.53it/s]
 88%|████████▊ | 57949/65536 [9:59:29<1:20:57,  1.56it/s]
 88%|████████▊ | 57950/65536 [9:59:29<1:20:20,  1.57it/s]
 88%|████████▊ | 57951/65536 [9:59:30<1:20:00,  1.58it/s]
 88%|████████▊ | 57952/65536 [9:59:31<1:19:40,  1.59it/s]
 88%|��███████▊ | 57953/65536 [9:59:31<1:20:03,  1.58it/s]
 88%|████████▊ | 57954/65536 [9:59:32<1:19:31,  1.59it/s]
 88%|████████▊ | 57955/65536 [9:59:32<1:18:31,  1.61it/s]
 88%|████████▊ | 57956/65536 [9:59:33<1:19:37,  1.59it/s]
 88%|████████▊ | 57957/65536 [9:59:34<1:19:27,  1.59it/s]
 88%|████████▊ | 57958/65536 [9:59:34<1:17:54,  1.62it/s]
 88%|████████▊ | 57959/65536 [9:59:35<1:17:21,  1.63it/s]
 88%|████████▊ | 57960/65536 [9:59:35<1:16:14,  1.66it/s]
                                                         
{'loss': 1.5493, 'learning_rate': 2.0435900575486708e-07, 'epoch': 3577.78}
+
 88%|████████▊ | 57960/65536 [9:59:35<1:16:14,  1.66it/s]
 88%|████████▊ | 57961/65536 [9:59:36<1:16:23,  1.65it/s]
 88%|████████▊ | 57962/65536 [9:59:37<1:17:19,  1.63it/s]
 88%|████████▊ | 57963/65536 [9:59:37<1:18:24,  1.61it/s]
 88%|████████▊ | 57964/65536 [9:59:38<1:22:15,  1.53it/s]
 88%|████████▊ | 57965/65536 [9:59:39<1:19:30,  1.59it/s]
 88%|████████▊ | 57966/65536 [9:59:39<1:17:41,  1.62it/s]
 88%|████████▊ | 57967/65536 [9:59:40<1:21:44,  1.54it/s]
 88%|████████▊ | 57968/65536 [9:59:41<1:21:29,  1.55it/s]
 88%|████████▊ | 57969/65536 [9:59:41<1:21:33,  1.55it/s]
 88%|████████▊ | 57970/65536 [9:59:42<1:20:46,  1.56it/s]
 88%|████████▊ | 57971/65536 [9:59:43<1:20:56,  1.56it/s]
 88%|████████▊ | 57972/65536 [9:59:43<1:18:32,  1.61it/s]
 88%|████████▊ | 57973/65536 [9:59:44<1:17:11,  1.63it/s]
 88%|████████▊ | 57974/65536 [9:59:44<1:18:30,  1.61it/s]
 88%|████████▊ | 57975/65536 [9:59:45<1:17:05,  1.63it/s]
 88%|████████▊ | 57976/65536 [9:59:46<1:17:04,  1.63it/s]
 88%|████████▊ | 57977/65536 [9:59:46<1:16:03,  1.66it/s]
 88%|████████▊ | 57978/65536 [9:59:47<1:16:23,  1.65it/s]
 88%|████████▊ | 57979/65536 [9:59:47<1:16:57,  1.64it/s]
 88%|████████▊ | 57980/65536 [9:59:48<1:21:30,  1.54it/s]
                                                         
{'loss': 1.505, 'learning_rate': 2.0408350679564094e-07, 'epoch': 3579.01}
+
 88%|████████▊ | 57980/65536 [9:59:48<1:21:30,  1.54it/s]
 88%|████████▊ | 57981/65536 [9:59:49<1:20:03,  1.57it/s]
 88%|████████▊ | 57982/65536 [9:59:49<1:21:17,  1.55it/s]
 88%|████████▊ | 57983/65536 [9:59:50<1:20:14,  1.57it/s]
 88%|████████▊ | 57984/65536 [9:59:51<1:20:29,  1.56it/s]
 88%|████████▊ | 57985/65536 [9:59:51<1:20:23,  1.57it/s]
 88%|████████▊ | 57986/65536 [9:59:52<1:18:19,  1.61it/s]
 88%|████████▊ | 57987/65536 [9:59:53<1:20:00,  1.57it/s]
 88%|████████▊ | 57988/65536 [9:59:53<1:17:49,  1.62it/s]
 88%|████████▊ | 57989/65536 [9:59:54<1:18:01,  1.61it/s]
 88%|████████▊ | 57990/65536 [9:59:54<1:16:25,  1.65it/s]
 88%|████████▊ | 57991/65536 [9:59:55<1:16:21,  1.65it/s]
 88%|████████▊ | 57992/65536 [9:59:56<1:17:04,  1.63it/s]
 88%|████████▊ | 57993/65536 [9:59:56<1:16:34,  1.64it/s]
 88%|████████▊ | 57994/65536 [9:59:57<1:16:20,  1.65it/s]
 88%|████████▊ | 57995/65536 [9:59:57<1:15:45,  1.66it/s]
 88%|████████▊ | 57996/65536 [9:59:58<1:15:57,  1.65it/s]
 88%|████████▊ | 57997/65536 [9:59:59<1:19:48,  1.57it/s]
 88%|████████▊ | 57998/65536 [9:59:59<1:17:51,  1.61it/s]
 88%|████████▊ | 57999/65536 [10:00:00<1:18:10,  1.61it/s]
 89%|████████▊ | 58000/65536 [10:00:00<1:17:15,  1.63it/s]
                                                          
{'loss': 1.5339, 'learning_rate': 2.038080078364148e-07, 'epoch': 3580.25}
+
 89%|████████▊ | 58000/65536 [10:00:00<1:17:15,  1.63it/s]
 89%|████████▊ | 58001/65536 [10:00:01<1:17:03,  1.63it/s]
 89%|████████▊ | 58002/65536 [10:00:02<1:16:39,  1.64it/s]
 89%|████████▊ | 58003/65536 [10:00:02<1:16:26,  1.64it/s]
 89%|████████▊ | 58004/65536 [10:00:03<1:14:57,  1.67it/s]
 89%|████████▊ | 58005/65536 [10:00:03<1:15:36,  1.66it/s]
 89%|████████▊ | 58006/65536 [10:00:04<1:16:26,  1.64it/s]
 89%|████████▊ | 58007/65536 [10:00:05<1:15:34,  1.66it/s]
 89%|████████▊ | 58008/65536 [10:00:05<1:15:37,  1.66it/s]
 89%|████████▊ | 58009/65536 [10:00:06<1:15:04,  1.67it/s]
 89%|████████▊ | 58010/65536 [10:00:06<1:15:48,  1.65it/s]
 89%|████████▊ | 58011/65536 [10:00:07<1:16:38,  1.64it/s]
 89%|████████▊ | 58012/65536 [10:00:08<1:17:29,  1.62it/s]
 89%|████████▊ | 58013/65536 [10:00:08<1:19:45,  1.57it/s]
 89%|████████▊ | 58014/65536 [10:00:09<1:19:00,  1.59it/s]
 89%|████████▊ | 58015/65536 [10:00:10<1:17:17,  1.62it/s]
 89%|████████▊ | 58016/65536 [10:00:10<1:17:19,  1.62it/s]
 89%|████████▊ | 58017/65536 [10:00:11<1:16:17,  1.64it/s]
 89%|████████▊ | 58018/65536 [10:00:11<1:16:26,  1.64it/s]
 89%|████████▊ | 58019/65536 [10:00:12<1:16:57,  1.63it/s]
 89%|████████▊ | 58020/65536 [10:00:13<1:17:48,  1.61it/s]
                                                          
{'loss': 1.5404, 'learning_rate': 2.0353250887718869e-07, 'epoch': 3581.48}
+
 89%|████████▊ | 58020/65536 [10:00:13<1:17:48,  1.61it/s]
 89%|████████▊ | 58021/65536 [10:00:13<1:19:26,  1.58it/s]
 89%|████████▊ | 58022/65536 [10:00:14<1:17:49,  1.61it/s]
 89%|████████▊ | 58023/65536 [10:00:15<1:16:26,  1.64it/s]
 89%|████████▊ | 58024/65536 [10:00:15<1:19:25,  1.58it/s]
 89%|████████▊ | 58025/65536 [10:00:16<1:17:52,  1.61it/s]
 89%|████████▊ | 58026/65536 [10:00:16<1:18:07,  1.60it/s]
 89%|████████▊ | 58027/65536 [10:00:17<1:17:01,  1.62it/s]
 89%|████████▊ | 58028/65536 [10:00:18<1:17:41,  1.61it/s]
 89%|████████▊ | 58029/65536 [10:00:18<1:21:00,  1.54it/s]
 89%|████████▊ | 58030/65536 [10:00:19<1:19:00,  1.58it/s]
 89%|████████▊ | 58031/65536 [10:00:20<1:17:27,  1.62it/s]
 89%|████████▊ | 58032/65536 [10:00:20<1:19:21,  1.58it/s]
 89%|████████▊ | 58033/65536 [10:00:21<1:17:25,  1.62it/s]
 89%|████████▊ | 58034/65536 [10:00:21<1:18:11,  1.60it/s]
 89%|████████▊ | 58035/65536 [10:00:22<1:16:32,  1.63it/s]
 89%|████████▊ | 58036/65536 [10:00:23<1:16:43,  1.63it/s]
 89%|████████▊ | 58037/65536 [10:00:23<1:18:05,  1.60it/s]
 89%|████████▊ | 58038/65536 [10:00:24<1:15:30,  1.66it/s]
 89%|████████▊ | 58039/65536 [10:00:24<1:16:22,  1.64it/s]
 89%|████████▊ | 58040/65536 [10:00:25<1:16:41,  1.63it/s]
                                                          
{'loss': 1.5498, 'learning_rate': 2.0325700991796254e-07, 'epoch': 3582.72}
+
 89%|████████▊ | 58040/65536 [10:00:25<1:16:41,  1.63it/s]
 89%|████████▊ | 58041/65536 [10:00:26<1:17:22,  1.61it/s]
 89%|████████▊ | 58042/65536 [10:00:26<1:17:46,  1.61it/s]
 89%|████████▊ | 58043/65536 [10:00:27<1:15:48,  1.65it/s]
 89%|████████▊ | 58044/65536 [10:00:28<1:16:25,  1.63it/s]
 89%|████████▊ | 58045/65536 [10:00:28<1:18:15,  1.60it/s]
 89%|████████▊ | 58046/65536 [10:00:29<1:17:50,  1.60it/s]
 89%|████████▊ | 58047/65536 [10:00:29<1:16:37,  1.63it/s]
 89%|████████▊ | 58048/65536 [10:00:30<1:15:54,  1.64it/s]
 89%|████████▊ | 58049/65536 [10:00:31<1:16:42,  1.63it/s]
 89%|████████▊ | 58050/65536 [10:00:31<1:17:39,  1.61it/s]
 89%|████████▊ | 58051/65536 [10:00:32<1:16:07,  1.64it/s]
 89%|████████▊ | 58052/65536 [10:00:33<1:16:50,  1.62it/s]
 89%|████████▊ | 58053/65536 [10:00:33<1:17:17,  1.61it/s]
 89%|████████▊ | 58054/65536 [10:00:34<1:17:05,  1.62it/s]
 89%|████████▊ | 58055/65536 [10:00:34<1:15:27,  1.65it/s]
 89%|████████▊ | 58056/65536 [10:00:35<1:16:36,  1.63it/s]
 89%|████████▊ | 58057/65536 [10:00:36<1:17:38,  1.61it/s]
 89%|████████▊ | 58058/65536 [10:00:36<1:17:14,  1.61it/s]
 89%|████████▊ | 58059/65536 [10:00:37<1:18:11,  1.59it/s]
 89%|████████▊ | 58060/65536 [10:00:37<1:16:59,  1.62it/s]
                                                          
{'loss': 1.5171, 'learning_rate': 2.029815109587364e-07, 'epoch': 3583.95}
+
 89%|████████▊ | 58060/65536 [10:00:37<1:16:59,  1.62it/s]
 89%|████████▊ | 58061/65536 [10:00:38<1:19:32,  1.57it/s]
 89%|████████▊ | 58062/65536 [10:00:39<1:18:55,  1.58it/s]
 89%|████████▊ | 58063/65536 [10:00:39<1:18:30,  1.59it/s]
 89%|████████▊ | 58064/65536 [10:00:40<1:17:30,  1.61it/s]
 89%|████████▊ | 58065/65536 [10:00:41<1:17:18,  1.61it/s]
 89%|████████▊ | 58066/65536 [10:00:41<1:16:47,  1.62it/s]
 89%|████████▊ | 58067/65536 [10:00:42<1:16:35,  1.63it/s]
 89%|████████▊ | 58068/65536 [10:00:43<1:18:47,  1.58it/s]
 89%|████████▊ | 58069/65536 [10:00:43<1:21:57,  1.52it/s]
 89%|��███████▊ | 58070/65536 [10:00:44<1:20:09,  1.55it/s]
 89%|████████▊ | 58071/65536 [10:00:44<1:18:15,  1.59it/s]
 89%|████████▊ | 58072/65536 [10:00:45<1:16:18,  1.63it/s]
 89%|████████▊ | 58073/65536 [10:00:46<1:17:39,  1.60it/s]
 89%|████████▊ | 58074/65536 [10:00:46<1:16:17,  1.63it/s]
 89%|████████▊ | 58075/65536 [10:00:47<1:16:53,  1.62it/s]
 89%|████████▊ | 58076/65536 [10:00:47<1:16:28,  1.63it/s]
 89%|████████▊ | 58077/65536 [10:00:48<1:15:02,  1.66it/s]
 89%|████████▊ | 58078/65536 [10:00:49<1:17:02,  1.61it/s]
 89%|████████▊ | 58079/65536 [10:00:49<1:16:16,  1.63it/s]
 89%|████████▊ | 58080/65536 [10:00:50<1:16:48,  1.62it/s]
                                                          
{'loss': 1.5456, 'learning_rate': 2.0270601199951026e-07, 'epoch': 3585.19}
+
 89%|████████▊ | 58080/65536 [10:00:50<1:16:48,  1.62it/s]
 89%|████████▊ | 58081/65536 [10:00:51<1:16:05,  1.63it/s]
 89%|████████▊ | 58082/65536 [10:00:51<1:16:18,  1.63it/s]
 89%|████████▊ | 58083/65536 [10:00:52<1:17:43,  1.60it/s]
 89%|████████▊ | 58084/65536 [10:00:52<1:18:25,  1.58it/s]
 89%|████████▊ | 58085/65536 [10:00:53<1:15:40,  1.64it/s]
 89%|████████▊ | 58086/65536 [10:00:54<1:14:40,  1.66it/s]
 89%|████████▊ | 58087/65536 [10:00:54<1:16:51,  1.62it/s]
 89%|████████▊ | 58088/65536 [10:00:55<1:17:28,  1.60it/s]
 89%|████████▊ | 58089/65536 [10:00:56<1:16:29,  1.62it/s]
 89%|████████▊ | 58090/65536 [10:00:56<1:17:07,  1.61it/s]
 89%|████████▊ | 58091/65536 [10:00:57<1:15:05,  1.65it/s]
 89%|████████▊ | 58092/65536 [10:00:57<1:15:41,  1.64it/s]
 89%|████████▊ | 58093/65536 [10:00:58<1:15:24,  1.65it/s]
 89%|████████▊ | 58094/65536 [10:00:59<1:17:25,  1.60it/s]
 89%|████████▊ | 58095/65536 [10:00:59<1:15:31,  1.64it/s]
 89%|████████▊ | 58096/65536 [10:01:00<1:16:00,  1.63it/s]
 89%|████████▊ | 58097/65536 [10:01:00<1:17:22,  1.60it/s]
 89%|████████▊ | 58098/65536 [10:01:01<1:16:39,  1.62it/s]
 89%|████████▊ | 58099/65536 [10:01:02<1:18:11,  1.59it/s]
 89%|████████▊ | 58100/65536 [10:01:02<1:17:30,  1.60it/s]
                                                          
{'loss': 1.5425, 'learning_rate': 2.0243051304028404e-07, 'epoch': 3586.42}
+
 89%|████████▊ | 58100/65536 [10:01:02<1:17:30,  1.60it/s]
 89%|████████▊ | 58101/65536 [10:01:03<1:17:21,  1.60it/s]
 89%|████████▊ | 58102/65536 [10:01:04<1:16:26,  1.62it/s]
 89%|████████▊ | 58103/65536 [10:01:04<1:15:45,  1.64it/s]
 89%|████████▊ | 58104/65536 [10:01:05<1:16:44,  1.61it/s]
 89%|████████▊ | 58105/65536 [10:01:05<1:18:18,  1.58it/s]
 89%|████████▊ | 58106/65536 [10:01:06<1:18:08,  1.58it/s]
 89%|████████▊ | 58107/65536 [10:01:07<1:15:07,  1.65it/s]
 89%|████████▊ | 58108/65536 [10:01:07<1:15:54,  1.63it/s]
 89%|████████▊ | 58109/65536 [10:01:08<1:17:23,  1.60it/s]
 89%|████████▊ | 58110/65536 [10:01:09<1:19:46,  1.55it/s]
 89%|████████▊ | 58111/65536 [10:01:09<1:18:34,  1.58it/s]
 89%|████████▊ | 58112/65536 [10:01:10<1:19:06,  1.56it/s]
 89%|████████▊ | 58113/65536 [10:01:10<1:18:40,  1.57it/s]
 89%|████████▊ | 58114/65536 [10:01:11<1:17:07,  1.60it/s]
 89%|████████▊ | 58115/65536 [10:01:12<1:18:14,  1.58it/s]
 89%|████████▊ | 58116/65536 [10:01:12<1:18:13,  1.58it/s]
 89%|████████▊ | 58117/65536 [10:01:13<1:17:42,  1.59it/s]
 89%|████████▊ | 58118/65536 [10:01:14<1:17:45,  1.59it/s]
 89%|████████▊ | 58119/65536 [10:01:14<1:21:01,  1.53it/s]
 89%|████████▊ | 58120/65536 [10:01:15<1:22:45,  1.49it/s]
                                                          
{'loss': 1.4761, 'learning_rate': 2.021550140810579e-07, 'epoch': 3587.65}
+
 89%|████████▊ | 58120/65536 [10:01:15<1:22:45,  1.49it/s]
 89%|████████▊ | 58121/65536 [10:01:16<1:21:34,  1.52it/s]
 89%|████████▊ | 58122/65536 [10:01:16<1:19:44,  1.55it/s]
 89%|████████▊ | 58123/65536 [10:01:17<1:18:30,  1.57it/s]
 89%|████████▊ | 58124/65536 [10:01:17<1:17:44,  1.59it/s]
 89%|████████▊ | 58125/65536 [10:01:18<1:19:18,  1.56it/s]
 89%|████████▊ | 58126/65536 [10:01:19<1:19:33,  1.55it/s]
 89%|████████▊ | 58127/65536 [10:01:20<1:23:03,  1.49it/s]
 89%|████████▊ | 58128/65536 [10:01:20<1:23:22,  1.48it/s]
 89%|████████▊ | 58129/65536 [10:01:21<1:21:27,  1.52it/s]
 89%|████████▊ | 58130/65536 [10:01:21<1:19:57,  1.54it/s]
 89%|████████▊ | 58131/65536 [10:01:22<1:19:48,  1.55it/s]
 89%|████████▊ | 58132/65536 [10:01:23<1:16:55,  1.60it/s]
 89%|████████▊ | 58133/65536 [10:01:23<1:17:12,  1.60it/s]
 89%|████████▊ | 58134/65536 [10:01:24<1:16:58,  1.60it/s]
 89%|████████▊ | 58135/65536 [10:01:25<1:15:20,  1.64it/s]
 89%|████████▊ | 58136/65536 [10:01:25<1:14:59,  1.64it/s]
 89%|████████▊ | 58137/65536 [10:01:26<1:14:40,  1.65it/s]
 89%|████████▊ | 58138/65536 [10:01:26<1:13:54,  1.67it/s]
 89%|████████▊ | 58139/65536 [10:01:27<1:12:47,  1.69it/s]
 89%|████████▊ | 58140/65536 [10:01:27<1:13:38,  1.67it/s]
                                                          
{'loss': 1.5021, 'learning_rate': 2.0187951512183176e-07, 'epoch': 3588.89}
+
 89%|████████▊ | 58140/65536 [10:01:27<1:13:38,  1.67it/s]
 89%|████████▊ | 58141/65536 [10:01:28<1:14:52,  1.65it/s]
 89%|████████▊ | 58142/65536 [10:01:29<1:17:31,  1.59it/s]
 89%|████████▊ | 58143/65536 [10:01:29<1:15:47,  1.63it/s]
 89%|████████▊ | 58144/65536 [10:01:30<1:15:48,  1.63it/s]
 89%|████████▊ | 58145/65536 [10:01:31<1:15:12,  1.64it/s]
 89%|████████▊ | 58146/65536 [10:01:31<1:15:26,  1.63it/s]
 89%|████████▊ | 58147/65536 [10:01:32<1:14:02,  1.66it/s]
 89%|████████▊ | 58148/65536 [10:01:32<1:15:01,  1.64it/s]
 89%|████████▊ | 58149/65536 [10:01:33<1:14:26,  1.65it/s]
 89%|████████▊ | 58150/65536 [10:01:34<1:16:06,  1.62it/s]
 89%|████████▊ | 58151/65536 [10:01:34<1:18:15,  1.57it/s]
 89%|████████▊ | 58152/65536 [10:01:35<1:16:44,  1.60it/s]
 89%|████████▊ | 58153/65536 [10:01:36<1:16:11,  1.61it/s]
 89%|████████▊ | 58154/65536 [10:01:36<1:16:45,  1.60it/s]
 89%|████████▊ | 58155/65536 [10:01:37<1:16:50,  1.60it/s]
 89%|████████▊ | 58156/65536 [10:01:37<1:17:27,  1.59it/s]
 89%|████████▊ | 58157/65536 [10:01:38<1:15:57,  1.62it/s]
 89%|████████▊ | 58158/65536 [10:01:39<1:15:51,  1.62it/s]
 89%|████████▊ | 58159/65536 [10:01:39<1:18:24,  1.57it/s]
 89%|████████▊ | 58160/65536 [10:01:40<1:16:38,  1.60it/s]
                                                          
{'loss': 1.5373, 'learning_rate': 2.0160401616260562e-07, 'epoch': 3590.12}
+
 89%|████████▊ | 58160/65536 [10:01:40<1:16:38,  1.60it/s]
 89%|████████▊ | 58161/65536 [10:01:41<1:17:55,  1.58it/s]
 89%|████████▊ | 58162/65536 [10:01:41<1:17:52,  1.58it/s]
 89%|████████▊ | 58163/65536 [10:01:42<1:19:37,  1.54it/s]
 89%|████████▉ | 58164/65536 [10:01:43<1:18:11,  1.57it/s]
 89%|████████▉ | 58165/65536 [10:01:43<1:16:30,  1.61it/s]
 89%|████████▉ | 58166/65536 [10:01:44<1:16:08,  1.61it/s]
 89%|████████▉ | 58167/65536 [10:01:44<1:16:26,  1.61it/s]
 89%|████████▉ | 58168/65536 [10:01:45<1:15:39,  1.62it/s]
 89%|████████▉ | 58169/65536 [10:01:46<1:17:31,  1.58it/s]
 89%|████████▉ | 58170/65536 [10:01:46<1:17:50,  1.58it/s]
 89%|████████▉ | 58171/65536 [10:01:47<1:16:18,  1.61it/s]
 89%|████████▉ | 58172/65536 [10:01:47<1:16:36,  1.60it/s]
 89%|████████▉ | 58173/65536 [10:01:48<1:15:43,  1.62it/s]
 89%|████████▉ | 58174/65536 [10:01:49<1:17:37,  1.58it/s]
 89%|████████▉ | 58175/65536 [10:01:49<1:19:55,  1.53it/s]
 89%|████████▉ | 58176/65536 [10:01:50<1:19:27,  1.54it/s]
 89%|████████▉ | 58177/65536 [10:01:51<1:19:57,  1.53it/s]
 89%|████████▉ | 58178/65536 [10:01:51<1:18:50,  1.56it/s]
 89%|████████▉ | 58179/65536 [10:01:52<1:18:28,  1.56it/s]
 89%|████████▉ | 58180/65536 [10:01:53<1:19:29,  1.54it/s]
                                                          
{'loss': 1.5045, 'learning_rate': 2.0132851720337948e-07, 'epoch': 3591.36}
+
 89%|████████▉ | 58180/65536 [10:01:53<1:19:29,  1.54it/s]
 89%|████████▉ | 58181/65536 [10:01:53<1:18:51,  1.55it/s]
 89%|████████▉ | 58182/65536 [10:01:54<1:18:19,  1.56it/s]
 89%|████████▉ | 58183/65536 [10:01:55<1:18:06,  1.57it/s]
 89%|████████▉ | 58184/65536 [10:01:55<1:15:16,  1.63it/s]
 89%|████████▉ | 58185/65536 [10:01:56<1:13:38,  1.66it/s]
 89%|████████▉ | 58186/65536 [10:01:56<1:14:31,  1.64it/s]
 89%|████████▉ | 58187/65536 [10:01:57<1:12:16,  1.69it/s]
 89%|████████▉ | 58188/65536 [10:01:58<1:17:06,  1.59it/s]
 89%|████████▉ | 58189/65536 [10:01:58<1:16:14,  1.61it/s]
 89%|████████▉ | 58190/65536 [10:01:59<1:14:45,  1.64it/s]
 89%|████████▉ | 58191/65536 [10:01:59<1:17:04,  1.59it/s]
 89%|████████▉ | 58192/65536 [10:02:00<1:15:00,  1.63it/s]
 89%|████████▉ | 58193/65536 [10:02:01<1:14:11,  1.65it/s]
 89%|████████▉ | 58194/65536 [10:02:01<1:13:49,  1.66it/s]
 89%|████████▉ | 58195/65536 [10:02:02<1:13:54,  1.66it/s]
 89%|████████▉ | 58196/65536 [10:02:02<1:14:42,  1.64it/s]
 89%|████████▉ | 58197/65536 [10:02:03<1:14:01,  1.65it/s]
 89%|████████▉ | 58198/65536 [10:02:04<1:13:06,  1.67it/s]
 89%|████████▉ | 58199/65536 [10:02:04<1:13:39,  1.66it/s]
 89%|████████▉ | 58200/65536 [10:02:05<1:14:04,  1.65it/s]
                                                          
{'loss': 1.5985, 'learning_rate': 2.0105301824415334e-07, 'epoch': 3592.59}
+
 89%|████████▉ | 58200/65536 [10:02:05<1:14:04,  1.65it/s]
 89%|████████▉ | 58201/65536 [10:02:06<1:16:30,  1.60it/s]
 89%|████████▉ | 58202/65536 [10:02:06<1:16:18,  1.60it/s]
 89%|████████▉ | 58203/65536 [10:02:07<1:16:21,  1.60it/s]
 89%|████████▉ | 58204/65536 [10:02:07<1:14:44,  1.63it/s]
 89%|████████▉ | 58205/65536 [10:02:08<1:14:03,  1.65it/s]
 89%|████████▉ | 58206/65536 [10:02:09<1:15:33,  1.62it/s]
 89%|████████▉ | 58207/65536 [10:02:09<1:19:14,  1.54it/s]
 89%|████████▉ | 58208/65536 [10:02:10<1:18:11,  1.56it/s]
 89%|████████▉ | 58209/65536 [10:02:11<1:16:16,  1.60it/s]
 89%|████████▉ | 58210/65536 [10:02:11<1:15:37,  1.61it/s]
 89%|████████▉ | 58211/65536 [10:02:12<1:14:41,  1.63it/s]
 89%|████████▉ | 58212/65536 [10:02:12<1:14:24,  1.64it/s]
 89%|████████▉ | 58213/65536 [10:02:13<1:15:52,  1.61it/s]
 89%|████████▉ | 58214/65536 [10:02:14<1:15:22,  1.62it/s]
 89%|████████▉ | 58215/65536 [10:02:14<1:15:48,  1.61it/s]
 89%|████████▉ | 58216/65536 [10:02:15<1:14:21,  1.64it/s]
 89%|████████▉ | 58217/65536 [10:02:15<1:14:00,  1.65it/s]
 89%|████████▉ | 58218/65536 [10:02:16<1:13:48,  1.65it/s]
 89%|████████▉ | 58219/65536 [10:02:17<1:15:12,  1.62it/s]
 89%|████████▉ | 58220/65536 [10:02:17<1:15:12,  1.62it/s]
                                                          
{'loss': 1.5325, 'learning_rate': 2.007775192849271e-07, 'epoch': 3593.83}
+
 89%|████████▉ | 58220/65536 [10:02:17<1:15:12,  1.62it/s]
 89%|████████▉ | 58221/65536 [10:02:18<1:17:36,  1.57it/s]
 89%|████████▉ | 58222/65536 [10:02:19<1:16:16,  1.60it/s]
 89%|████████▉ | 58223/65536 [10:02:19<1:17:10,  1.58it/s]
 89%|████████▉ | 58224/65536 [10:02:20<1:17:34,  1.57it/s]
 89%|████████▉ | 58225/65536 [10:02:20<1:16:52,  1.59it/s]
 89%|████████▉ | 58226/65536 [10:02:21<1:15:48,  1.61it/s]
 89%|████████▉ | 58227/65536 [10:02:22<1:16:20,  1.60it/s]
 89%|████████▉ | 58228/65536 [10:02:22<1:14:18,  1.64it/s]
 89%|████████▉ | 58229/65536 [10:02:23<1:13:31,  1.66it/s]
 89%|████████▉ | 58230/65536 [10:02:23<1:12:23,  1.68it/s]
 89%|████████▉ | 58231/65536 [10:02:24<1:14:17,  1.64it/s]
 89%|████████▉ | 58232/65536 [10:02:25<1:16:05,  1.60it/s]
 89%|████████▉ | 58233/65536 [10:02:25<1:15:43,  1.61it/s]
 89%|████████▉ | 58234/65536 [10:02:26<1:17:32,  1.57it/s]
 89%|████████▉ | 58235/65536 [10:02:27<1:17:08,  1.58it/s]
 89%|████████▉ | 58236/65536 [10:02:27<1:15:43,  1.61it/s]
 89%|████████▉ | 58237/65536 [10:02:28<1:15:43,  1.61it/s]
 89%|████████▉ | 58238/65536 [10:02:28<1:14:23,  1.64it/s]
 89%|████████▉ | 58239/65536 [10:02:29<1:15:28,  1.61it/s]
 89%|████████▉ | 58240/65536 [10:02:30<1:19:37,  1.53it/s]
                                                          
{'loss': 1.5484, 'learning_rate': 2.0050202032570095e-07, 'epoch': 3595.06}
+
 89%|████████▉ | 58240/65536 [10:02:30<1:19:37,  1.53it/s]
 89%|████████▉ | 58241/65536 [10:02:30<1:18:35,  1.55it/s]
 89%|████████▉ | 58242/65536 [10:02:31<1:19:40,  1.53it/s]
 89%|████████▉ | 58243/65536 [10:02:32<1:19:43,  1.52it/s]
 89%|████████▉ | 58244/65536 [10:02:32<1:18:48,  1.54it/s]
 89%|████████▉ | 58245/65536 [10:02:33<1:17:54,  1.56it/s]
 89%|████████▉ | 58246/65536 [10:02:34<1:16:08,  1.60it/s]
 89%|████████▉ | 58247/65536 [10:02:34<1:15:26,  1.61it/s]
 89%|████████▉ | 58248/65536 [10:02:35<1:14:13,  1.64it/s]
 89%|████████▉ | 58249/65536 [10:02:35<1:13:14,  1.66it/s]
 89%|████████▉ | 58250/65536 [10:02:36<1:13:03,  1.66it/s]
 89%|████████▉ | 58251/65536 [10:02:37<1:13:04,  1.66it/s]
 89%|████████▉ | 58252/65536 [10:02:37<1:13:03,  1.66it/s]
 89%|████████▉ | 58253/65536 [10:02:38<1:12:37,  1.67it/s]
 89%|████████▉ | 58254/65536 [10:02:38<1:15:04,  1.62it/s]
 89%|████████▉ | 58255/65536 [10:02:39<1:15:00,  1.62it/s]
 89%|████████▉ | 58256/65536 [10:02:40<1:17:11,  1.57it/s]
 89%|████████▉ | 58257/65536 [10:02:40<1:15:09,  1.61it/s]
 89%|████████▉ | 58258/65536 [10:02:41<1:16:12,  1.59it/s]
 89%|████████▉ | 58259/65536 [10:02:42<1:15:45,  1.60it/s]
 89%|████████▉ | 58260/65536 [10:02:42<1:14:04,  1.64it/s]
                                                          
{'loss': 1.5396, 'learning_rate': 2.002265213664748e-07, 'epoch': 3596.3}
+
 89%|████████▉ | 58260/65536 [10:02:42<1:14:04,  1.64it/s]
 89%|████████▉ | 58261/65536 [10:02:43<1:14:20,  1.63it/s]
 89%|████████▉ | 58262/65536 [10:02:43<1:15:56,  1.60it/s]
 89%|████████▉ | 58263/65536 [10:02:44<1:15:24,  1.61it/s]
 89%|████████▉ | 58264/65536 [10:02:45<1:14:03,  1.64it/s]
 89%|████████▉ | 58265/65536 [10:02:45<1:13:16,  1.65it/s]
 89%|████████▉ | 58266/65536 [10:02:46<1:14:11,  1.63it/s]
 89%|████████▉ | 58267/65536 [10:02:47<1:15:58,  1.59it/s]
 89%|████████▉ | 58268/65536 [10:02:47<1:15:39,  1.60it/s]
 89%|████████▉ | 58269/65536 [10:02:48<1:15:43,  1.60it/s]
 89%|████████▉ | 58270/65536 [10:02:48<1:15:51,  1.60it/s]
 89%|████████▉ | 58271/65536 [10:02:49<1:14:55,  1.62it/s]
 89%|████████▉ | 58272/65536 [10:02:50<1:17:10,  1.57it/s]
 89%|████████▉ | 58273/65536 [10:02:50<1:15:58,  1.59it/s]
 89%|████████▉ | 58274/65536 [10:02:51<1:14:20,  1.63it/s]
 89%|████████▉ | 58275/65536 [10:02:51<1:13:23,  1.65it/s]
 89%|████████▉ | 58276/65536 [10:02:52<1:16:30,  1.58it/s]
 89%|████████▉ | 58277/65536 [10:02:53<1:15:06,  1.61it/s]
 89%|████████▉ | 58278/65536 [10:02:53<1:13:59,  1.63it/s]
 89%|████████▉ | 58279/65536 [10:02:54<1:13:02,  1.66it/s]
 89%|████████▉ | 58280/65536 [10:02:55<1:16:01,  1.59it/s]
                                                          
{'loss': 1.5119, 'learning_rate': 1.9995102240724867e-07, 'epoch': 3597.53}
+
 89%|████████▉ | 58280/65536 [10:02:55<1:16:01,  1.59it/s]
 89%|████████▉ | 58281/65536 [10:02:55<1:16:54,  1.57it/s]
 89%|████████▉ | 58282/65536 [10:02:56<1:16:29,  1.58it/s]
 89%|████████▉ | 58283/65536 [10:02:57<1:15:24,  1.60it/s]
 89%|████████▉ | 58284/65536 [10:02:57<1:14:47,  1.62it/s]
 89%|████████▉ | 58285/65536 [10:02:58<1:14:40,  1.62it/s]
 89%|████████▉ | 58286/65536 [10:02:58<1:13:40,  1.64it/s]
 89%|████████▉ | 58287/65536 [10:02:59<1:15:18,  1.60it/s]
 89%|████████▉ | 58288/65536 [10:03:00<1:16:32,  1.58it/s]
 89%|████████▉ | 58289/65536 [10:03:00<1:16:05,  1.59it/s]
 89%|████████▉ | 58290/65536 [10:03:01<1:19:10,  1.53it/s]
 89%|████████▉ | 58291/65536 [10:03:02<1:17:36,  1.56it/s]
 89%|████████▉ | 58292/65536 [10:03:02<1:16:25,  1.58it/s]
 89%|████████▉ | 58293/65536 [10:03:03<1:15:14,  1.60it/s]
 89%|████████▉ | 58294/65536 [10:03:03<1:14:25,  1.62it/s]
 89%|████████▉ | 58295/65536 [10:03:04<1:16:34,  1.58it/s]
 89%|████████▉ | 58296/65536 [10:03:05<1:15:39,  1.60it/s]
 89%|████████▉ | 58297/65536 [10:03:05<1:15:52,  1.59it/s]
 89%|████████▉ | 58298/65536 [10:03:06<1:16:33,  1.58it/s]
 89%|████████▉ | 58299/65536 [10:03:07<1:16:49,  1.57it/s]
 89%|████████▉ | 58300/65536 [10:03:07<1:14:49,  1.61it/s]
                                                          
{'loss': 1.5147, 'learning_rate': 1.9967552344802252e-07, 'epoch': 3598.77}
+
 89%|████████▉ | 58300/65536 [10:03:07<1:14:49,  1.61it/s]
 89%|████████▉ | 58301/65536 [10:03:08<1:13:37,  1.64it/s]
 89%|████████▉ | 58302/65536 [10:03:08<1:12:11,  1.67it/s]
 89%|████████▉ | 58303/65536 [10:03:09<1:12:22,  1.67it/s]
 89%|████████▉ | 58304/65536 [10:03:10<1:14:22,  1.62it/s]
 89%|████████▉ | 58305/65536 [10:03:10<1:15:43,  1.59it/s]
 89%|████████▉ | 58306/65536 [10:03:11<1:13:34,  1.64it/s]
 89%|████████▉ | 58307/65536 [10:03:11<1:14:10,  1.62it/s]
 89%|████████▉ | 58308/65536 [10:03:12<1:14:16,  1.62it/s]
 89%|████████▉ | 58309/65536 [10:03:13<1:14:43,  1.61it/s]
 89%|████████▉ | 58310/65536 [10:03:13<1:14:06,  1.63it/s]
 89%|████████▉ | 58311/65536 [10:03:14<1:13:13,  1.64it/s]
 89%|████████▉ | 58312/65536 [10:03:15<1:14:50,  1.61it/s]
 89%|████████▉ | 58313/65536 [10:03:15<1:13:40,  1.63it/s]
 89%|████████▉ | 58314/65536 [10:03:16<1:14:15,  1.62it/s]
 89%|████████▉ | 58315/65536 [10:03:16<1:13:36,  1.63it/s]
 89%|████████▉ | 58316/65536 [10:03:17<1:12:10,  1.67it/s]
 89%|████████▉ | 58317/65536 [10:03:18<1:13:57,  1.63it/s]
 89%|████████▉ | 58318/65536 [10:03:18<1:15:32,  1.59it/s]
 89%|████████▉ | 58319/65536 [10:03:19<1:14:49,  1.61it/s]
 89%|████████▉ | 58320/65536 [10:03:20<1:15:49,  1.59it/s]
                                                          
{'loss': 1.5411, 'learning_rate': 1.9940002448879638e-07, 'epoch': 3600.0}
+
 89%|████████▉ | 58320/65536 [10:03:20<1:15:49,  1.59it/s]
 89%|████████▉ | 58321/65536 [10:03:20<1:16:06,  1.58it/s]
 89%|████████▉ | 58322/65536 [10:03:21<1:18:22,  1.53it/s]
 89%|████████▉ | 58323/65536 [10:03:21<1:15:58,  1.58it/s]
 89%|████████▉ | 58324/65536 [10:03:22<1:17:07,  1.56it/s]
 89%|████████▉ | 58325/65536 [10:03:23<1:16:03,  1.58it/s]
 89%|████████▉ | 58326/65536 [10:03:23<1:16:38,  1.57it/s]
 89%|████████▉ | 58327/65536 [10:03:24<1:15:04,  1.60it/s]
 89%|████████▉ | 58328/65536 [10:03:25<1:13:08,  1.64it/s]
 89%|████████▉ | 58329/65536 [10:03:25<1:12:46,  1.65it/s]
 89%|████████▉ | 58330/65536 [10:03:26<1:14:00,  1.62it/s]
 89%|████████▉ | 58331/65536 [10:03:26<1:14:05,  1.62it/s]
 89%|████████▉ | 58332/65536 [10:03:27<1:12:38,  1.65it/s]
 89%|████████▉ | 58333/65536 [10:03:28<1:13:09,  1.64it/s]
 89%|████████▉ | 58334/65536 [10:03:28<1:12:48,  1.65it/s]
 89%|████████▉ | 58335/65536 [10:03:29<1:14:11,  1.62it/s]
 89%|████████▉ | 58336/65536 [10:03:29<1:14:35,  1.61it/s]
 89%|████████▉ | 58337/65536 [10:03:30<1:16:38,  1.57it/s]
 89%|████████▉ | 58338/65536 [10:03:31<1:15:47,  1.58it/s]
 89%|████████▉ | 58339/65536 [10:03:31<1:15:20,  1.59it/s]
 89%|████████▉ | 58340/65536 [10:03:32<1:13:59,  1.62it/s]
                                                          
{'loss': 1.5351, 'learning_rate': 1.9912452552957024e-07, 'epoch': 3601.23}
+
 89%|████████▉ | 58340/65536 [10:03:32<1:13:59,  1.62it/s]
 89%|████████▉ | 58341/65536 [10:03:33<1:14:03,  1.62it/s]
 89%|████████▉ | 58342/65536 [10:03:33<1:14:12,  1.62it/s]
 89%|████████▉ | 58343/65536 [10:03:34<1:13:22,  1.63it/s]
 89%|████████▉ | 58344/65536 [10:03:34<1:12:07,  1.66it/s]
 89%|████████▉ | 58345/65536 [10:03:35<1:13:13,  1.64it/s]
 89%|████████▉ | 58346/65536 [10:03:36<1:14:21,  1.61it/s]
 89%|████████▉ | 58347/65536 [10:03:36<1:16:21,  1.57it/s]
 89%|████████▉ | 58348/65536 [10:03:37<1:14:06,  1.62it/s]
 89%|████████▉ | 58349/65536 [10:03:38<1:15:17,  1.59it/s]
 89%|████████▉ | 58350/65536 [10:03:38<1:15:13,  1.59it/s]
 89%|████████▉ | 58351/65536 [10:03:39<1:15:09,  1.59it/s]
 89%|████████▉ | 58352/65536 [10:03:39<1:15:16,  1.59it/s]
 89%|████████▉ | 58353/65536 [10:03:40<1:17:32,  1.54it/s]
 89%|████████▉ | 58354/65536 [10:03:41<1:17:15,  1.55it/s]
 89%|████████▉ | 58355/65536 [10:03:41<1:17:20,  1.55it/s]
 89%|████████▉ | 58356/65536 [10:03:42<1:14:46,  1.60it/s]
 89%|████████▉ | 58357/65536 [10:03:43<1:14:57,  1.60it/s]
 89%|████████▉ | 58358/65536 [10:03:43<1:14:02,  1.62it/s]
 89%|████████▉ | 58359/65536 [10:03:44<1:12:44,  1.64it/s]
 89%|████████▉ | 58360/65536 [10:03:44<1:12:44,  1.64it/s]
                                                          
{'loss': 1.5331, 'learning_rate': 1.98849026570344e-07, 'epoch': 3602.47}
+
 89%|████████▉ | 58360/65536 [10:03:44<1:12:44,  1.64it/s]
 89%|████████▉ | 58361/65536 [10:03:45<1:12:16,  1.65it/s]
 89%|████████▉ | 58362/65536 [10:03:46<1:12:47,  1.64it/s]
 89%|████████▉ | 58363/65536 [10:03:46<1:14:34,  1.60it/s]
 89%|████████▉ | 58364/65536 [10:03:47<1:15:34,  1.58it/s]
 89%|████████▉ | 58365/65536 [10:03:48<1:13:40,  1.62it/s]
 89%|████████▉ | 58366/65536 [10:03:48<1:13:38,  1.62it/s]
 89%|████████▉ | 58367/65536 [10:03:49<1:12:56,  1.64it/s]
 89%|████████▉ | 58368/65536 [10:03:49<1:13:24,  1.63it/s]
 89%|████████▉ | 58369/65536 [10:03:50<1:15:48,  1.58it/s]
 89%|████████▉ | 58370/65536 [10:03:51<1:15:30,  1.58it/s]
 89%|████████▉ | 58371/65536 [10:03:51<1:14:15,  1.61it/s]
 89%|████████▉ | 58372/65536 [10:03:52<1:13:42,  1.62it/s]
 89%|████████▉ | 58373/65536 [10:03:52<1:12:05,  1.66it/s]
 89%|████████▉ | 58374/65536 [10:03:53<1:11:23,  1.67it/s]
 89%|████████▉ | 58375/65536 [10:03:54<1:11:45,  1.66it/s]
 89%|████████▉ | 58376/65536 [10:03:54<1:10:34,  1.69it/s]
 89%|████████▉ | 58377/65536 [10:03:55<1:12:49,  1.64it/s]
 89%|████████▉ | 58378/65536 [10:03:55<1:11:54,  1.66it/s]
 89%|████████▉ | 58379/65536 [10:03:56<1:12:18,  1.65it/s]
 89%|████████▉ | 58380/65536 [10:03:57<1:11:15,  1.67it/s]
                                                          
{'loss': 1.5585, 'learning_rate': 1.9857352761111788e-07, 'epoch': 3603.7}
+
 89%|████████▉ | 58380/65536 [10:03:57<1:11:15,  1.67it/s]
 89%|████████▉ | 58381/65536 [10:03:57<1:11:56,  1.66it/s]
 89%|████████▉ | 58382/65536 [10:03:58<1:14:08,  1.61it/s]
 89%|████████▉ | 58383/65536 [10:03:59<1:13:35,  1.62it/s]
 89%|████████▉ | 58384/65536 [10:03:59<1:15:37,  1.58it/s]
 89%|████████▉ | 58385/65536 [10:04:00<1:17:36,  1.54it/s]
 89%|████████▉ | 58386/65536 [10:04:00<1:14:58,  1.59it/s]
 89%|████████▉ | 58387/65536 [10:04:01<1:14:10,  1.61it/s]
 89%|████████▉ | 58388/65536 [10:04:02<1:16:13,  1.56it/s]
 89%|████████▉ | 58389/65536 [10:04:02<1:14:48,  1.59it/s]
 89%|████████▉ | 58390/65536 [10:04:03<1:15:01,  1.59it/s]
 89%|████████▉ | 58391/65536 [10:04:04<1:13:02,  1.63it/s]
 89%|████████▉ | 58392/65536 [10:04:04<1:12:58,  1.63it/s]
 89%|████████▉ | 58393/65536 [10:04:05<1:10:39,  1.68it/s]
 89%|████████▉ | 58394/65536 [10:04:05<1:10:25,  1.69it/s]
 89%|████████▉ | 58395/65536 [10:04:06<1:12:16,  1.65it/s]
 89%|████████▉ | 58396/65536 [10:04:07<1:12:01,  1.65it/s]
 89%|████████▉ | 58397/65536 [10:04:07<1:12:54,  1.63it/s]
 89%|████████▉ | 58398/65536 [10:04:08<1:11:49,  1.66it/s]
 89%|████████▉ | 58399/65536 [10:04:08<1:13:06,  1.63it/s]
 89%|████████▉ | 58400/65536 [10:04:09<1:16:11,  1.56it/s]
                                                          
{'loss': 1.5598, 'learning_rate': 1.9829802865189174e-07, 'epoch': 3604.94}
+
 89%|████████▉ | 58400/65536 [10:04:09<1:16:11,  1.56it/s]
 89%|████████▉ | 58401/65536 [10:04:10<1:14:57,  1.59it/s]
 89%|████████▉ | 58402/65536 [10:04:10<1:18:07,  1.52it/s]
 89%|████████▉ | 58403/65536 [10:04:11<1:14:50,  1.59it/s]
 89%|████████▉ | 58404/65536 [10:04:12<1:14:04,  1.60it/s]
 89%|████████▉ | 58405/65536 [10:04:12<1:15:02,  1.58it/s]
 89%|████████▉ | 58406/65536 [10:04:13<1:13:27,  1.62it/s]
 89%|████████▉ | 58407/65536 [10:04:13<1:13:12,  1.62it/s]
 89%|████████▉ | 58408/65536 [10:04:14<1:11:47,  1.65it/s]
 89%|████████▉ | 58409/65536 [10:04:15<1:12:45,  1.63it/s]
 89%|████████▉ | 58410/65536 [10:04:15<1:13:08,  1.62it/s]
 89%|████████▉ | 58411/65536 [10:04:16<1:14:27,  1.59it/s]
 89%|████████▉ | 58412/65536 [10:04:17<1:14:46,  1.59it/s]
 89%|████████▉ | 58413/65536 [10:04:17<1:14:15,  1.60it/s]
 89%|████████▉ | 58414/65536 [10:04:18<1:14:53,  1.58it/s]
 89%|████████▉ | 58415/65536 [10:04:18<1:14:24,  1.59it/s]
 89%|████████▉ | 58416/65536 [10:04:19<1:15:29,  1.57it/s]
 89%|████████▉ | 58417/65536 [10:04:20<1:14:05,  1.60it/s]
 89%|████████▉ | 58418/65536 [10:04:20<1:15:47,  1.57it/s]
 89%|████████▉ | 58419/65536 [10:04:21<1:15:41,  1.57it/s]
 89%|████████▉ | 58420/65536 [10:04:22<1:14:52,  1.58it/s]
                                                          
{'loss': 1.5326, 'learning_rate': 1.980225296926656e-07, 'epoch': 3606.17}
+
 89%|████████▉ | 58420/65536 [10:04:22<1:14:52,  1.58it/s]
 89%|████████▉ | 58421/65536 [10:04:22<1:14:18,  1.60it/s]
 89%|████████▉ | 58422/65536 [10:04:23<1:12:23,  1.64it/s]
 89%|████████▉ | 58423/65536 [10:04:23<1:12:42,  1.63it/s]
 89%|████████▉ | 58424/65536 [10:04:24<1:11:34,  1.66it/s]
 89%|████████▉ | 58425/65536 [10:04:25<1:10:17,  1.69it/s]
 89%|████████▉ | 58426/65536 [10:04:25<1:11:07,  1.67it/s]
 89%|████████▉ | 58427/65536 [10:04:26<1:11:40,  1.65it/s]
 89%|████████▉ | 58428/65536 [10:04:26<1:12:52,  1.63it/s]
 89%|████████▉ | 58429/65536 [10:04:27<1:11:54,  1.65it/s]
 89%|████████▉ | 58430/65536 [10:04:28<1:13:40,  1.61it/s]
 89%|████████▉ | 58431/65536 [10:04:28<1:14:44,  1.58it/s]
 89%|████████▉ | 58432/65536 [10:04:29<1:13:34,  1.61it/s]
 89%|████████▉ | 58433/65536 [10:04:30<1:13:00,  1.62it/s]
 89%|████████▉ | 58434/65536 [10:04:30<1:16:29,  1.55it/s]
 89%|████████▉ | 58435/65536 [10:04:31<1:15:46,  1.56it/s]
 89%|████████▉ | 58436/65536 [10:04:32<1:15:18,  1.57it/s]
 89%|████████▉ | 58437/65536 [10:04:32<1:13:13,  1.62it/s]
 89%|████████▉ | 58438/65536 [10:04:33<1:12:31,  1.63it/s]
 89%|████████▉ | 58439/65536 [10:04:33<1:11:35,  1.65it/s]
 89%|████████▉ | 58440/65536 [10:04:34<1:12:56,  1.62it/s]
                                                          
{'loss': 1.5338, 'learning_rate': 1.9774703073343946e-07, 'epoch': 3607.41}
+
 89%|████████▉ | 58440/65536 [10:04:34<1:12:56,  1.62it/s]
 89%|████████▉ | 58441/65536 [10:04:35<1:11:35,  1.65it/s]
 89%|████████▉ | 58442/65536 [10:04:35<1:15:01,  1.58it/s]
 89%|████████▉ | 58443/65536 [10:04:36<1:14:12,  1.59it/s]
 89%|████████▉ | 58444/65536 [10:04:36<1:12:09,  1.64it/s]
 89%|████████▉ | 58445/65536 [10:04:37<1:11:20,  1.66it/s]
 89%|████████▉ | 58446/65536 [10:04:38<1:13:32,  1.61it/s]
 89%|████████▉ | 58447/65536 [10:04:38<1:12:10,  1.64it/s]
 89%|████████▉ | 58448/65536 [10:04:39<1:12:15,  1.63it/s]
 89%|████████▉ | 58449/65536 [10:04:40<1:15:04,  1.57it/s]
 89%|████████▉ | 58450/65536 [10:04:40<1:14:40,  1.58it/s]
 89%|████████▉ | 58451/65536 [10:04:41<1:14:51,  1.58it/s]
 89%|████████▉ | 58452/65536 [10:04:41<1:14:32,  1.58it/s]
 89%|████████▉ | 58453/65536 [10:04:42<1:14:16,  1.59it/s]
 89%|████████▉ | 58454/65536 [10:04:43<1:12:19,  1.63it/s]
 89%|████████▉ | 58455/65536 [10:04:43<1:12:38,  1.62it/s]
 89%|████████▉ | 58456/65536 [10:04:44<1:12:40,  1.62it/s]
 89%|████████▉ | 58457/65536 [10:04:44<1:11:05,  1.66it/s]
 89%|████████▉ | 58458/65536 [10:04:45<1:10:24,  1.68it/s]
 89%|████████▉ | 58459/65536 [10:04:46<1:11:08,  1.66it/s]
 89%|████████▉ | 58460/65536 [10:04:46<1:11:36,  1.65it/s]
                                                          
{'loss': 1.5881, 'learning_rate': 1.9747153177421334e-07, 'epoch': 3608.64}
+
 89%|████████▉ | 58460/65536 [10:04:46<1:11:36,  1.65it/s]
 89%|████████▉ | 58461/65536 [10:04:47<1:13:04,  1.61it/s]
 89%|████████▉ | 58462/65536 [10:04:48<1:14:34,  1.58it/s]
 89%|████████▉ | 58463/65536 [10:04:48<1:14:50,  1.58it/s]
 89%|████████▉ | 58464/65536 [10:04:49<1:14:22,  1.58it/s]
 89%|████████▉ | 58465/65536 [10:04:50<1:15:55,  1.55it/s]
 89%|████████▉ | 58466/65536 [10:04:50<1:18:05,  1.51it/s]
 89%|████████▉ | 58467/65536 [10:04:51<1:17:13,  1.53it/s]
 89%|████████▉ | 58468/65536 [10:04:51<1:16:48,  1.53it/s]
 89%|████████▉ | 58469/65536 [10:04:52<1:16:12,  1.55it/s]
 89%|████████▉ | 58470/65536 [10:04:53<1:16:13,  1.55it/s]
 89%|████████▉ | 58471/65536 [10:04:53<1:15:30,  1.56it/s]
 89%|████████▉ | 58472/65536 [10:04:54<1:14:04,  1.59it/s]
 89%|████████▉ | 58473/65536 [10:04:55<1:12:06,  1.63it/s]
 89%|████████▉ | 58474/65536 [10:04:55<1:13:10,  1.61it/s]
 89%|████████▉ | 58475/65536 [10:04:56<1:12:36,  1.62it/s]
 89%|████████▉ | 58476/65536 [10:04:56<1:13:08,  1.61it/s]
 89%|████████▉ | 58477/65536 [10:04:57<1:12:15,  1.63it/s]
 89%|████████▉ | 58478/65536 [10:04:58<1:13:24,  1.60it/s]
 89%|████████▉ | 58479/65536 [10:04:58<1:12:56,  1.61it/s]
 89%|���███████▉ | 58480/65536 [10:04:59<1:13:07,  1.61it/s]
                                                          
{'loss': 1.5337, 'learning_rate': 1.9719603281498717e-07, 'epoch': 3609.88}
+
 89%|████████▉ | 58480/65536 [10:04:59<1:13:07,  1.61it/s]
 89%|████████▉ | 58481/65536 [10:05:00<1:12:34,  1.62it/s]
 89%|████████▉ | 58482/65536 [10:05:00<1:11:13,  1.65it/s]
 89%|████████▉ | 58483/65536 [10:05:01<1:14:45,  1.57it/s]
 89%|████████▉ | 58484/65536 [10:05:01<1:14:21,  1.58it/s]
 89%|████████▉ | 58485/65536 [10:05:02<1:12:19,  1.62it/s]
 89%|████████▉ | 58486/65536 [10:05:03<1:11:58,  1.63it/s]
 89%|████████▉ | 58487/65536 [10:05:03<1:11:54,  1.63it/s]
 89%|████████▉ | 58488/65536 [10:05:04<1:11:33,  1.64it/s]
 89%|████████▉ | 58489/65536 [10:05:05<1:17:14,  1.52it/s]
 89%|████████▉ | 58490/65536 [10:05:05<1:14:16,  1.58it/s]
 89%|████████▉ | 58491/65536 [10:05:06<1:13:41,  1.59it/s]
 89%|████████▉ | 58492/65536 [10:05:06<1:12:48,  1.61it/s]
 89%|████████▉ | 58493/65536 [10:05:07<1:15:57,  1.55it/s]
 89%|████████▉ | 58494/65536 [10:05:08<1:15:28,  1.55it/s]
 89%|████████▉ | 58495/65536 [10:05:08<1:14:14,  1.58it/s]
 89%|████████▉ | 58496/65536 [10:05:09<1:13:11,  1.60it/s]
 89%|████████▉ | 58497/65536 [10:05:10<1:14:45,  1.57it/s]
 89%|████████▉ | 58498/65536 [10:05:10<1:13:03,  1.61it/s]
 89%|████████▉ | 58499/65536 [10:05:11<1:16:46,  1.53it/s]
 89%|████████▉ | 58500/65536 [10:05:12<1:17:32,  1.51it/s]
                                                          
{'loss': 1.5572, 'learning_rate': 1.9692053385576095e-07, 'epoch': 3611.11}
+
 89%|████████▉ | 58500/65536 [10:05:12<1:17:32,  1.51it/s]
 89%|████████▉ | 58501/65536 [10:05:12<1:16:24,  1.53it/s]
 89%|████████▉ | 58502/65536 [10:05:13<1:14:18,  1.58it/s]
 89%|████████▉ | 58503/65536 [10:05:13<1:14:05,  1.58it/s]
 89%|████████▉ | 58504/65536 [10:05:14<1:14:18,  1.58it/s]
 89%|████████▉ | 58505/65536 [10:05:15<1:13:26,  1.60it/s]
 89%|████████▉ | 58506/65536 [10:05:15<1:13:44,  1.59it/s]
 89%|████████▉ | 58507/65536 [10:05:16<1:13:09,  1.60it/s]
 89%|████████▉ | 58508/65536 [10:05:17<1:11:46,  1.63it/s]
 89%|████████▉ | 58509/65536 [10:05:17<1:12:11,  1.62it/s]
 89%|████████▉ | 58510/65536 [10:05:18<1:10:19,  1.67it/s]
 89%|████████▉ | 58511/65536 [10:05:18<1:10:43,  1.66it/s]
 89%|████████▉ | 58512/65536 [10:05:19<1:09:51,  1.68it/s]
 89%|████████▉ | 58513/65536 [10:05:20<1:09:02,  1.70it/s]
 89%|████████▉ | 58514/65536 [10:05:20<1:10:10,  1.67it/s]
 89%|████████▉ | 58515/65536 [10:05:21<1:14:17,  1.57it/s]
 89%|████████▉ | 58516/65536 [10:05:21<1:13:46,  1.59it/s]
 89%|████████▉ | 58517/65536 [10:05:22<1:12:00,  1.62it/s]
 89%|████████▉ | 58518/65536 [10:05:23<1:10:36,  1.66it/s]
 89%|████████▉ | 58519/65536 [10:05:23<1:11:35,  1.63it/s]
 89%|████████▉ | 58520/65536 [10:05:24<1:13:23,  1.59it/s]
                                                          
{'loss': 1.5736, 'learning_rate': 1.966450348965348e-07, 'epoch': 3612.35}
+
 89%|████████▉ | 58520/65536 [10:05:24<1:13:23,  1.59it/s]
 89%|████████▉ | 58521/65536 [10:05:25<1:16:57,  1.52it/s]
 89%|████████▉ | 58522/65536 [10:05:25<1:17:21,  1.51it/s]
 89%|████████▉ | 58523/65536 [10:05:26<1:16:15,  1.53it/s]
 89%|████████▉ | 58524/65536 [10:05:27<1:15:22,  1.55it/s]
 89%|████████▉ | 58525/65536 [10:05:27<1:13:45,  1.58it/s]
 89%|████████▉ | 58526/65536 [10:05:28<1:12:09,  1.62it/s]
 89%|████████▉ | 58527/65536 [10:05:28<1:11:17,  1.64it/s]
 89%|████████▉ | 58528/65536 [10:05:29<1:11:28,  1.63it/s]
 89%|████████▉ | 58529/65536 [10:05:30<1:10:49,  1.65it/s]
 89%|████████▉ | 58530/65536 [10:05:30<1:11:50,  1.63it/s]
 89%|████████▉ | 58531/65536 [10:05:31<1:14:46,  1.56it/s]
 89%|████████▉ | 58532/65536 [10:05:32<1:13:05,  1.60it/s]
 89%|████████▉ | 58533/65536 [10:05:32<1:14:38,  1.56it/s]
 89%|████████▉ | 58534/65536 [10:05:33<1:13:05,  1.60it/s]
 89%|████████▉ | 58535/65536 [10:05:33<1:11:24,  1.63it/s]
 89%|████████▉ | 58536/65536 [10:05:34<1:10:46,  1.65it/s]
 89%|████████▉ | 58537/65536 [10:05:35<1:11:16,  1.64it/s]
 89%|████��███▉ | 58538/65536 [10:05:35<1:09:43,  1.67it/s]
 89%|████████▉ | 58539/65536 [10:05:36<1:10:40,  1.65it/s]
 89%|████████▉ | 58540/65536 [10:05:36<1:11:09,  1.64it/s]
                                                          
{'loss': 1.5843, 'learning_rate': 1.9636953593730867e-07, 'epoch': 3613.58}
+
 89%|████████▉ | 58540/65536 [10:05:36<1:11:09,  1.64it/s]
 89%|████████▉ | 58541/65536 [10:05:37<1:10:37,  1.65it/s]
 89%|████████▉ | 58542/65536 [10:05:38<1:10:08,  1.66it/s]
 89%|████████▉ | 58543/65536 [10:05:38<1:10:35,  1.65it/s]
 89%|████████▉ | 58544/65536 [10:05:39<1:11:49,  1.62it/s]
 89%|████████▉ | 58545/65536 [10:05:39<1:13:32,  1.58it/s]
 89%|████████▉ | 58546/65536 [10:05:40<1:12:02,  1.62it/s]
 89%|████████▉ | 58547/65536 [10:05:41<1:15:13,  1.55it/s]
 89%|████████▉ | 58548/65536 [10:05:41<1:16:40,  1.52it/s]
 89%|████████▉ | 58549/65536 [10:05:42<1:15:24,  1.54it/s]
 89%|████████▉ | 58550/65536 [10:05:43<1:13:50,  1.58it/s]
 89%|████████▉ | 58551/65536 [10:05:43<1:12:43,  1.60it/s]
 89%|████████▉ | 58552/65536 [10:05:44<1:11:32,  1.63it/s]
 89%|████████▉ | 58553/65536 [10:05:45<1:11:48,  1.62it/s]
 89%|████████▉ | 58554/65536 [10:05:45<1:11:21,  1.63it/s]
 89%|████████▉ | 58555/65536 [10:05:46<1:12:17,  1.61it/s]
 89%|████████▉ | 58556/65536 [10:05:46<1:12:20,  1.61it/s]
 89%|████████▉ | 58557/65536 [10:05:47<1:13:35,  1.58it/s]
 89%|████████▉ | 58558/65536 [10:05:48<1:12:14,  1.61it/s]
 89%|████████▉ | 58559/65536 [10:05:48<1:11:31,  1.63it/s]
 89%|████████▉ | 58560/65536 [10:05:49<1:11:57,  1.62it/s]
                                                          
{'loss': 1.5126, 'learning_rate': 1.9609403697808253e-07, 'epoch': 3614.81}
+
 89%|████████▉ | 58560/65536 [10:05:49<1:11:57,  1.62it/s]
 89%|████████▉ | 58561/65536 [10:05:49<1:11:55,  1.62it/s]
 89%|████████▉ | 58562/65536 [10:05:50<1:10:19,  1.65it/s]
 89%|████████▉ | 58563/65536 [10:05:51<1:11:36,  1.62it/s]
 89%|████████▉ | 58564/65536 [10:05:51<1:13:30,  1.58it/s]
 89%|████████▉ | 58565/65536 [10:05:52<1:11:21,  1.63it/s]
 89%|████████▉ | 58566/65536 [10:05:53<1:11:49,  1.62it/s]
 89%|████████▉ | 58567/65536 [10:05:53<1:14:15,  1.56it/s]
 89%|████████▉ | 58568/65536 [10:05:54<1:12:05,  1.61it/s]
 89%|████████▉ | 58569/65536 [10:05:54<1:12:02,  1.61it/s]
 89%|████████▉ | 58570/65536 [10:05:55<1:12:56,  1.59it/s]
 89%|████████▉ | 58571/65536 [10:05:56<1:14:29,  1.56it/s]
 89%|████████▉ | 58572/65536 [10:05:56<1:13:52,  1.57it/s]
 89%|████████▉ | 58573/65536 [10:05:57<1:13:35,  1.58it/s]
 89%|████████▉ | 58574/65536 [10:05:58<1:13:12,  1.58it/s]
 89%|████████▉ | 58575/65536 [10:05:58<1:12:54,  1.59it/s]
 89%|████████▉ | 58576/65536 [10:05:59<1:13:38,  1.58it/s]
 89%|████████▉ | 58577/65536 [10:06:00<1:14:52,  1.55it/s]
 89%|████████▉ | 58578/65536 [10:06:00<1:13:06,  1.59it/s]
 89%|████████▉ | 58579/65536 [10:06:01<1:11:50,  1.61it/s]
 89%|████████▉ | 58580/65536 [10:06:01<1:13:57,  1.57it/s]
                                                          
{'loss': 1.566, 'learning_rate': 1.958185380188564e-07, 'epoch': 3616.05}
+
 89%|████████▉ | 58580/65536 [10:06:01<1:13:57,  1.57it/s]
 89%|████████▉ | 58581/65536 [10:06:02<1:13:40,  1.57it/s]
 89%|████████▉ | 58582/65536 [10:06:03<1:13:12,  1.58it/s]
 89%|████████▉ | 58583/65536 [10:06:03<1:11:23,  1.62it/s]
 89%|████████▉ | 58584/65536 [10:06:04<1:13:00,  1.59it/s]
 89%|████████▉ | 58585/65536 [10:06:05<1:14:26,  1.56it/s]
 89%|████████▉ | 58586/65536 [10:06:05<1:12:39,  1.59it/s]
 89%|████████▉ | 58587/65536 [10:06:06<1:13:05,  1.58it/s]
 89%|████████▉ | 58588/65536 [10:06:06<1:12:45,  1.59it/s]
 89%|████████▉ | 58589/65536 [10:06:07<1:14:10,  1.56it/s]
 89%|████████▉ | 58590/65536 [10:06:08<1:12:46,  1.59it/s]
 89%|████████▉ | 58591/65536 [10:06:08<1:11:48,  1.61it/s]
 89%|████████▉ | 58592/65536 [10:06:09<1:11:15,  1.62it/s]
 89%|████████▉ | 58593/65536 [10:06:10<1:11:06,  1.63it/s]
 89%|████████▉ | 58594/65536 [10:06:10<1:11:05,  1.63it/s]
 89%|████████▉ | 58595/65536 [10:06:11<1:09:17,  1.67it/s]
 89%|████████▉ | 58596/65536 [10:06:11<1:11:45,  1.61it/s]
 89%|████████▉ | 58597/65536 [10:06:12<1:11:41,  1.61it/s]
 89%|████████▉ | 58598/65536 [10:06:13<1:11:02,  1.63it/s]
 89%|████████▉ | 58599/65536 [10:06:13<1:10:31,  1.64it/s]
 89%|████████▉ | 58600/65536 [10:06:14<1:11:05,  1.63it/s]
                                                          
{'loss': 1.5558, 'learning_rate': 1.9554303905963025e-07, 'epoch': 3617.28}
+
 89%|████████▉ | 58600/65536 [10:06:14<1:11:05,  1.63it/s]
 89%|████████▉ | 58601/65536 [10:06:14<1:10:41,  1.64it/s]
 89%|████████▉ | 58602/65536 [10:06:15<1:10:55,  1.63it/s]
 89%|████████▉ | 58603/65536 [10:06:16<1:10:51,  1.63it/s]
 89%|████████▉ | 58604/65536 [10:06:16<1:13:17,  1.58it/s]
 89%|████████▉ | 58605/65536 [10:06:17<1:13:25,  1.57it/s]
 89%|████████▉ | 58606/65536 [10:06:18<1:13:37,  1.57it/s]
 89%|████████▉ | 58607/65536 [10:06:18<1:11:33,  1.61it/s]
 89%|████████▉ | 58608/65536 [10:06:19<1:11:38,  1.61it/s]
 89%|████████▉ | 58609/65536 [10:06:19<1:10:51,  1.63it/s]
 89%|████████▉ | 58610/65536 [10:06:20<1:12:09,  1.60it/s]
 89%|████████▉ | 58611/65536 [10:06:21<1:10:47,  1.63it/s]
 89%|████████▉ | 58612/65536 [10:06:21<1:13:15,  1.58it/s]
 89%|████████▉ | 58613/65536 [10:06:22<1:14:38,  1.55it/s]
 89%|████████▉ | 58614/65536 [10:06:23<1:16:01,  1.52it/s]
 89%|████████▉ | 58615/65536 [10:06:23<1:16:35,  1.51it/s]
 89%|████████▉ | 58616/65536 [10:06:24<1:14:34,  1.55it/s]
 89%|████████▉ | 58617/65536 [10:06:25<1:13:16,  1.57it/s]
 89%|████████▉ | 58618/65536 [10:06:25<1:12:00,  1.60it/s]
 89%|████████▉ | 58619/65536 [10:06:26<1:10:15,  1.64it/s]
 89%|████████▉ | 58620/65536 [10:06:26<1:11:53,  1.60it/s]
                                                          
{'loss': 1.529, 'learning_rate': 1.952675401004041e-07, 'epoch': 3618.52}
+
 89%|████████▉ | 58620/65536 [10:06:26<1:11:53,  1.60it/s]
 89%|████████▉ | 58621/65536 [10:06:27<1:12:35,  1.59it/s]
 89%|████████▉ | 58622/65536 [10:06:28<1:11:39,  1.61it/s]
 89%|████████▉ | 58623/65536 [10:06:28<1:12:34,  1.59it/s]
 89%|████████▉ | 58624/65536 [10:06:29<1:10:22,  1.64it/s]
 89%|████████▉ | 58625/65536 [10:06:30<1:11:50,  1.60it/s]
 89%|████████▉ | 58626/65536 [10:06:30<1:11:08,  1.62it/s]
 89%|████████▉ | 58627/65536 [10:06:31<1:10:09,  1.64it/s]
 89%|████████▉ | 58628/65536 [10:06:31<1:12:28,  1.59it/s]
 89%|████████▉ | 58629/65536 [10:06:32<1:10:53,  1.62it/s]
 89%|████████▉ | 58630/65536 [10:06:33<1:10:22,  1.64it/s]
 89%|████████▉ | 58631/65536 [10:06:33<1:10:42,  1.63it/s]
 89%|████████▉ | 58632/65536 [10:06:34<1:10:54,  1.62it/s]
 89%|████████▉ | 58633/65536 [10:06:35<1:11:39,  1.61it/s]
 89%|████████▉ | 58634/65536 [10:06:35<1:11:14,  1.61it/s]
 89%|████████▉ | 58635/65536 [10:06:36<1:10:38,  1.63it/s]
 89%|████████▉ | 58636/65536 [10:06:36<1:10:05,  1.64it/s]
 89%|████████▉ | 58637/65536 [10:06:37<1:09:46,  1.65it/s]
 89%|████████▉ | 58638/65536 [10:06:38<1:11:54,  1.60it/s]
 89%|████████▉ | 58639/65536 [10:06:38<1:12:17,  1.59it/s]
 89%|████████▉ | 58640/65536 [10:06:39<1:12:30,  1.58it/s]
                                                          
{'loss': 1.5571, 'learning_rate': 1.9499204114117786e-07, 'epoch': 3619.75}
+
 89%|████████▉ | 58640/65536 [10:06:39<1:12:30,  1.58it/s]
 89%|████████▉ | 58641/65536 [10:06:39<1:12:19,  1.59it/s]
 89%|████████▉ | 58642/65536 [10:06:40<1:10:51,  1.62it/s]
 89%|████████▉ | 58643/65536 [10:06:41<1:09:50,  1.65it/s]
 89%|████████▉ | 58644/65536 [10:06:41<1:11:23,  1.61it/s]
 89%|████████▉ | 58645/65536 [10:06:42<1:13:10,  1.57it/s]
 89%|████████▉ | 58646/65536 [10:06:43<1:11:35,  1.60it/s]
 89%|████████▉ | 58647/65536 [10:06:43<1:11:19,  1.61it/s]
 89%|████████▉ | 58648/65536 [10:06:44<1:10:02,  1.64it/s]
 89%|████████▉ | 58649/65536 [10:06:44<1:09:36,  1.65it/s]
 89%|████████▉ | 58650/65536 [10:06:45<1:08:06,  1.69it/s]
 89%|████████▉ | 58651/65536 [10:06:46<1:08:12,  1.68it/s]
 89%|████████▉ | 58652/65536 [10:06:46<1:07:40,  1.70it/s]
 89%|████████▉ | 58653/65536 [10:06:47<1:10:29,  1.63it/s]
 89%|████████▉ | 58654/65536 [10:06:47<1:11:18,  1.61it/s]
 90%|████████▉ | 58655/65536 [10:06:48<1:10:23,  1.63it/s]
 90%|████████▉ | 58656/65536 [10:06:49<1:10:41,  1.62it/s]
 90%|████████▉ | 58657/65536 [10:06:49<1:13:22,  1.56it/s]
 90%|████████▉ | 58658/65536 [10:06:50<1:12:11,  1.59it/s]
 90%|████████▉ | 58659/65536 [10:06:51<1:12:33,  1.58it/s]
 90%|████████▉ | 58660/65536 [10:06:51<1:10:49,  1.62it/s]
                                                          
{'loss': 1.582, 'learning_rate': 1.9471654218195172e-07, 'epoch': 3620.99}
+
 90%|████████▉ | 58660/65536 [10:06:51<1:10:49,  1.62it/s]
 90%|████████▉ | 58661/65536 [10:06:52<1:12:59,  1.57it/s]
 90%|████████▉ | 58662/65536 [10:06:53<1:14:56,  1.53it/s]
 90%|████████▉ | 58663/65536 [10:06:53<1:12:55,  1.57it/s]
 90%|████████▉ | 58664/65536 [10:06:54<1:13:05,  1.57it/s]
 90%|████████▉ | 58665/65536 [10:06:54<1:10:15,  1.63it/s]
 90%|████████▉ | 58666/65536 [10:06:55<1:10:06,  1.63it/s]
 90%|████████▉ | 58667/65536 [10:06:56<1:12:10,  1.59it/s]
 90%|████████▉ | 58668/65536 [10:06:56<1:10:59,  1.61it/s]
 90%|████████▉ | 58669/65536 [10:06:57<1:10:59,  1.61it/s]
 90%|████████▉ | 58670/65536 [10:06:57<1:11:02,  1.61it/s]
 90%|████████▉ | 58671/65536 [10:06:58<1:10:05,  1.63it/s]
 90%|████████▉ | 58672/65536 [10:06:59<1:09:27,  1.65it/s]
 90%|████████▉ | 58673/65536 [10:06:59<1:08:59,  1.66it/s]
 90%|████████▉ | 58674/65536 [10:07:00<1:11:05,  1.61it/s]
 90%|████████▉ | 58675/65536 [10:07:01<1:10:54,  1.61it/s]
 90%|████████▉ | 58676/65536 [10:07:01<1:09:02,  1.66it/s]
 90%|████████▉ | 58677/65536 [10:07:02<1:11:42,  1.59it/s]
 90%|████████▉ | 58678/65536 [10:07:02<1:10:05,  1.63it/s]
 90%|████████▉ | 58679/65536 [10:07:03<1:10:28,  1.62it/s]
 90%|████████▉ | 58680/65536 [10:07:04<1:09:25,  1.65it/s]
                                                          
{'loss': 1.5444, 'learning_rate': 1.9444104322272558e-07, 'epoch': 3622.22}
+
 90%|████████▉ | 58680/65536 [10:07:04<1:09:25,  1.65it/s]
 90%|████████▉ | 58681/65536 [10:07:04<1:08:58,  1.66it/s]
 90%|████████▉ | 58682/65536 [10:07:05<1:08:29,  1.67it/s]
 90%|████████▉ | 58683/65536 [10:07:05<1:08:33,  1.67it/s]
 90%|████████▉ | 58684/65536 [10:07:06<1:08:38,  1.66it/s]
 90%|████████▉ | 58685/65536 [10:07:07<1:08:20,  1.67it/s]
 90%|████████▉ | 58686/65536 [10:07:07<1:08:42,  1.66it/s]
 90%|████████▉ | 58687/65536 [10:07:08<1:11:43,  1.59it/s]
 90%|████████▉ | 58688/65536 [10:07:08<1:11:49,  1.59it/s]
 90%|████████▉ | 58689/65536 [10:07:09<1:11:29,  1.60it/s]
 90%|████████▉ | 58690/65536 [10:07:10<1:11:17,  1.60it/s]
 90%|████████▉ | 58691/65536 [10:07:10<1:10:46,  1.61it/s]
 90%|████████▉ | 58692/65536 [10:07:11<1:10:46,  1.61it/s]
 90%|████████▉ | 58693/65536 [10:07:12<1:15:11,  1.52it/s]
 90%|████████▉ | 58694/65536 [10:07:12<1:13:36,  1.55it/s]
 90%|████████▉ | 58695/65536 [10:07:13<1:14:20,  1.53it/s]
 90%|████████▉ | 58696/65536 [10:07:14<1:13:11,  1.56it/s]
 90%|████████▉ | 58697/65536 [10:07:14<1:14:06,  1.54it/s]
 90%|████████▉ | 58698/65536 [10:07:15<1:11:09,  1.60it/s]
 90%|████████▉ | 58699/65536 [10:07:15<1:11:22,  1.60it/s]
 90%|████████▉ | 58700/65536 [10:07:16<1:11:43,  1.59it/s]
                                                          
{'loss': 1.5387, 'learning_rate': 1.9416554426349944e-07, 'epoch': 3623.46}
+
 90%|████████▉ | 58700/65536 [10:07:16<1:11:43,  1.59it/s]
 90%|████████▉ | 58701/65536 [10:07:17<1:13:27,  1.55it/s]
 90%|████████▉ | 58702/65536 [10:07:17<1:13:20,  1.55it/s]
 90%|████████▉ | 58703/65536 [10:07:18<1:11:50,  1.59it/s]
 90%|████████▉ | 58704/65536 [10:07:19<1:11:03,  1.60it/s]
 90%|████████▉ | 58705/65536 [10:07:19<1:08:52,  1.65it/s]
 90%|████████▉ | 58706/65536 [10:07:20<1:10:02,  1.63it/s]
 90%|████████▉ | 58707/65536 [10:07:20<1:09:55,  1.63it/s]
 90%|████████▉ | 58708/65536 [10:07:21<1:08:40,  1.66it/s]
 90%|████████▉ | 58709/65536 [10:07:22<1:10:37,  1.61it/s]
 90%|████████▉ | 58710/65536 [10:07:22<1:09:30,  1.64it/s]
 90%|████████▉ | 58711/65536 [10:07:23<1:09:49,  1.63it/s]
 90%|████████▉ | 58712/65536 [10:07:24<1:10:02,  1.62it/s]
 90%|████████▉ | 58713/65536 [10:07:24<1:09:31,  1.64it/s]
 90%|████████▉ | 58714/65536 [10:07:25<1:09:11,  1.64it/s]
 90%|████████▉ | 58715/65536 [10:07:25<1:10:02,  1.62it/s]
 90%|████████▉ | 58716/65536 [10:07:26<1:10:36,  1.61it/s]
 90%|████████▉ | 58717/65536 [10:07:27<1:12:47,  1.56it/s]
 90%|████████▉ | 58718/65536 [10:07:27<1:11:46,  1.58it/s]
 90%|████████▉ | 58719/65536 [10:07:28<1:10:14,  1.62it/s]
 90%|████████▉ | 58720/65536 [10:07:29<1:11:20,  1.59it/s]
                                                          
{'loss': 1.5513, 'learning_rate': 1.938900453042733e-07, 'epoch': 3624.69}
+
 90%|████████▉ | 58720/65536 [10:07:29<1:11:20,  1.59it/s]
 90%|████████▉ | 58721/65536 [10:07:29<1:10:47,  1.60it/s]
 90%|████████▉ | 58722/65536 [10:07:30<1:11:02,  1.60it/s]
 90%|████████▉ | 58723/65536 [10:07:30<1:11:46,  1.58it/s]
 90%|████████▉ | 58724/65536 [10:07:31<1:11:44,  1.58it/s]
 90%|████████▉ | 58725/65536 [10:07:32<1:10:14,  1.62it/s]
 90%|████████▉ | 58726/65536 [10:07:32<1:12:29,  1.57it/s]
 90%|████████▉ | 58727/65536 [10:07:33<1:11:20,  1.59it/s]
 90%|████████▉ | 58728/65536 [10:07:34<1:12:09,  1.57it/s]
 90%|████████▉ | 58729/65536 [10:07:34<1:10:39,  1.61it/s]
 90%|████████▉ | 58730/65536 [10:07:35<1:10:49,  1.60it/s]
 90%|████████▉ | 58731/65536 [10:07:35<1:10:01,  1.62it/s]
 90%|████████▉ | 58732/65536 [10:07:36<1:10:44,  1.60it/s]
 90%|████████▉ | 58733/65536 [10:07:37<1:11:13,  1.59it/s]
 90%|████████▉ | 58734/65536 [10:07:37<1:10:12,  1.61it/s]
 90%|████████▉ | 58735/65536 [10:07:38<1:10:04,  1.62it/s]
 90%|████████▉ | 58736/65536 [10:07:38<1:09:21,  1.63it/s]
 90%|████████▉ | 58737/65536 [10:07:39<1:08:15,  1.66it/s]
 90%|████████▉ | 58738/65536 [10:07:40<1:11:03,  1.59it/s]
 90%|████████▉ | 58739/65536 [10:07:40<1:10:57,  1.60it/s]
 90%|████████▉ | 58740/65536 [10:07:41<1:10:29,  1.61it/s]
                                                          
{'loss': 1.512, 'learning_rate': 1.9361454634504715e-07, 'epoch': 3625.93}
+
 90%|████████▉ | 58740/65536 [10:07:41<1:10:29,  1.61it/s]
 90%|████████▉ | 58741/65536 [10:07:42<1:09:23,  1.63it/s]
 90%|████████▉ | 58742/65536 [10:07:42<1:10:38,  1.60it/s]
 90%|████████▉ | 58743/65536 [10:07:43<1:09:37,  1.63it/s]
 90%|████████▉ | 58744/65536 [10:07:43<1:10:45,  1.60it/s]
 90%|████████▉ | 58745/65536 [10:07:44<1:10:43,  1.60it/s]
 90%|████████▉ | 58746/65536 [10:07:45<1:10:40,  1.60it/s]
 90%|████████▉ | 58747/65536 [10:07:45<1:10:45,  1.60it/s]
 90%|████████▉ | 58748/65536 [10:07:46<1:11:08,  1.59it/s]
 90%|████████▉ | 58749/65536 [10:07:47<1:09:18,  1.63it/s]
 90%|████████▉ | 58750/65536 [10:07:47<1:09:02,  1.64it/s]
 90%|████████▉ | 58751/65536 [10:07:48<1:08:57,  1.64it/s]
 90%|████████▉ | 58752/65536 [10:07:48<1:11:37,  1.58it/s]
 90%|████████▉ | 58753/65536 [10:07:49<1:11:48,  1.57it/s]
 90%|████████▉ | 58754/65536 [10:07:50<1:11:42,  1.58it/s]
 90%|████████▉ | 58755/65536 [10:07:50<1:09:33,  1.62it/s]
 90%|████████▉ | 58756/65536 [10:07:51<1:09:22,  1.63it/s]
 90%|████████▉ | 58757/65536 [10:07:52<1:10:20,  1.61it/s]
 90%|████████▉ | 58758/65536 [10:07:52<1:15:51,  1.49it/s]
 90%|████████▉ | 58759/65536 [10:07:53<1:14:44,  1.51it/s]
 90%|████████▉ | 58760/65536 [10:07:54<1:14:12,  1.52it/s]
                                                          
{'loss': 1.5541, 'learning_rate': 1.9333904738582093e-07, 'epoch': 3627.16}
+
 90%|████████▉ | 58760/65536 [10:07:54<1:14:12,  1.52it/s]
 90%|████████▉ | 58761/65536 [10:07:54<1:12:22,  1.56it/s]
 90%|████████▉ | 58762/65536 [10:07:55<1:11:14,  1.58it/s]
 90%|████████▉ | 58763/65536 [10:07:55<1:10:53,  1.59it/s]
 90%|████████▉ | 58764/65536 [10:07:56<1:09:01,  1.64it/s]
 90%|████████▉ | 58765/65536 [10:07:57<1:08:59,  1.64it/s]
 90%|████████▉ | 58766/65536 [10:07:57<1:08:09,  1.66it/s]
 90%|████████▉ | 58767/65536 [10:07:58<1:08:29,  1.65it/s]
 90%|████████▉ | 58768/65536 [10:07:58<1:06:28,  1.70it/s]
 90%|████████▉ | 58769/65536 [10:07:59<1:07:58,  1.66it/s]
 90%|████████▉ | 58770/65536 [10:08:00<1:07:50,  1.66it/s]
 90%|████████▉ | 58771/65536 [10:08:00<1:08:18,  1.65it/s]
 90%|████████▉ | 58772/65536 [10:08:01<1:09:11,  1.63it/s]
 90%|████████▉ | 58773/65536 [10:08:02<1:09:45,  1.62it/s]
 90%|████████▉ | 58774/65536 [10:08:02<1:11:23,  1.58it/s]
 90%|████████▉ | 58775/65536 [10:08:03<1:11:30,  1.58it/s]
 90%|████████▉ | 58776/65536 [10:08:03<1:11:19,  1.58it/s]
 90%|████████▉ | 58777/65536 [10:08:04<1:10:23,  1.60it/s]
 90%|████████▉ | 58778/65536 [10:08:05<1:08:33,  1.64it/s]
 90%|████████▉ | 58779/65536 [10:08:05<1:10:12,  1.60it/s]
 90%|████████▉ | 58780/65536 [10:08:06<1:08:58,  1.63it/s]
                                                          
{'loss': 1.5458, 'learning_rate': 1.930635484265948e-07, 'epoch': 3628.4}
+
 90%|████████▉ | 58780/65536 [10:08:06<1:08:58,  1.63it/s]
 90%|████████▉ | 58781/65536 [10:08:07<1:11:53,  1.57it/s]
 90%|████████▉ | 58782/65536 [10:08:07<1:10:39,  1.59it/s]
 90%|████████▉ | 58783/65536 [10:08:08<1:10:11,  1.60it/s]
 90%|████████▉ | 58784/65536 [10:08:08<1:10:15,  1.60it/s]
 90%|████████▉ | 58785/65536 [10:08:09<1:09:01,  1.63it/s]
 90%|████████▉ | 58786/65536 [10:08:10<1:08:59,  1.63it/s]
 90%|████████▉ | 58787/65536 [10:08:10<1:06:52,  1.68it/s]
 90%|████████▉ | 58788/65536 [10:08:11<1:07:43,  1.66it/s]
 90%|████████▉ | 58789/65536 [10:08:11<1:09:33,  1.62it/s]
 90%|████████▉ | 58790/65536 [10:08:12<1:11:09,  1.58it/s]
 90%|████████▉ | 58791/65536 [10:08:13<1:10:52,  1.59it/s]
 90%|████████▉ | 58792/65536 [10:08:13<1:10:29,  1.59it/s]
 90%|████████▉ | 58793/65536 [10:08:14<1:09:04,  1.63it/s]
 90%|████████▉ | 58794/65536 [10:08:14<1:07:17,  1.67it/s]
 90%|████████▉ | 58795/65536 [10:08:15<1:06:27,  1.69it/s]
 90%|████████▉ | 58796/65536 [10:08:16<1:07:04,  1.67it/s]
 90%|████████▉ | 58797/65536 [10:08:16<1:08:32,  1.64it/s]
 90%|████████▉ | 58798/65536 [10:08:17<1:10:03,  1.60it/s]
 90%|████████▉ | 58799/65536 [10:08:18<1:11:00,  1.58it/s]
 90%|████████▉ | 58800/65536 [10:08:18<1:09:59,  1.60it/s]
                                                          
{'loss': 1.5554, 'learning_rate': 1.9278804946736865e-07, 'epoch': 3629.63}
+
 90%|████████▉ | 58800/65536 [10:08:18<1:09:59,  1.60it/s]
 90%|████████▉ | 58801/65536 [10:08:19<1:08:10,  1.65it/s]
 90%|████████▉ | 58802/65536 [10:08:19<1:08:20,  1.64it/s]
 90%|████████▉ | 58803/65536 [10:08:20<1:08:00,  1.65it/s]
 90%|████████▉ | 58804/65536 [10:08:21<1:09:30,  1.61it/s]
 90%|████████▉ | 58805/65536 [10:08:21<1:09:33,  1.61it/s]
 90%|████████▉ | 58806/65536 [10:08:22<1:10:48,  1.58it/s]
 90%|████████▉ | 58807/65536 [10:08:23<1:12:14,  1.55it/s]
 90%|████████▉ | 58808/65536 [10:08:23<1:11:13,  1.57it/s]
 90%|████████▉ | 58809/65536 [10:08:24<1:12:43,  1.54it/s]
 90%|████████▉ | 58810/65536 [10:08:24<1:09:59,  1.60it/s]
 90%|████████▉ | 58811/65536 [10:08:25<1:09:28,  1.61it/s]
 90%|████████▉ | 58812/65536 [10:08:26<1:10:10,  1.60it/s]
 90%|████████▉ | 58813/65536 [10:08:26<1:08:59,  1.62it/s]
 90%|████████▉ | 58814/65536 [10:08:27<1:08:09,  1.64it/s]
 90%|████████▉ | 58815/65536 [10:08:28<1:09:44,  1.61it/s]
 90%|████████▉ | 58816/65536 [10:08:28<1:08:15,  1.64it/s]
 90%|████████▉ | 58817/65536 [10:08:29<1:07:44,  1.65it/s]
 90%|████████▉ | 58818/65536 [10:08:29<1:08:49,  1.63it/s]
 90%|████████▉ | 58819/65536 [10:08:30<1:09:19,  1.61it/s]
 90%|████████▉ | 58820/65536 [10:08:31<1:11:51,  1.56it/s]
                                                          
{'loss': 1.5525, 'learning_rate': 1.9251255050814248e-07, 'epoch': 3630.86}
+
 90%|████████▉ | 58820/65536 [10:08:31<1:11:51,  1.56it/s]
 90%|████████▉ | 58821/65536 [10:08:31<1:10:50,  1.58it/s]
 90%|████████▉ | 58822/65536 [10:08:32<1:09:49,  1.60it/s]
 90%|████████▉ | 58823/65536 [10:08:33<1:11:16,  1.57it/s]
 90%|████████▉ | 58824/65536 [10:08:33<1:12:13,  1.55it/s]
 90%|████████▉ | 58825/65536 [10:08:34<1:12:17,  1.55it/s]
 90%|████████▉ | 58826/65536 [10:08:35<1:12:48,  1.54it/s]
 90%|████████▉ | 58827/65536 [10:08:35<1:09:47,  1.60it/s]
 90%|████████▉ | 58828/65536 [10:08:36<1:09:00,  1.62it/s]
 90%|████████▉ | 58829/65536 [10:08:36<1:09:03,  1.62it/s]
 90%|████████▉ | 58830/65536 [10:08:37<1:08:38,  1.63it/s]
 90%|████████▉ | 58831/65536 [10:08:38<1:08:28,  1.63it/s]
 90%|████████▉ | 58832/65536 [10:08:38<1:08:04,  1.64it/s]
 90%|████████▉ | 58833/65536 [10:08:39<1:08:01,  1.64it/s]
 90%|████████▉ | 58834/65536 [10:08:39<1:08:29,  1.63it/s]
 90%|████████▉ | 58835/65536 [10:08:40<1:08:38,  1.63it/s]
 90%|████████▉ | 58836/65536 [10:08:41<1:08:33,  1.63it/s]
 90%|████████▉ | 58837/65536 [10:08:41<1:07:42,  1.65it/s]
 90%|████████▉ | 58838/65536 [10:08:42<1:07:41,  1.65it/s]
 90%|████████▉ | 58839/65536 [10:08:42<1:09:43,  1.60it/s]
 90%|████████▉ | 58840/65536 [10:08:43<1:08:59,  1.62it/s]
                                                          
{'loss': 1.5987, 'learning_rate': 1.9223705154891637e-07, 'epoch': 3632.1}
+
 90%|████████▉ | 58840/65536 [10:08:43<1:08:59,  1.62it/s]
 90%|████████▉ | 58841/65536 [10:08:44<1:09:03,  1.62it/s]
 90%|████████▉ | 58842/65536 [10:08:44<1:10:31,  1.58it/s]
 90%|████████▉ | 58843/65536 [10:08:45<1:08:55,  1.62it/s]
 90%|████████▉ | 58844/65536 [10:08:46<1:08:44,  1.62it/s]
 90%|████████▉ | 58845/65536 [10:08:46<1:07:52,  1.64it/s]
 90%|████████▉ | 58846/65536 [10:08:47<1:08:05,  1.64it/s]
 90%|████████▉ | 58847/65536 [10:08:47<1:07:00,  1.66it/s]
 90%|████████▉ | 58848/65536 [10:08:48<1:06:50,  1.67it/s]
 90%|████████▉ | 58849/65536 [10:08:49<1:08:57,  1.62it/s]
 90%|████████▉ | 58850/65536 [10:08:49<1:09:08,  1.61it/s]
 90%|████████▉ | 58851/65536 [10:08:50<1:08:15,  1.63it/s]
 90%|████████▉ | 58852/65536 [10:08:51<1:10:49,  1.57it/s]
 90%|████████▉ | 58853/65536 [10:08:51<1:09:16,  1.61it/s]
 90%|████████▉ | 58854/65536 [10:08:52<1:08:12,  1.63it/s]
 90%|████████▉ | 58855/65536 [10:08:52<1:09:50,  1.59it/s]
 90%|████████▉ | 58856/65536 [10:08:53<1:11:33,  1.56it/s]
 90%|████████▉ | 58857/65536 [10:08:54<1:10:09,  1.59it/s]
 90%|████████▉ | 58858/65536 [10:08:54<1:10:00,  1.59it/s]
 90%|████████▉ | 58859/65536 [10:08:55<1:09:10,  1.61it/s]
 90%|████████▉ | 58860/65536 [10:08:56<1:12:01,  1.54it/s]
                                                          
{'loss': 1.5321, 'learning_rate': 1.9196155258969028e-07, 'epoch': 3633.33}
+
 90%|████████▉ | 58860/65536 [10:08:56<1:12:01,  1.54it/s]
 90%|████████▉ | 58861/65536 [10:08:56<1:10:47,  1.57it/s]
 90%|████████▉ | 58862/65536 [10:08:57<1:10:29,  1.58it/s]
 90%|████████▉ | 58863/65536 [10:08:57<1:10:32,  1.58it/s]
 90%|████████▉ | 58864/65536 [10:08:58<1:09:42,  1.60it/s]
 90%|████████▉ | 58865/65536 [10:08:59<1:08:44,  1.62it/s]
 90%|████████▉ | 58866/65536 [10:08:59<1:08:38,  1.62it/s]
 90%|████████▉ | 58867/65536 [10:09:00<1:08:19,  1.63it/s]
 90%|████████▉ | 58868/65536 [10:09:00<1:06:53,  1.66it/s]
 90%|████████▉ | 58869/65536 [10:09:01<1:06:20,  1.67it/s]
 90%|████████▉ | 58870/65536 [10:09:02<1:07:22,  1.65it/s]
 90%|████████▉ | 58871/65536 [10:09:02<1:10:40,  1.57it/s]
 90%|████████▉ | 58872/65536 [10:09:03<1:09:50,  1.59it/s]
 90%|████████▉ | 58873/65536 [10:09:04<1:09:26,  1.60it/s]
 90%|████████▉ | 58874/65536 [10:09:04<1:09:42,  1.59it/s]
 90%|████████▉ | 58875/65536 [10:09:05<1:08:46,  1.61it/s]
 90%|████████▉ | 58876/65536 [10:09:06<1:10:35,  1.57it/s]
 90%|████████▉ | 58877/65536 [10:09:06<1:08:37,  1.62it/s]
 90%|████████▉ | 58878/65536 [10:09:07<1:08:48,  1.61it/s]
 90%|████████▉ | 58879/65536 [10:09:07<1:08:51,  1.61it/s]
 90%|████████▉ | 58880/65536 [10:09:08<1:08:15,  1.63it/s]
                                                          
{'loss': 1.5373, 'learning_rate': 1.916860536304641e-07, 'epoch': 3634.57}
+
 90%|████████▉ | 58880/65536 [10:09:08<1:08:15,  1.63it/s]
 90%|████████▉ | 58881/65536 [10:09:09<1:09:49,  1.59it/s]
 90%|████████▉ | 58882/65536 [10:09:09<1:09:52,  1.59it/s]
 90%|████████▉ | 58883/65536 [10:09:10<1:07:39,  1.64it/s]
 90%|████████▉ | 58884/65536 [10:09:10<1:07:52,  1.63it/s]
 90%|████████▉ | 58885/65536 [10:09:11<1:07:27,  1.64it/s]
 90%|████████▉ | 58886/65536 [10:09:12<1:05:47,  1.68it/s]
 90%|██████��█▉ | 58887/65536 [10:09:12<1:05:09,  1.70it/s]
 90%|████████▉ | 58888/65536 [10:09:13<1:09:15,  1.60it/s]
 90%|████████▉ | 58889/65536 [10:09:13<1:08:11,  1.62it/s]
 90%|████████▉ | 58890/65536 [10:09:14<1:07:53,  1.63it/s]
 90%|████████▉ | 58891/65536 [10:09:15<1:06:25,  1.67it/s]
 90%|████████▉ | 58892/65536 [10:09:15<1:05:47,  1.68it/s]
 90%|████████▉ | 58893/65536 [10:09:16<1:06:39,  1.66it/s]
 90%|████████▉ | 58894/65536 [10:09:16<1:07:30,  1.64it/s]
 90%|████████▉ | 58895/65536 [10:09:17<1:06:12,  1.67it/s]
 90%|████████▉ | 58896/65536 [10:09:18<1:07:13,  1.65it/s]
 90%|████████▉ | 58897/65536 [10:09:18<1:06:14,  1.67it/s]
 90%|████████▉ | 58898/65536 [10:09:19<1:07:53,  1.63it/s]
 90%|████████▉ | 58899/65536 [10:09:20<1:08:14,  1.62it/s]
 90%|████████▉ | 58900/65536 [10:09:20<1:07:09,  1.65it/s]
                                                          
{'loss': 1.567, 'learning_rate': 1.9141055467123787e-07, 'epoch': 3635.8}
+
 90%|████████▉ | 58900/65536 [10:09:20<1:07:09,  1.65it/s]
 90%|████████▉ | 58901/65536 [10:09:21<1:06:42,  1.66it/s]
 90%|████████▉ | 58902/65536 [10:09:21<1:08:29,  1.61it/s]
 90%|████████▉ | 58903/65536 [10:09:22<1:09:22,  1.59it/s]
 90%|████████▉ | 58904/65536 [10:09:23<1:10:30,  1.57it/s]
 90%|████████▉ | 58905/65536 [10:09:23<1:08:28,  1.61it/s]
 90%|████████▉ | 58906/65536 [10:09:24<1:08:45,  1.61it/s]
 90%|████████▉ | 58907/65536 [10:09:24<1:08:15,  1.62it/s]
 90%|████████▉ | 58908/65536 [10:09:25<1:09:55,  1.58it/s]
 90%|████████▉ | 58909/65536 [10:09:26<1:09:18,  1.59it/s]
 90%|████████▉ | 58910/65536 [10:09:26<1:09:07,  1.60it/s]
 90%|████████▉ | 58911/65536 [10:09:27<1:09:21,  1.59it/s]
 90%|████████▉ | 58912/65536 [10:09:28<1:10:10,  1.57it/s]
 90%|████████▉ | 58913/65536 [10:09:28<1:07:54,  1.63it/s]
 90%|████████▉ | 58914/65536 [10:09:29<1:06:44,  1.65it/s]
 90%|████████▉ | 58915/65536 [10:09:29<1:07:13,  1.64it/s]
 90%|████████▉ | 58916/65536 [10:09:30<1:07:49,  1.63it/s]
 90%|████████▉ | 58917/65536 [10:09:31<1:08:09,  1.62it/s]
 90%|████████▉ | 58918/65536 [10:09:31<1:07:28,  1.63it/s]
 90%|████████▉ | 58919/65536 [10:09:32<1:06:58,  1.65it/s]
 90%|████████▉ | 58920/65536 [10:09:33<1:09:34,  1.58it/s]
                                                          
{'loss': 1.5553, 'learning_rate': 1.9113505571201172e-07, 'epoch': 3637.04}
+
 90%|████████▉ | 58920/65536 [10:09:33<1:09:34,  1.58it/s]
 90%|████████▉ | 58921/65536 [10:09:33<1:11:10,  1.55it/s]
 90%|████████▉ | 58922/65536 [10:09:34<1:09:32,  1.58it/s]
 90%|████████▉ | 58923/65536 [10:09:34<1:08:38,  1.61it/s]
 90%|████████▉ | 58924/65536 [10:09:35<1:07:46,  1.63it/s]
 90%|████████▉ | 58925/65536 [10:09:36<1:07:39,  1.63it/s]
 90%|████████▉ | 58926/65536 [10:09:36<1:08:02,  1.62it/s]
 90%|████████▉ | 58927/65536 [10:09:37<1:07:32,  1.63it/s]
 90%|████████▉ | 58928/65536 [10:09:38<1:09:05,  1.59it/s]
 90%|████████▉ | 58929/65536 [10:09:38<1:09:53,  1.58it/s]
 90%|████████▉ | 58930/65536 [10:09:39<1:09:13,  1.59it/s]
 90%|████████▉ | 58931/65536 [10:09:39<1:08:17,  1.61it/s]
 90%|████████▉ | 58932/65536 [10:09:40<1:08:51,  1.60it/s]
 90%|████████▉ | 58933/65536 [10:09:41<1:09:36,  1.58it/s]
 90%|████████▉ | 58934/65536 [10:09:41<1:07:56,  1.62it/s]
 90%|████████▉ | 58935/65536 [10:09:42<1:07:58,  1.62it/s]
 90%|████████▉ | 58936/65536 [10:09:43<1:09:18,  1.59it/s]
 90%|████████▉ | 58937/65536 [10:09:43<1:07:27,  1.63it/s]
 90%|████████▉ | 58938/65536 [10:09:44<1:07:46,  1.62it/s]
 90%|████████▉ | 58939/65536 [10:09:44<1:08:35,  1.60it/s]
 90%|████████▉ | 58940/65536 [10:09:45<1:08:56,  1.59it/s]
                                                          
{'loss': 1.5572, 'learning_rate': 1.9085955675278558e-07, 'epoch': 3638.27}
+
 90%|████████▉ | 58940/65536 [10:09:45<1:08:56,  1.59it/s]
 90%|████████▉ | 58941/65536 [10:09:46<1:09:27,  1.58it/s]
 90%|████████▉ | 58942/65536 [10:09:46<1:08:23,  1.61it/s]
 90%|████████▉ | 58943/65536 [10:09:47<1:07:33,  1.63it/s]
 90%|████████▉ | 58944/65536 [10:09:47<1:06:06,  1.66it/s]
 90%|████████▉ | 58945/65536 [10:09:48<1:05:49,  1.67it/s]
 90%|████████▉ | 58946/65536 [10:09:49<1:06:24,  1.65it/s]
 90%|████████▉ | 58947/65536 [10:09:49<1:04:24,  1.71it/s]
 90%|████████▉ | 58948/65536 [10:09:50<1:07:08,  1.64it/s]
 90%|████████▉ | 58949/65536 [10:09:50<1:08:00,  1.61it/s]
 90%|████████▉ | 58950/65536 [10:09:51<1:08:56,  1.59it/s]
 90%|████████▉ | 58951/65536 [10:09:52<1:08:09,  1.61it/s]
 90%|████████▉ | 58952/65536 [10:09:52<1:09:41,  1.57it/s]
 90%|████████▉ | 58953/65536 [10:09:53<1:09:56,  1.57it/s]
 90%|████████▉ | 58954/65536 [10:09:54<1:10:45,  1.55it/s]
 90%|████████▉ | 58955/65536 [10:09:54<1:08:51,  1.59it/s]
 90%|████████▉ | 58956/65536 [10:09:55<1:07:46,  1.62it/s]
 90%|████████▉ | 58957/65536 [10:09:56<1:08:09,  1.61it/s]
 90%|████████▉ | 58958/65536 [10:09:56<1:08:42,  1.60it/s]
 90%|████████▉ | 58959/65536 [10:09:57<1:07:42,  1.62it/s]
 90%|████████▉ | 58960/65536 [10:09:57<1:07:14,  1.63it/s]
                                                          
{'loss': 1.5682, 'learning_rate': 1.9058405779355944e-07, 'epoch': 3639.51}
+
 90%|████████▉ | 58960/65536 [10:09:57<1:07:14,  1.63it/s]
 90%|████████▉ | 58961/65536 [10:09:58<1:06:17,  1.65it/s]
 90%|████████▉ | 58962/65536 [10:09:59<1:05:45,  1.67it/s]
 90%|████████▉ | 58963/65536 [10:09:59<1:07:01,  1.63it/s]
 90%|████████▉ | 58964/65536 [10:10:00<1:06:05,  1.66it/s]
 90%|████████▉ | 58965/65536 [10:10:00<1:07:09,  1.63it/s]
 90%|████████▉ | 58966/65536 [10:10:01<1:07:40,  1.62it/s]
 90%|████████▉ | 58967/65536 [10:10:02<1:08:12,  1.61it/s]
 90%|████████▉ | 58968/65536 [10:10:02<1:07:11,  1.63it/s]
 90%|████████▉ | 58969/65536 [10:10:03<1:08:23,  1.60it/s]
 90%|████████▉ | 58970/65536 [10:10:04<1:08:07,  1.61it/s]
 90%|████████▉ | 58971/65536 [10:10:04<1:06:25,  1.65it/s]
 90%|████████▉ | 58972/65536 [10:10:05<1:10:38,  1.55it/s]
 90%|████████▉ | 58973/65536 [10:10:05<1:09:46,  1.57it/s]
 90%|████████▉ | 58974/65536 [10:10:06<1:08:44,  1.59it/s]
 90%|████████▉ | 58975/65536 [10:10:07<1:09:44,  1.57it/s]
 90%|████████▉ | 58976/65536 [10:10:07<1:07:52,  1.61it/s]
 90%|████████▉ | 58977/65536 [10:10:08<1:09:14,  1.58it/s]
 90%|████████▉ | 58978/65536 [10:10:09<1:09:03,  1.58it/s]
 90%|████████▉ | 58979/65536 [10:10:09<1:08:02,  1.61it/s]
 90%|████████▉ | 58980/65536 [10:10:10<1:07:12,  1.63it/s]
                                                          
{'loss': 1.5635, 'learning_rate': 1.903085588343333e-07, 'epoch': 3640.74}
+
 90%|████████▉ | 58980/65536 [10:10:10<1:07:12,  1.63it/s]
 90%|████████▉ | 58981/65536 [10:10:10<1:07:36,  1.62it/s]
 90%|████████▉ | 58982/65536 [10:10:11<1:08:15,  1.60it/s]
 90%|█████████ | 58983/65536 [10:10:12<1:07:51,  1.61it/s]
 90%|█████████ | 58984/65536 [10:10:12<1:08:12,  1.60it/s]
 90%|█████████ | 58985/65536 [10:10:13<1:09:37,  1.57it/s]
 90%|█████████ | 58986/65536 [10:10:14<1:08:24,  1.60it/s]
 90%|█████████ | 58987/65536 [10:10:14<1:08:22,  1.60it/s]
 90%|█████████ | 58988/65536 [10:10:15<1:07:21,  1.62it/s]
 90%|█████████ | 58989/65536 [10:10:15<1:07:00,  1.63it/s]
 90%|█████████ | 58990/65536 [10:10:16<1:08:18,  1.60it/s]
 90%|█████████ | 58991/65536 [10:10:17<1:07:01,  1.63it/s]
 90%|█████████ | 58992/65536 [10:10:17<1:07:08,  1.62it/s]
 90%|█████████ | 58993/65536 [10:10:18<1:06:39,  1.64it/s]
 90%|█████████ | 58994/65536 [10:10:18<1:05:25,  1.67it/s]
 90%|█████████ | 58995/65536 [10:10:19<1:08:53,  1.58it/s]
 90%|█████████ | 58996/65536 [10:10:20<1:06:57,  1.63it/s]
 90%|█████████ | 58997/65536 [10:10:20<1:07:17,  1.62it/s]
 90%|█████████ | 58998/65536 [10:10:21<1:08:08,  1.60it/s]
 90%|█████████ | 58999/65536 [10:10:22<1:06:33,  1.64it/s]
 90%|█████████ | 59000/65536 [10:10:22<1:06:23,  1.64it/s]
                                                          
{'loss': 1.5635, 'learning_rate': 1.9003305987510716e-07, 'epoch': 3641.98}
+
 90%|█████████ | 59000/65536 [10:10:22<1:06:23,  1.64it/s]
 90%|█████████ | 59001/65536 [10:10:23<1:10:12,  1.55it/s]
 90%|█████████ | 59002/65536 [10:10:24<1:11:03,  1.53it/s]
 90%|█████████ | 59003/65536 [10:10:24<1:08:56,  1.58it/s]
 90%|█████████ | 59004/65536 [10:10:25<1:07:25,  1.61it/s]
 90%|█████████ | 59005/65536 [10:10:25<1:06:42,  1.63it/s]
 90%|█████████ | 59006/65536 [10:10:26<1:06:16,  1.64it/s]
 90%|█████████ | 59007/65536 [10:10:27<1:06:40,  1.63it/s]
 90%|█████████ | 59008/65536 [10:10:27<1:08:05,  1.60it/s]
 90%|█████████ | 59009/65536 [10:10:28<1:08:14,  1.59it/s]
 90%|█████████ | 59010/65536 [10:10:28<1:07:18,  1.62it/s]
 90%|█████████ | 59011/65536 [10:10:29<1:07:27,  1.61it/s]
 90%|█████████ | 59012/65536 [10:10:30<1:06:56,  1.62it/s]
 90%|█████████ | 59013/65536 [10:10:30<1:05:29,  1.66it/s]
 90%|█████████ | 59014/65536 [10:10:31<1:05:18,  1.66it/s]
 90%|█████████ | 59015/65536 [10:10:31<1:06:04,  1.64it/s]
 90%|█████████ | 59016/65536 [10:10:32<1:06:27,  1.64it/s]
 90%|█████████ | 59017/65536 [10:10:33<1:07:35,  1.61it/s]
 90%|█████████ | 59018/65536 [10:10:33<1:07:19,  1.61it/s]
 90%|█████████ | 59019/65536 [10:10:34<1:06:37,  1.63it/s]
 90%|█████████ | 59020/65536 [10:10:35<1:06:49,  1.62it/s]
                                                          
{'loss': 1.5658, 'learning_rate': 1.8975756091588102e-07, 'epoch': 3643.21}
+
 90%|█████████ | 59020/65536 [10:10:35<1:06:49,  1.62it/s]
 90%|█████████ | 59021/65536 [10:10:35<1:06:44,  1.63it/s]
 90%|█████████ | 59022/65536 [10:10:36<1:07:50,  1.60it/s]
 90%|█████████ | 59023/65536 [10:10:37<1:09:34,  1.56it/s]
 90%|█████████ | 59024/65536 [10:10:37<1:08:35,  1.58it/s]
 90%|█████████ | 59025/65536 [10:10:38<1:10:03,  1.55it/s]
 90%|█████████ | 59026/65536 [10:10:38<1:08:01,  1.59it/s]
 90%|█████████ | 59027/65536 [10:10:39<1:07:01,  1.62it/s]
 90%|█████████ | 59028/65536 [10:10:40<1:06:45,  1.62it/s]
 90%|█████████ | 59029/65536 [10:10:40<1:06:39,  1.63it/s]
 90%|█████████ | 59030/65536 [10:10:41<1:06:09,  1.64it/s]
 90%|█████████ | 59031/65536 [10:10:41<1:06:42,  1.63it/s]
 90%|█████████ | 59032/65536 [10:10:42<1:05:04,  1.67it/s]
 90%|█████████ | 59033/65536 [10:10:43<1:08:25,  1.58it/s]
 90%|█████████ | 59034/65536 [10:10:43<1:09:04,  1.57it/s]
 90%|█████████ | 59035/65536 [10:10:44<1:09:02,  1.57it/s]
 90%|█████████ | 59036/65536 [10:10:45<1:08:46,  1.58it/s]
 90%|█████████ | 59037/65536 [10:10:45<1:08:05,  1.59it/s]
 90%|█████████ | 59038/65536 [10:10:46<1:08:45,  1.57it/s]
 90%|█████████ | 59039/65536 [10:10:46<1:07:05,  1.61it/s]
 90%|█████████ | 59040/65536 [10:10:47<1:06:09,  1.64it/s]
                                                          
{'loss': 1.5657, 'learning_rate': 1.8948206195665477e-07, 'epoch': 3644.44}
+
 90%|█████████ | 59040/65536 [10:10:47<1:06:09,  1.64it/s]
 90%|█████████ | 59041/65536 [10:10:48<1:05:26,  1.65it/s]
 90%|█████████ | 59042/65536 [10:10:48<1:05:07,  1.66it/s]
 90%|█████████ | 59043/65536 [10:10:49<1:05:57,  1.64it/s]
 90%|█████████ | 59044/65536 [10:10:49<1:05:58,  1.64it/s]
 90%|█████████ | 59045/65536 [10:10:50<1:06:01,  1.64it/s]
 90%|█████████ | 59046/65536 [10:10:51<1:06:40,  1.62it/s]
 90%|█████████ | 59047/65536 [10:10:51<1:06:29,  1.63it/s]
 90%|█████████ | 59048/65536 [10:10:52<1:07:19,  1.61it/s]
 90%|█████████ | 59049/65536 [10:10:53<1:07:20,  1.61it/s]
 90%|█████████ | 59050/65536 [10:10:53<1:10:35,  1.53it/s]
 90%|█████████ | 59051/65536 [10:10:54<1:10:21,  1.54it/s]
 90%|█████████ | 59052/65536 [10:10:55<1:10:28,  1.53it/s]
 90%|█████████ | 59053/65536 [10:10:55<1:09:10,  1.56it/s]
 90%|█████████ | 59054/65536 [10:10:56<1:09:51,  1.55it/s]
 90%|█████████ | 59055/65536 [10:10:56<1:07:05,  1.61it/s]
 90%|█████████ | 59056/65536 [10:10:57<1:04:56,  1.66it/s]
 90%|█████████ | 59057/65536 [10:10:58<1:04:17,  1.68it/s]
 90%|█████████ | 59058/65536 [10:10:58<1:05:30,  1.65it/s]
 90%|█████████ | 59059/65536 [10:10:59<1:05:05,  1.66it/s]
 90%|█████████ | 59060/65536 [10:10:59<1:04:55,  1.66it/s]
                                                          
{'loss': 1.589, 'learning_rate': 1.8920656299742863e-07, 'epoch': 3645.68}
+
 90%|█████████ | 59060/65536 [10:10:59<1:04:55,  1.66it/s]
 90%|█████████ | 59061/65536 [10:11:00<1:05:13,  1.65it/s]
 90%|█████████ | 59062/65536 [10:11:01<1:05:31,  1.65it/s]
 90%|█████████ | 59063/65536 [10:11:01<1:06:42,  1.62it/s]
 90%|█████████ | 59064/65536 [10:11:02<1:06:08,  1.63it/s]
 90%|█████████ | 59065/65536 [10:11:02<1:05:07,  1.66it/s]
 90%|█████████ | 59066/65536 [10:11:03<1:07:50,  1.59it/s]
 90%|█████████ | 59067/65536 [10:11:04<1:09:00,  1.56it/s]
 90%|█████████ | 59068/65536 [10:11:04<1:07:07,  1.61it/s]
 90%|█████████ | 59069/65536 [10:11:05<1:08:47,  1.57it/s]
 90%|█████████ | 59070/65536 [10:11:06<1:07:58,  1.59it/s]
 90%|█████████ | 59071/65536 [10:11:06<1:06:56,  1.61it/s]
 90%|█████████ | 59072/65536 [10:11:07<1:06:45,  1.61it/s]
 90%|█████████ | 59073/65536 [10:11:08<1:07:06,  1.61it/s]
 90%|█████████ | 59074/65536 [10:11:08<1:08:23,  1.57it/s]
 90%|█████████ | 59075/65536 [10:11:09<1:06:50,  1.61it/s]
 90%|█████████ | 59076/65536 [10:11:09<1:07:01,  1.61it/s]
 90%|█████████ | 59077/65536 [10:11:10<1:06:24,  1.62it/s]
 90%|█████████ | 59078/65536 [10:11:11<1:05:20,  1.65it/s]
 90%|█████████ | 59079/65536 [10:11:11<1:03:57,  1.68it/s]
 90%|█████████ | 59080/65536 [10:11:12<1:05:28,  1.64it/s]
                                                          
{'loss': 1.5261, 'learning_rate': 1.889310640382025e-07, 'epoch': 3646.91}
+
 90%|█████████ | 59080/65536 [10:11:12<1:05:28,  1.64it/s]
 90%|█████████ | 59081/65536 [10:11:12<1:06:14,  1.62it/s]
 90%|█████████ | 59082/65536 [10:11:13<1:07:35,  1.59it/s]
 90%|█████████ | 59083/65536 [10:11:14<1:06:08,  1.63it/s]
 90%|█████████ | 59084/65536 [10:11:14<1:06:26,  1.62it/s]
 90%|█████████ | 59085/65536 [10:11:15<1:05:58,  1.63it/s]
 90%|█████████ | 59086/65536 [10:11:16<1:07:52,  1.58it/s]
 90%|█████████ | 59087/65536 [10:11:16<1:06:06,  1.63it/s]
 90%|█████████ | 59088/65536 [10:11:17<1:05:24,  1.64it/s]
 90%|█████████ | 59089/65536 [10:11:17<1:05:24,  1.64it/s]
 90%|█████████ | 59090/65536 [10:11:18<1:04:29,  1.67it/s]
 90%|█████████ | 59091/65536 [10:11:19<1:06:27,  1.62it/s]
 90%|█████████ | 59092/65536 [10:11:19<1:07:00,  1.60it/s]
 90%|█████████ | 59093/65536 [10:11:20<1:07:38,  1.59it/s]
 90%|█████████ | 59094/65536 [10:11:21<1:07:27,  1.59it/s]
 90%|█████████ | 59095/65536 [10:11:21<1:06:16,  1.62it/s]
 90%|█████████ | 59096/65536 [10:11:22<1:08:04,  1.58it/s]
 90%|█████████ | 59097/65536 [10:11:22<1:06:22,  1.62it/s]
 90%|█████████ | 59098/65536 [10:11:23<1:08:16,  1.57it/s]
 90%|█████████ | 59099/65536 [10:11:24<1:07:07,  1.60it/s]
 90%|█████████ | 59100/65536 [10:11:24<1:08:23,  1.57it/s]
                                                          
{'loss': 1.5241, 'learning_rate': 1.8865556507897635e-07, 'epoch': 3648.15}
+
 90%|█████████ | 59100/65536 [10:11:24<1:08:23,  1.57it/s]
 90%|█████████ | 59101/65536 [10:11:25<1:07:28,  1.59it/s]
 90%|█████████ | 59102/65536 [10:11:26<1:08:44,  1.56it/s]
 90%|█████████ | 59103/65536 [10:11:26<1:06:48,  1.60it/s]
 90%|█████████ | 59104/65536 [10:11:27<1:05:58,  1.63it/s]
 90%|█████████ | 59105/65536 [10:11:27<1:09:11,  1.55it/s]
 90%|█████████ | 59106/65536 [10:11:28<1:07:10,  1.60it/s]
 90%|█████████ | 59107/65536 [10:11:29<1:08:12,  1.57it/s]
 90%|█████████ | 59108/65536 [10:11:29<1:07:16,  1.59it/s]
 90%|█████████ | 59109/65536 [10:11:30<1:05:52,  1.63it/s]
 90%|█████████ | 59110/65536 [10:11:30<1:04:50,  1.65it/s]
 90%|█████████ | 59111/65536 [10:11:31<1:05:42,  1.63it/s]
 90%|█████████ | 59112/65536 [10:11:32<1:04:54,  1.65it/s]
 90%|█████████ | 59113/65536 [10:11:32<1:04:16,  1.67it/s]
 90%|█████████ | 59114/65536 [10:11:33<1:06:20,  1.61it/s]
 90%|█████████ | 59115/65536 [10:11:34<1:05:25,  1.64it/s]
 90%|█████████ | 59116/65536 [10:11:34<1:07:12,  1.59it/s]
 90%|█████████ | 59117/65536 [10:11:35<1:06:55,  1.60it/s]
 90%|█████████ | 59118/65536 [10:11:35<1:06:57,  1.60it/s]
 90%|█████████ | 59119/65536 [10:11:36<1:05:18,  1.64it/s]
 90%|█████████ | 59120/65536 [10:11:37<1:04:48,  1.65it/s]
                                                          
{'loss': 1.5414, 'learning_rate': 1.883800661197502e-07, 'epoch': 3649.38}
+
 90%|█████████ | 59120/65536 [10:11:37<1:04:48,  1.65it/s]
 90%|█████████ | 59121/65536 [10:11:37<1:04:53,  1.65it/s]
 90%|█████████ | 59122/65536 [10:11:38<1:04:30,  1.66it/s]
 90%|█████████ | 59123/65536 [10:11:39<1:07:07,  1.59it/s]
 90%|█████████ | 59124/65536 [10:11:39<1:06:44,  1.60it/s]
 90%|█████████ | 59125/65536 [10:11:40<1:09:09,  1.55it/s]
 90%|█████████ | 59126/65536 [10:11:40<1:07:06,  1.59it/s]
 90%|█████████ | 59127/65536 [10:11:41<1:06:40,  1.60it/s]
 90%|█████████ | 59128/65536 [10:11:42<1:05:23,  1.63it/s]
 90%|█████████ | 59129/65536 [10:11:42<1:03:45,  1.67it/s]
 90%|█████████ | 59130/65536 [10:11:43<1:04:13,  1.66it/s]
 90%|█████████ | 59131/65536 [10:11:44<1:07:44,  1.58it/s]
 90%|█████████ | 59132/65536 [10:11:44<1:08:07,  1.57it/s]
 90%|█████████ | 59133/65536 [10:11:45<1:07:28,  1.58it/s]
 90%|█████████ | 59134/65536 [10:11:45<1:09:04,  1.54it/s]
 90%|█████████ | 59135/65536 [10:11:46<1:07:24,  1.58it/s]
 90%|█████████ | 59136/65536 [10:11:47<1:05:09,  1.64it/s]
 90%|█████████ | 59137/65536 [10:11:47<1:05:36,  1.63it/s]
 90%|█████████ | 59138/65536 [10:11:48<1:04:40,  1.65it/s]
 90%|█████████ | 59139/65536 [10:11:48<1:05:50,  1.62it/s]
 90%|█████████ | 59140/65536 [10:11:49<1:03:48,  1.67it/s]
                                                          
{'loss': 1.5475, 'learning_rate': 1.8810456716052407e-07, 'epoch': 3650.62}
+
 90%|█████████ | 59140/65536 [10:11:49<1:03:48,  1.67it/s]
 90%|█████████ | 59141/65536 [10:11:50<1:04:18,  1.66it/s]
 90%|█████████ | 59142/65536 [10:11:50<1:04:37,  1.65it/s]
 90%|█████████ | 59143/65536 [10:11:51<1:03:32,  1.68it/s]
 90%|█████████ | 59144/65536 [10:11:51<1:05:04,  1.64it/s]
 90%|█████████ | 59145/65536 [10:11:52<1:06:10,  1.61it/s]
 90%|█████████ | 59146/65536 [10:11:53<1:05:24,  1.63it/s]
 90%|█████████ | 59147/65536 [10:11:53<1:07:39,  1.57it/s]
 90%|█████████ | 59148/65536 [10:11:54<1:06:44,  1.60it/s]
 90%|█████████ | 59149/65536 [10:11:55<1:05:53,  1.62it/s]
 90%|█████████ | 59150/65536 [10:11:55<1:06:23,  1.60it/s]
 90%|█████████ | 59151/65536 [10:11:56<1:07:31,  1.58it/s]
 90%|█████████ | 59152/65536 [10:11:57<1:07:26,  1.58it/s]
 90%|█████████ | 59153/65536 [10:11:57<1:06:25,  1.60it/s]
 90%|█████████ | 59154/65536 [10:11:58<1:05:57,  1.61it/s]
 90%|█████████ | 59155/65536 [10:11:58<1:06:38,  1.60it/s]
 90%|█████████ | 59156/65536 [10:11:59<1:05:55,  1.61it/s]
 90%|█████████ | 59157/65536 [10:12:00<1:07:02,  1.59it/s]
 90%|█████████ | 59158/65536 [10:12:00<1:06:03,  1.61it/s]
 90%|█████████ | 59159/65536 [10:12:01<1:06:29,  1.60it/s]
 90%|█████████ | 59160/65536 [10:12:02<1:06:08,  1.61it/s]
                                                          
{'loss': 1.5151, 'learning_rate': 1.8782906820129795e-07, 'epoch': 3651.85}
+
 90%|█████████ | 59160/65536 [10:12:02<1:06:08,  1.61it/s]
 90%|█████████ | 59161/65536 [10:12:02<1:07:22,  1.58it/s]
 90%|█████████ | 59162/65536 [10:12:03<1:06:47,  1.59it/s]
 90%|█████████ | 59163/65536 [10:12:03<1:07:21,  1.58it/s]
 90%|█████████ | 59164/65536 [10:12:04<1:06:12,  1.60it/s]
 90%|█████████ | 59165/65536 [10:12:05<1:05:01,  1.63it/s]
 90%|█████████ | 59166/65536 [10:12:05<1:05:01,  1.63it/s]
 90%|█████████ | 59167/65536 [10:12:06<1:04:27,  1.65it/s]
 90%|█████████ | 59168/65536 [10:12:06<1:04:29,  1.65it/s]
 90%|█████████ | 59169/65536 [10:12:07<1:06:41,  1.59it/s]
 90%|█████████ | 59170/65536 [10:12:08<1:05:48,  1.61it/s]
 90%|█████████ | 59171/65536 [10:12:08<1:05:52,  1.61it/s]
 90%|█████████ | 59172/65536 [10:12:09<1:05:54,  1.61it/s]
 90%|█████████ | 59173/65536 [10:12:10<1:04:33,  1.64it/s]
 90%|█████████ | 59174/65536 [10:12:10<1:04:04,  1.65it/s]
 90%|█████████ | 59175/65536 [10:12:11<1:05:09,  1.63it/s]
 90%|█████████ | 59176/65536 [10:12:11<1:04:40,  1.64it/s]
 90%|█████████ | 59177/65536 [10:12:12<1:04:24,  1.65it/s]
 90%|█████████ | 59178/65536 [10:12:13<1:07:12,  1.58it/s]
 90%|█████████ | 59179/65536 [10:12:13<1:08:02,  1.56it/s]
 90%|█████████ | 59180/65536 [10:12:14<1:06:49,  1.59it/s]
                                                          
{'loss': 1.5475, 'learning_rate': 1.8755356924207168e-07, 'epoch': 3653.09}
+
 90%|█████████ | 59180/65536 [10:12:14<1:06:49,  1.59it/s]
 90%|█████████ | 59181/65536 [10:12:15<1:07:04,  1.58it/s]
 90%|█████████ | 59182/65536 [10:12:15<1:05:17,  1.62it/s]
 90%|█████████ | 59183/65536 [10:12:16<1:05:16,  1.62it/s]
 90%|█████████ | 59184/65536 [10:12:16<1:06:01,  1.60it/s]
 90%|█████████ | 59185/65536 [10:12:17<1:05:40,  1.61it/s]
 90%|█████████ | 59186/65536 [10:12:18<1:05:19,  1.62it/s]
 90%|█████████ | 59187/65536 [10:12:18<1:05:26,  1.62it/s]
 90%|█████████ | 59188/65536 [10:12:19<1:04:17,  1.65it/s]
 90%|█████████ | 59189/65536 [10:12:20<1:06:32,  1.59it/s]
 90%|█████████ | 59190/65536 [10:12:20<1:05:52,  1.61it/s]
 90%|█████████ | 59191/65536 [10:12:21<1:05:26,  1.62it/s]
 90%|█████████ | 59192/65536 [10:12:21<1:06:05,  1.60it/s]
 90%|█████████ | 59193/65536 [10:12:22<1:05:58,  1.60it/s]
 90%|█████████ | 59194/65536 [10:12:23<1:03:58,  1.65it/s]
 90%|█████████ | 59195/65536 [10:12:23<1:07:31,  1.57it/s]
 90%|█████████ | 59196/65536 [10:12:24<1:05:32,  1.61it/s]
 90%|█████████ | 59197/65536 [10:12:25<1:06:48,  1.58it/s]
 90%|█████████ | 59198/65536 [10:12:25<1:06:28,  1.59it/s]
 90%|█████████ | 59199/65536 [10:12:26<1:06:09,  1.60it/s]
 90%|█████████ | 59200/65536 [10:12:26<1:05:06,  1.62it/s]
                                                          
{'loss': 1.5143, 'learning_rate': 1.8727807028284556e-07, 'epoch': 3654.32}
+
 90%|█████████ | 59200/65536 [10:12:26<1:05:06,  1.62it/s]
 90%|█████████ | 59201/65536 [10:12:27<1:05:17,  1.62it/s]
 90%|█████████ | 59202/65536 [10:12:28<1:05:03,  1.62it/s]
 90%|█████████ | 59203/65536 [10:12:28<1:06:18,  1.59it/s]
 90%|█████████ | 59204/65536 [10:12:29<1:04:58,  1.62it/s]
 90%|█████████ | 59205/65536 [10:12:29<1:06:30,  1.59it/s]
 90%|█████████ | 59206/65536 [10:12:30<1:06:39,  1.58it/s]
 90%|█████████ | 59207/65536 [10:12:31<1:08:01,  1.55it/s]
 90%|█████████ | 59208/65536 [10:12:31<1:06:26,  1.59it/s]
 90%|█████████ | 59209/65536 [10:12:32<1:06:42,  1.58it/s]
 90%|█████████ | 59210/65536 [10:12:33<1:05:55,  1.60it/s]
 90%|█████████ | 59211/65536 [10:12:33<1:05:24,  1.61it/s]
 90%|█████████ | 59212/65536 [10:12:34<1:07:11,  1.57it/s]
 90%|█████████ | 59213/65536 [10:12:35<1:06:15,  1.59it/s]
 90%|█████████ | 59214/65536 [10:12:35<1:05:17,  1.61it/s]
 90%|█████████ | 59215/65536 [10:12:36<1:05:26,  1.61it/s]
 90%|█████████ | 59216/65536 [10:12:36<1:05:40,  1.60it/s]
 90%|█████████ | 59217/65536 [10:12:37<1:05:12,  1.62it/s]
 90%|█████████ | 59218/65536 [10:12:38<1:04:26,  1.63it/s]
 90%|█████████ | 59219/65536 [10:12:38<1:05:21,  1.61it/s]
 90%|█████████ | 59220/65536 [10:12:39<1:05:09,  1.62it/s]
                                                          
{'loss': 1.5238, 'learning_rate': 1.8700257132361948e-07, 'epoch': 3655.56}
+
 90%|█████████ | 59220/65536 [10:12:39<1:05:09,  1.62it/s]
 90%|█████████ | 59221/65536 [10:12:39<1:03:21,  1.66it/s]
 90%|█████████ | 59222/65536 [10:12:40<1:02:42,  1.68it/s]
 90%|█████████ | 59223/65536 [10:12:41<1:03:33,  1.66it/s]
 90%|█████████ | 59224/65536 [10:12:41<1:03:48,  1.65it/s]
 90%|█████████ | 59225/65536 [10:12:42<1:03:25,  1.66it/s]
 90%|█████████ | 59226/65536 [10:12:42<1:05:57,  1.59it/s]
 90%|█████████ | 59227/65536 [10:12:43<1:06:27,  1.58it/s]
 90%|█████████ | 59228/65536 [10:12:44<1:08:10,  1.54it/s]
 90%|█████████ | 59229/65536 [10:12:44<1:06:47,  1.57it/s]
 90%|█████████ | 59230/65536 [10:12:45<1:07:25,  1.56it/s]
 90%|█████████ | 59231/65536 [10:12:46<1:06:40,  1.58it/s]
 90%|█████████ | 59232/65536 [10:12:46<1:05:09,  1.61it/s]
 90%|█████████ | 59233/65536 [10:12:47<1:05:27,  1.60it/s]
 90%|█████████ | 59234/65536 [10:12:48<1:04:14,  1.64it/s]
 90%|█████████ | 59235/65536 [10:12:48<1:03:43,  1.65it/s]
 90%|█████████ | 59236/65536 [10:12:49<1:05:22,  1.61it/s]
 90%|█████████ | 59237/65536 [10:12:49<1:04:29,  1.63it/s]
 90%|█████████ | 59238/65536 [10:12:50<1:03:23,  1.66it/s]
 90%|█���███████ | 59239/65536 [10:12:51<1:04:43,  1.62it/s]
 90%|█████████ | 59240/65536 [10:12:51<1:04:02,  1.64it/s]
                                                          
{'loss': 1.5603, 'learning_rate': 1.867270723643933e-07, 'epoch': 3656.79}
+
 90%|█████████ | 59240/65536 [10:12:51<1:04:02,  1.64it/s]
 90%|█████████ | 59241/65536 [10:12:52<1:04:59,  1.61it/s]
 90%|█████████ | 59242/65536 [10:12:52<1:03:46,  1.64it/s]
 90%|█████████ | 59243/65536 [10:12:53<1:04:44,  1.62it/s]
 90%|█████████ | 59244/65536 [10:12:54<1:06:59,  1.57it/s]
 90%|█████████ | 59245/65536 [10:12:54<1:07:33,  1.55it/s]
 90%|█████████ | 59246/65536 [10:12:55<1:06:30,  1.58it/s]
 90%|█████████ | 59247/65536 [10:12:56<1:06:28,  1.58it/s]
 90%|█████████ | 59248/65536 [10:12:56<1:05:22,  1.60it/s]
 90%|█████████ | 59249/65536 [10:12:57<1:04:24,  1.63it/s]
 90%|█████████ | 59250/65536 [10:12:57<1:03:49,  1.64it/s]
 90%|█████████ | 59251/65536 [10:12:58<1:04:21,  1.63it/s]
 90%|█████████ | 59252/65536 [10:12:59<1:03:53,  1.64it/s]
 90%|█████████ | 59253/65536 [10:12:59<1:02:29,  1.68it/s]
 90%|█████████ | 59254/65536 [10:13:00<1:04:41,  1.62it/s]
 90%|█████████ | 59255/65536 [10:13:00<1:04:20,  1.63it/s]
 90%|█████████ | 59256/65536 [10:13:01<1:03:26,  1.65it/s]
 90%|█████████ | 59257/65536 [10:13:02<1:03:49,  1.64it/s]
 90%|█████████ | 59258/65536 [10:13:02<1:04:53,  1.61it/s]
 90%|█████████ | 59259/65536 [10:13:03<1:04:38,  1.62it/s]
 90%|█████████ | 59260/65536 [10:13:04<1:07:07,  1.56it/s]
                                                          
{'loss': 1.5396, 'learning_rate': 1.8645157340516717e-07, 'epoch': 3658.02}
+
 90%|█████████ | 59260/65536 [10:13:04<1:07:07,  1.56it/s]
 90%|█████████ | 59261/65536 [10:13:04<1:08:01,  1.54it/s]
 90%|█████████ | 59262/65536 [10:13:05<1:08:01,  1.54it/s]
 90%|█████████ | 59263/65536 [10:13:06<1:09:44,  1.50it/s]
 90%|█████████ | 59264/65536 [10:13:06<1:07:54,  1.54it/s]
 90%|█████████ | 59265/65536 [10:13:07<1:06:46,  1.57it/s]
 90%|█████████ | 59266/65536 [10:13:08<1:06:13,  1.58it/s]
 90%|█████████ | 59267/65536 [10:13:08<1:07:31,  1.55it/s]
 90%|█████████ | 59268/65536 [10:13:09<1:07:02,  1.56it/s]
 90%|█████████ | 59269/65536 [10:13:09<1:05:33,  1.59it/s]
 90%|█████████ | 59270/65536 [10:13:10<1:03:55,  1.63it/s]
 90%|█████████ | 59271/65536 [10:13:11<1:03:40,  1.64it/s]
 90%|█████████ | 59272/65536 [10:13:11<1:03:19,  1.65it/s]
 90%|█████████ | 59273/65536 [10:13:12<1:03:52,  1.63it/s]
 90%|█████████ | 59274/65536 [10:13:12<1:04:51,  1.61it/s]
 90%|█████████ | 59275/65536 [10:13:13<1:04:51,  1.61it/s]
 90%|█████████ | 59276/65536 [10:13:14<1:07:28,  1.55it/s]
 90%|█████████ | 59277/65536 [10:13:14<1:06:28,  1.57it/s]
 90%|█████████ | 59278/65536 [10:13:15<1:05:19,  1.60it/s]
 90%|█████████ | 59279/65536 [10:13:16<1:05:57,  1.58it/s]
 90%|█████████ | 59280/65536 [10:13:16<1:06:35,  1.57it/s]
                                                          
{'loss': 1.5327, 'learning_rate': 1.8617607444594103e-07, 'epoch': 3659.26}
+
 90%|█████████ | 59280/65536 [10:13:16<1:06:35,  1.57it/s]
 90%|█████████ | 59281/65536 [10:13:17<1:05:39,  1.59it/s]
 90%|█████████ | 59282/65536 [10:13:18<1:07:59,  1.53it/s]
 90%|█████████ | 59283/65536 [10:13:18<1:07:21,  1.55it/s]
 90%|█████████ | 59284/65536 [10:13:19<1:07:32,  1.54it/s]
 90%|█████████ | 59285/65536 [10:13:19<1:05:32,  1.59it/s]
 90%|█████████ | 59286/65536 [10:13:20<1:05:54,  1.58it/s]
 90%|█████████ | 59287/65536 [10:13:21<1:06:14,  1.57it/s]
 90%|█████████ | 59288/65536 [10:13:21<1:04:35,  1.61it/s]
 90%|█████████ | 59289/65536 [10:13:22<1:04:03,  1.63it/s]
 90%|█████████ | 59290/65536 [10:13:23<1:06:53,  1.56it/s]
 90%|█████████ | 59291/65536 [10:13:23<1:04:38,  1.61it/s]
 90%|█████████ | 59292/65536 [10:13:24<1:03:14,  1.65it/s]
 90%|█████████ | 59293/65536 [10:13:25<1:07:19,  1.55it/s]
 90%|█████████ | 59294/65536 [10:13:25<1:08:09,  1.53it/s]
 90%|█████████ | 59295/65536 [10:13:26<1:08:12,  1.53it/s]
 90%|█████████ | 59296/65536 [10:13:27<1:07:29,  1.54it/s]
 90%|█████��███ | 59297/65536 [10:13:27<1:05:53,  1.58it/s]
 90%|█████████ | 59298/65536 [10:13:28<1:04:30,  1.61it/s]
 90%|█████████ | 59299/65536 [10:13:28<1:04:19,  1.62it/s]
 90%|█████████ | 59300/65536 [10:13:29<1:03:22,  1.64it/s]
                                                          
{'loss': 1.5517, 'learning_rate': 1.8590057548671478e-07, 'epoch': 3660.49}
+
 90%|█████████ | 59300/65536 [10:13:29<1:03:22,  1.64it/s]
 90%|█████████ | 59301/65536 [10:13:29<1:02:12,  1.67it/s]
 90%|█████████ | 59302/65536 [10:13:30<1:03:05,  1.65it/s]
 90%|█████████ | 59303/65536 [10:13:31<1:01:58,  1.68it/s]
 90%|█████████ | 59304/65536 [10:13:31<1:02:36,  1.66it/s]
 90%|█████████ | 59305/65536 [10:13:32<1:04:30,  1.61it/s]
 90%|█████████ | 59306/65536 [10:13:33<1:03:38,  1.63it/s]
 90%|█████████ | 59307/65536 [10:13:33<1:02:28,  1.66it/s]
 90%|█████████ | 59308/65536 [10:13:34<1:02:37,  1.66it/s]
 90%|█████████ | 59309/65536 [10:13:34<1:05:13,  1.59it/s]
 90%|█████████ | 59310/65536 [10:13:35<1:04:29,  1.61it/s]
 91%|█████████ | 59311/65536 [10:13:36<1:03:33,  1.63it/s]
 91%|█████████ | 59312/65536 [10:13:36<1:06:04,  1.57it/s]
 91%|█████████ | 59313/65536 [10:13:37<1:05:44,  1.58it/s]
 91%|█████████ | 59314/65536 [10:13:38<1:07:42,  1.53it/s]
 91%|█████████ | 59315/65536 [10:13:38<1:06:56,  1.55it/s]
 91%|█████████ | 59316/65536 [10:13:39<1:04:57,  1.60it/s]
 91%|█████████ | 59317/65536 [10:13:39<1:03:57,  1.62it/s]
 91%|█████████ | 59318/65536 [10:13:40<1:03:21,  1.64it/s]
 91%|█████████ | 59319/65536 [10:13:41<1:03:10,  1.64it/s]
 91%|█████████ | 59320/65536 [10:13:41<1:03:21,  1.64it/s]
                                                          
{'loss': 1.5484, 'learning_rate': 1.8562507652748864e-07, 'epoch': 3661.73}
+
 91%|█████████ | 59320/65536 [10:13:41<1:03:21,  1.64it/s]
 91%|█████████ | 59321/65536 [10:13:42<1:03:53,  1.62it/s]
 91%|█████████ | 59322/65536 [10:13:43<1:05:11,  1.59it/s]
 91%|█████████ | 59323/65536 [10:13:43<1:09:13,  1.50it/s]
 91%|█████████ | 59324/65536 [10:13:44<1:09:36,  1.49it/s]
 91%|█████████ | 59325/65536 [10:13:45<1:09:35,  1.49it/s]
 91%|█████████ | 59326/65536 [10:13:45<1:07:25,  1.53it/s]
 91%|█████████ | 59327/65536 [10:13:46<1:06:58,  1.55it/s]
 91%|█████████ | 59328/65536 [10:13:47<1:05:24,  1.58it/s]
 91%|█████████ | 59329/65536 [10:13:47<1:06:53,  1.55it/s]
 91%|█████████ | 59330/65536 [10:13:48<1:04:48,  1.60it/s]
 91%|█████████ | 59331/65536 [10:13:48<1:04:00,  1.62it/s]
 91%|█████████ | 59332/65536 [10:13:49<1:04:36,  1.60it/s]
 91%|█████████ | 59333/65536 [10:13:50<1:05:49,  1.57it/s]
 91%|█████████ | 59334/65536 [10:13:50<1:06:01,  1.57it/s]
 91%|█████████ | 59335/65536 [10:13:51<1:04:18,  1.61it/s]
 91%|█████████ | 59336/65536 [10:13:52<1:04:57,  1.59it/s]
 91%|█████████ | 59337/65536 [10:13:52<1:03:40,  1.62it/s]
 91%|█████████ | 59338/65536 [10:13:53<1:03:04,  1.64it/s]
 91%|█████████ | 59339/65536 [10:13:53<1:02:37,  1.65it/s]
 91%|█████████ | 59340/65536 [10:13:54<1:04:20,  1.61it/s]
                                                          
{'loss': 1.5254, 'learning_rate': 1.853495775682625e-07, 'epoch': 3662.96}
+
 91%|█████████ | 59340/65536 [10:13:54<1:04:20,  1.61it/s]
 91%|█████████ | 59341/65536 [10:13:55<1:05:48,  1.57it/s]
 91%|█████████ | 59342/65536 [10:13:55<1:05:21,  1.58it/s]
 91%|█████████ | 59343/65536 [10:13:56<1:04:49,  1.59it/s]
 91%|█████████ | 59344/65536 [10:13:57<1:05:26,  1.58it/s]
 91%|█████████ | 59345/65536 [10:13:57<1:04:29,  1.60it/s]
 91%|█████████ | 59346/65536 [10:13:58<1:03:12,  1.63it/s]
 91%|█████████ | 59347/65536 [10:13:58<1:03:04,  1.64it/s]
 91%|█████████ | 59348/65536 [10:13:59<1:02:00,  1.66it/s]
 91%|█████████ | 59349/65536 [10:14:00<1:03:01,  1.64it/s]
 91%|█████████ | 59350/65536 [10:14:00<1:03:39,  1.62it/s]
 91%|█████████ | 59351/65536 [10:14:01<1:05:05,  1.58it/s]
 91%|█████████ | 59352/65536 [10:14:01<1:05:16,  1.58it/s]
 91%|█████████ | 59353/65536 [10:14:02<1:04:56,  1.59it/s]
 91%|█████████ | 59354/65536 [10:14:03<1:02:59,  1.64it/s]
 91%|█████████ | 59355/65536 [10:14:03<1:03:13,  1.63it/s]
 91%|█████████ | 59356/65536 [10:14:04<1:04:05,  1.61it/s]
 91%|█████████ | 59357/65536 [10:14:05<1:05:09,  1.58it/s]
 91%|█████████ | 59358/65536 [10:14:05<1:05:29,  1.57it/s]
 91%|█████████ | 59359/65536 [10:14:06<1:06:29,  1.55it/s]
 91%|█████████ | 59360/65536 [10:14:07<1:05:16,  1.58it/s]
                                                          
{'loss': 1.5516, 'learning_rate': 1.8507407860903635e-07, 'epoch': 3664.2}
+
 91%|█████████ | 59360/65536 [10:14:07<1:05:16,  1.58it/s]
 91%|█████████ | 59361/65536 [10:14:07<1:06:07,  1.56it/s]
 91%|█████████ | 59362/65536 [10:14:08<1:04:22,  1.60it/s]
 91%|█████████ | 59363/65536 [10:14:08<1:03:33,  1.62it/s]
 91%|█████████ | 59364/65536 [10:14:09<1:02:16,  1.65it/s]
 91%|█████████ | 59365/65536 [10:14:10<1:02:56,  1.63it/s]
 91%|█████████ | 59366/65536 [10:14:10<1:02:27,  1.65it/s]
 91%|█████████ | 59367/65536 [10:14:11<1:02:21,  1.65it/s]
 91%|█████████ | 59368/65536 [10:14:11<1:03:16,  1.62it/s]
 91%|█████████ | 59369/65536 [10:14:12<1:02:34,  1.64it/s]
 91%|█████████ | 59370/65536 [10:14:13<1:02:29,  1.64it/s]
 91%|█████████ | 59371/65536 [10:14:13<1:04:27,  1.59it/s]
 91%|█████████ | 59372/65536 [10:14:14<1:02:58,  1.63it/s]
 91%|█████████ | 59373/65536 [10:14:14<1:03:21,  1.62it/s]
 91%|█████████ | 59374/65536 [10:14:15<1:05:23,  1.57it/s]
 91%|█████████ | 59375/65536 [10:14:16<1:05:52,  1.56it/s]
 91%|█████████ | 59376/65536 [10:14:16<1:04:42,  1.59it/s]
 91%|█████████ | 59377/65536 [10:14:17<1:05:04,  1.58it/s]
 91%|█████████ | 59378/65536 [10:14:18<1:03:51,  1.61it/s]
 91%|█████████ | 59379/65536 [10:14:18<1:02:56,  1.63it/s]
 91%|█████████ | 59380/65536 [10:14:19<1:04:46,  1.58it/s]
                                                          
{'loss': 1.5704, 'learning_rate': 1.8479857964981021e-07, 'epoch': 3665.43}
+
 91%|█████████ | 59380/65536 [10:14:19<1:04:46,  1.58it/s]
 91%|█████████ | 59381/65536 [10:14:20<1:06:09,  1.55it/s]
 91%|█████████ | 59382/65536 [10:14:20<1:06:43,  1.54it/s]
 91%|█████████ | 59383/65536 [10:14:21<1:05:39,  1.56it/s]
 91%|█████████ | 59384/65536 [10:14:21<1:04:38,  1.59it/s]
 91%|█████████ | 59385/65536 [10:14:22<1:03:00,  1.63it/s]
 91%|█████████ | 59386/65536 [10:14:23<1:02:29,  1.64it/s]
 91%|█████████ | 59387/65536 [10:14:23<1:02:56,  1.63it/s]
 91%|█████████ | 59388/65536 [10:14:24<1:02:37,  1.64it/s]
 91%|█████████ | 59389/65536 [10:14:25<1:02:48,  1.63it/s]
 91%|█████████ | 59390/65536 [10:14:25<1:04:28,  1.59it/s]
 91%|█████████ | 59391/65536 [10:14:26<1:06:09,  1.55it/s]
 91%|█████████ | 59392/65536 [10:14:27<1:05:51,  1.55it/s]
 91%|█████████ | 59393/65536 [10:14:27<1:05:23,  1.57it/s]
 91%|█████████ | 59394/65536 [10:14:28<1:04:26,  1.59it/s]
 91%|█████████ | 59395/65536 [10:14:28<1:03:18,  1.62it/s]
 91%|█████████ | 59396/65536 [10:14:29<1:03:50,  1.60it/s]
 91%|█████████ | 59397/65536 [10:14:30<1:04:23,  1.59it/s]
 91%|█████████ | 59398/65536 [10:14:30<1:03:34,  1.61it/s]
 91%|█████████ | 59399/65536 [10:14:31<1:02:50,  1.63it/s]
 91%|█████████ | 59400/65536 [10:14:31<1:02:08,  1.65it/s]
                                                          
{'loss': 1.5279, 'learning_rate': 1.8452308069058407e-07, 'epoch': 3666.67}
+
 91%|█████████ | 59400/65536 [10:14:31<1:02:08,  1.65it/s]
 91%|█████████ | 59401/65536 [10:14:32<1:02:31,  1.64it/s]
 91%|█████████ | 59402/65536 [10:14:33<1:04:17,  1.59it/s]
 91%|█████████ | 59403/65536 [10:14:33<1:02:44,  1.63it/s]
 91%|█████████ | 59404/65536 [10:14:34<1:03:01,  1.62it/s]
 91%|█████████ | 59405/65536 [10:14:35<1:03:07,  1.62it/s]
 91%|█████████ | 59406/65536 [10:14:35<1:05:27,  1.56it/s]
 91%|█████████ | 59407/65536 [10:14:36<1:05:09,  1.57it/s]
 91%|█████████ | 59408/65536 [10:14:36<1:04:26,  1.59it/s]
 91%|█████████ | 59409/65536 [10:14:37<1:04:48,  1.58it/s]
 91%|█████████ | 59410/65536 [10:14:38<1:04:11,  1.59it/s]
 91%|█████████ | 59411/65536 [10:14:38<1:06:26,  1.54it/s]
 91%|█████████ | 59412/65536 [10:14:39<1:05:49,  1.55it/s]
 91%|█████████ | 59413/65536 [10:14:40<1:04:43,  1.58it/s]
 91%|█████████ | 59414/65536 [10:14:40<1:03:55,  1.60it/s]
 91%|█████████ | 59415/65536 [10:14:41<1:03:23,  1.61it/s]
 91%|█████████ | 59416/65536 [10:14:42<1:03:36,  1.60it/s]
 91%|█████████ | 59417/65536 [10:14:42<1:02:59,  1.62it/s]
 91%|█████████ | 59418/65536 [10:14:43<1:02:37,  1.63it/s]
 91%|█████████ | 59419/65536 [10:14:43<1:01:55,  1.65it/s]
 91%|█████████ | 59420/65536 [10:14:44<1:04:53,  1.57it/s]
                                                          
{'loss': 1.4852, 'learning_rate': 1.8424758173135793e-07, 'epoch': 3667.9}
+
 91%|█████████ | 59420/65536 [10:14:44<1:04:53,  1.57it/s]
 91%|█████████ | 59421/65536 [10:14:45<1:03:32,  1.60it/s]
 91%|█████████ | 59422/65536 [10:14:45<1:08:16,  1.49it/s]
 91%|█████████ | 59423/65536 [10:14:46<1:08:38,  1.48it/s]
 91%|█████████ | 59424/65536 [10:14:47<1:08:06,  1.50it/s]
 91%|█████████ | 59425/65536 [10:14:47<1:05:45,  1.55it/s]
 91%|█████████ | 59426/65536 [10:14:48<1:04:07,  1.59it/s]
 91%|█████████ | 59427/65536 [10:14:48<1:02:33,  1.63it/s]
 91%|█████████ | 59428/65536 [10:14:49<1:01:40,  1.65it/s]
 91%|█████████ | 59429/65536 [10:14:50<1:01:21,  1.66it/s]
 91%|█████████ | 59430/65536 [10:14:50<1:03:56,  1.59it/s]
 91%|█████████ | 59431/65536 [10:14:51<1:02:20,  1.63it/s]
 91%|█████████ | 59432/65536 [10:14:52<1:03:47,  1.59it/s]
 91%|█████████ | 59433/65536 [10:14:52<1:03:48,  1.59it/s]
 91%|█████████ | 59434/65536 [10:14:53<1:02:58,  1.61it/s]
 91%|█████████ | 59435/65536 [10:14:53<1:03:14,  1.61it/s]
 91%|█████████ | 59436/65536 [10:14:54<1:04:08,  1.59it/s]
 91%|█████████ | 59437/65536 [10:14:55<1:03:06,  1.61it/s]
 91%|█████████ | 59438/65536 [10:14:55<1:04:36,  1.57it/s]
 91%|█████████ | 59439/65536 [10:14:56<1:04:08,  1.58it/s]
 91%|█████████ | 59440/65536 [10:14:57<1:03:36,  1.60it/s]
                                                          
{'loss': 1.5694, 'learning_rate': 1.8397208277213168e-07, 'epoch': 3669.14}
+
 91%|█████████ | 59440/65536 [10:14:57<1:03:36,  1.60it/s]
 91%|█████████ | 59441/65536 [10:14:57<1:04:52,  1.57it/s]
 91%|█████████ | 59442/65536 [10:14:58<1:04:26,  1.58it/s]
 91%|█████████ | 59443/65536 [10:14:58<1:02:54,  1.61it/s]
 91%|█████████ | 59444/65536 [10:14:59<1:04:36,  1.57it/s]
 91%|█████████ | 59445/65536 [10:15:00<1:04:35,  1.57it/s]
 91%|█████████ | 59446/65536 [10:15:00<1:03:39,  1.59it/s]
 91%|█████████ | 59447/65536 [10:15:01<1:03:07,  1.61it/s]
 91%|█████████ | 59448/65536 [10:15:02<1:03:06,  1.61it/s]
 91%|█████████ | 59449/65536 [10:15:02<1:03:10,  1.61it/s]
 91%|█████████ | 59450/65536 [10:15:03<1:01:45,  1.64it/s]
 91%|█████████ | 59451/65536 [10:15:03<1:01:17,  1.65it/s]
 91%|█████████ | 59452/65536 [10:15:04<1:02:29,  1.62it/s]
 91%|█████████ | 59453/65536 [10:15:05<1:02:17,  1.63it/s]
 91%|█████████ | 59454/65536 [10:15:05<1:02:11,  1.63it/s]
 91%|█████████ | 59455/65536 [10:15:06<1:05:51,  1.54it/s]
 91%|█████████ | 59456/65536 [10:15:07<1:04:09,  1.58it/s]
 91%|█████████ | 59457/65536 [10:15:07<1:03:30,  1.60it/s]
 91%|█████████ | 59458/65536 [10:15:08<1:02:35,  1.62it/s]
 91%|█████████ | 59459/65536 [10:15:08<1:02:09,  1.63it/s]
 91%|█████████ | 59460/65536 [10:15:09<1:01:52,  1.64it/s]
                                                          
{'loss': 1.5283, 'learning_rate': 1.8369658381290554e-07, 'epoch': 3670.37}
+
 91%|█████████ | 59460/65536 [10:15:09<1:01:52,  1.64it/s]
 91%|█████████ | 59461/65536 [10:15:10<1:02:01,  1.63it/s]
 91%|█████████ | 59462/65536 [10:15:10<1:02:06,  1.63it/s]
 91%|█████████ | 59463/65536 [10:15:11<1:01:52,  1.64it/s]
 91%|█████████ | 59464/65536 [10:15:11<1:01:49,  1.64it/s]
 91%|█████████ | 59465/65536 [10:15:12<1:02:21,  1.62it/s]
 91%|█████████ | 59466/65536 [10:15:13<1:03:16,  1.60it/s]
 91%|█████████ | 59467/65536 [10:15:13<1:03:21,  1.60it/s]
 91%|█████████ | 59468/65536 [10:15:14<1:03:43,  1.59it/s]
 91%|█████████ | 59469/65536 [10:15:15<1:04:05,  1.58it/s]
 91%|█████████ | 59470/65536 [10:15:15<1:02:33,  1.62it/s]
 91%|█████████ | 59471/65536 [10:15:16<1:05:56,  1.53it/s]
 91%|█████████ | 59472/65536 [10:15:17<1:05:27,  1.54it/s]
 91%|█████████ | 59473/65536 [10:15:17<1:04:57,  1.56it/s]
 91%|█████████ | 59474/65536 [10:15:18<1:05:24,  1.54it/s]
 91%|█████████ | 59475/65536 [10:15:19<1:05:10,  1.55it/s]
 91%|█████████ | 59476/65536 [10:15:19<1:05:29,  1.54it/s]
 91%|█████████ | 59477/65536 [10:15:20<1:03:23,  1.59it/s]
 91%|█████████ | 59478/65536 [10:15:20<1:02:29,  1.62it/s]
 91%|█████████ | 59479/65536 [10:15:21<1:02:26,  1.62it/s]
 91%|█████████ | 59480/65536 [10:15:22<1:01:55,  1.63it/s]
                                                          
{'loss': 1.5324, 'learning_rate': 1.834210848536794e-07, 'epoch': 3671.6}
+
 91%|█████████ | 59480/65536 [10:15:22<1:01:55,  1.63it/s]
 91%|█████████ | 59481/65536 [10:15:22<1:01:29,  1.64it/s]
 91%|█████████ | 59482/65536 [10:15:23<1:00:33,  1.67it/s]
 91%|█████████ | 59483/65536 [10:15:23<1:00:29,  1.67it/s]
 91%|█████████ | 59484/65536 [10:15:24<1:01:33,  1.64it/s]
 91%|█████████ | 59485/65536 [10:15:25<1:01:10,  1.65it/s]
 91%|█████████ | 59486/65536 [10:15:25<1:02:37,  1.61it/s]
 91%|█████████ | 59487/65536 [10:15:26<1:05:46,  1.53it/s]
 91%|█████████ | 59488/65536 [10:15:27<1:05:00,  1.55it/s]
 91%|█████████ | 59489/65536 [10:15:27<1:03:18,  1.59it/s]
 91%|█████████ | 59490/65536 [10:15:28<1:02:37,  1.61it/s]
 91%|█████████ | 59491/65536 [10:15:28<1:01:19,  1.64it/s]
 91%|█████████ | 59492/65536 [10:15:29<1:00:51,  1.66it/s]
 91%|█████████ | 59493/65536 [10:15:30<1:02:41,  1.61it/s]
 91%|█████████ | 59494/65536 [10:15:30<1:01:45,  1.63it/s]
 91%|█████████ | 59495/65536 [10:15:31<1:03:14,  1.59it/s]
 91%|█████████ | 59496/65536 [10:15:32<1:02:15,  1.62it/s]
 91%|█████████ | 59497/65536 [10:15:32<1:03:00,  1.60it/s]
 91%|█████████ | 59498/65536 [10:15:33<1:01:56,  1.62it/s]
 91%|█████████ | 59499/65536 [10:15:33<1:01:40,  1.63it/s]
 91%|█████████ | 59500/65536 [10:15:34<1:01:00,  1.65it/s]
                                                          
{'loss': 1.5737, 'learning_rate': 1.831455858944533e-07, 'epoch': 3672.84}
+
 91%|█████████ | 59500/65536 [10:15:34<1:01:00,  1.65it/s]
 91%|█████████ | 59501/65536 [10:15:35<1:01:31,  1.64it/s]
 91%|█████████ | 59502/65536 [10:15:35<1:02:20,  1.61it/s]
 91%|█████████ | 59503/65536 [10:15:36<1:03:53,  1.57it/s]
 91%|█████████ | 59504/65536 [10:15:37<1:04:37,  1.56it/s]
 91%|█████████ | 59505/65536 [10:15:37<1:03:30,  1.58it/s]
 91%|█████████ | 59506/65536 [10:15:38<1:02:44,  1.60it/s]
 91%|█████████ | 59507/65536 [10:15:38<1:02:01,  1.62it/s]
 91%|█████████ | 59508/65536 [10:15:39<1:01:09,  1.64it/s]
 91%|█████████ | 59509/65536 [10:15:40<1:00:22,  1.66it/s]
 91%|█████████ | 59510/65536 [10:15:40<1:01:01,  1.65it/s]
 91%|█████████ | 59511/65536 [10:15:41<1:02:11,  1.61it/s]
 91%|█████████ | 59512/65536 [10:15:41<1:02:04,  1.62it/s]
 91%|█████████ | 59513/65536 [10:15:42<1:01:10,  1.64it/s]
 91%|█████████ | 59514/65536 [10:15:43<1:02:35,  1.60it/s]
 91%|█████████ | 59515/65536 [10:15:43<1:01:34,  1.63it/s]
 91%|█████████ | 59516/65536 [10:15:44<1:01:40,  1.63it/s]
 91%|█████████ | 59517/65536 [10:15:44<1:02:37,  1.60it/s]
 91%|█████████ | 59518/65536 [10:15:45<1:01:48,  1.62it/s]
 91%|█████████ | 59519/65536 [10:15:46<1:04:13,  1.56it/s]
 91%|█████████ | 59520/65536 [10:15:46<1:02:21,  1.61it/s]
                                                          
{'loss': 1.5304, 'learning_rate': 1.8287008693522715e-07, 'epoch': 3674.07}
+
 91%|█████████ | 59520/65536 [10:15:46<1:02:21,  1.61it/s]
 91%|█████████ | 59521/65536 [10:15:47<1:01:38,  1.63it/s]
 91%|█████████ | 59522/65536 [10:15:48<1:03:14,  1.58it/s]
 91%|█████████ | 59523/65536 [10:15:48<1:01:57,  1.62it/s]
 91%|█████████ | 59524/65536 [10:15:49<1:02:48,  1.60it/s]
 91%|█████████ | 59525/65536 [10:15:50<1:03:16,  1.58it/s]
 91%|█████████ | 59526/65536 [10:15:50<1:03:27,  1.58it/s]
 91%|█████████ | 59527/65536 [10:15:51<1:02:14,  1.61it/s]
 91%|█████████ | 59528/65536 [10:15:51<1:02:51,  1.59it/s]
 91%|█████████ | 59529/65536 [10:15:52<1:02:36,  1.60it/s]
 91%|█████████ | 59530/65536 [10:15:53<1:01:44,  1.62it/s]
 91%|█████████ | 59531/65536 [10:15:53<1:02:55,  1.59it/s]
 91%|█████████ | 59532/65536 [10:15:54<1:02:26,  1.60it/s]
 91%|█████████ | 59533/65536 [10:15:54<1:01:42,  1.62it/s]
 91%|█████████ | 59534/65536 [10:15:55<1:00:33,  1.65it/s]
 91%|█████████ | 59535/65536 [10:15:56<1:00:37,  1.65it/s]
 91%|█████████ | 59536/65536 [10:15:56<1:01:53,  1.62it/s]
 91%|█████████ | 59537/65536 [10:15:57<1:01:22,  1.63it/s]
 91%|█████████ | 59538/65536 [10:15:58<1:01:03,  1.64it/s]
 91%|█████████ | 59539/65536 [10:15:58<1:01:39,  1.62it/s]
 91%|█████████ | 59540/65536 [10:15:59<1:00:59,  1.64it/s]
                                                          
{'loss': 1.5675, 'learning_rate': 1.82594587976001e-07, 'epoch': 3675.31}
+
 91%|█████████ | 59540/65536 [10:15:59<1:00:59,  1.64it/s]
 91%|█████████ | 59541/65536 [10:15:59<1:02:02,  1.61it/s]
 91%|█████████ | 59542/65536 [10:16:00<1:01:38,  1.62it/s]
 91%|█████████ | 59543/65536 [10:16:01<1:01:09,  1.63it/s]
 91%|█████████ | 59544/65536 [10:16:01<1:01:59,  1.61it/s]
 91%|█████████ | 59545/65536 [10:16:02<1:03:04,  1.58it/s]
 91%|█████████ | 59546/65536 [10:16:03<1:03:13,  1.58it/s]
 91%|█████████ | 59547/65536 [10:16:03<1:03:36,  1.57it/s]
 91%|█████████ | 59548/65536 [10:16:04<1:01:26,  1.62it/s]
 91%|█████████ | 59549/65536 [10:16:04<1:01:14,  1.63it/s]
 91%|█████████ | 59550/65536 [10:16:05<1:01:57,  1.61it/s]
 91%|█████████ | 59551/65536 [10:16:06<1:01:09,  1.63it/s]
 91%|█████████ | 59552/65536 [10:16:06<1:03:40,  1.57it/s]
 91%|█████████ | 59553/65536 [10:16:07<1:01:47,  1.61it/s]
 91%|█████████ | 59554/65536 [10:16:07<1:01:26,  1.62it/s]
 91%|█████████ | 59555/65536 [10:16:08<1:01:38,  1.62it/s]
 91%|█████████ | 59556/65536 [10:16:09<1:02:21,  1.60it/s]
 91%|█████████ | 59557/65536 [10:16:09<1:02:04,  1.61it/s]
 91%|█████████ | 59558/65536 [10:16:10<1:00:22,  1.65it/s]
 91%|█████████ | 59559/65536 [10:16:10<59:37,  1.67it/s]  
 91%|█████████ | 59560/65536 [10:16:11<1:01:09,  1.63it/s]
                                                          
{'loss': 1.574, 'learning_rate': 1.8231908901677486e-07, 'epoch': 3676.54}
+
 91%|█████████ | 59560/65536 [10:16:11<1:01:09,  1.63it/s]
 91%|█████████ | 59561/65536 [10:16:12<1:03:00,  1.58it/s]
 91%|█████████ | 59562/65536 [10:16:12<1:01:54,  1.61it/s]
 91%|█████████ | 59563/65536 [10:16:13<1:00:51,  1.64it/s]
 91%|█████████ | 59564/65536 [10:16:14<1:02:32,  1.59it/s]
 91%|█████████ | 59565/65536 [10:16:14<1:00:54,  1.63it/s]
 91%|█████████ | 59566/65536 [10:16:15<1:01:48,  1.61it/s]
 91%|█████████ | 59567/65536 [10:16:15<1:00:01,  1.66it/s]
 91%|█████████ | 59568/65536 [10:16:16<1:01:45,  1.61it/s]
 91%|█████████ | 59569/65536 [10:16:17<1:01:52,  1.61it/s]
 91%|█████████ | 59570/65536 [10:16:17<1:01:04,  1.63it/s]
 91%|█████████ | 59571/65536 [10:16:18<1:02:29,  1.59it/s]
 91%|█████████ | 59572/65536 [10:16:19<1:02:05,  1.60it/s]
 91%|█████████ | 59573/65536 [10:16:19<1:01:30,  1.62it/s]
 91%|█████████ | 59574/65536 [10:16:20<1:01:20,  1.62it/s]
 91%|█████████ | 59575/65536 [10:16:20<1:00:33,  1.64it/s]
 91%|█████████ | 59576/65536 [10:16:21<59:39,  1.67it/s]  
 91%|█████████ | 59577/65536 [10:16:22<59:31,  1.67it/s]
 91%|█████████ | 59578/65536 [10:16:22<59:01,  1.68it/s]
 91%|█████████ | 59579/65536 [10:16:23<59:36,  1.67it/s]
 91%|█████████ | 59580/65536 [10:16:23<1:01:42,  1.61it/s]
                                                          
{'loss': 1.5253, 'learning_rate': 1.8204359005754862e-07, 'epoch': 3677.78}
+
 91%|█████████ | 59580/65536 [10:16:23<1:01:42,  1.61it/s]
 91%|█████████ | 59581/65536 [10:16:24<1:01:49,  1.61it/s]
 91%|█████████ | 59582/65536 [10:16:25<1:03:40,  1.56it/s]
 91%|█████████ | 59583/65536 [10:16:25<1:02:39,  1.58it/s]
 91%|█████████ | 59584/65536 [10:16:26<1:05:37,  1.51it/s]
 91%|█████████ | 59585/65536 [10:16:27<1:04:23,  1.54it/s]
 91%|█████████ | 59586/65536 [10:16:27<1:03:14,  1.57it/s]
 91%|█████████ | 59587/65536 [10:16:28<1:02:31,  1.59it/s]
 91%|█████████ | 59588/65536 [10:16:29<1:02:49,  1.58it/s]
 91%|█████████ | 59589/65536 [10:16:29<1:02:35,  1.58it/s]
 91%|█████████ | 59590/65536 [10:16:30<1:01:36,  1.61it/s]
 91%|█████████ | 59591/65536 [10:16:30<1:01:00,  1.62it/s]
 91%|█████████ | 59592/65536 [10:16:31<1:02:27,  1.59it/s]
 91%|█████████ | 59593/65536 [10:16:32<1:02:24,  1.59it/s]
 91%|█████████ | 59594/65536 [10:16:32<1:01:15,  1.62it/s]
 91%|█████████ | 59595/65536 [10:16:33<1:01:41,  1.60it/s]
 91%|█████████ | 59596/65536 [10:16:34<1:01:15,  1.62it/s]
 91%|█████████ | 59597/65536 [10:16:34<1:00:59,  1.62it/s]
 91%|█████████ | 59598/65536 [10:16:35<1:01:03,  1.62it/s]
 91%|█████████ | 59599/65536 [10:16:35<1:00:11,  1.64it/s]
 91%|█████████ | 59600/65536 [10:16:36<1:01:43,  1.60it/s]
                                                          
{'loss': 1.4923, 'learning_rate': 1.817680910983225e-07, 'epoch': 3679.01}
+
 91%|█████████ | 59600/65536 [10:16:36<1:01:43,  1.60it/s]
 91%|█████████ | 59601/65536 [10:16:37<1:02:14,  1.59it/s]
 91%|█████████ | 59602/65536 [10:16:37<1:02:43,  1.58it/s]
 91%|█████████ | 59603/65536 [10:16:38<1:03:43,  1.55it/s]
 91%|█████████ | 59604/65536 [10:16:39<1:03:38,  1.55it/s]
 91%|█████████ | 59605/65536 [10:16:39<1:04:27,  1.53it/s]
 91%|█████████ | 59606/65536 [10:16:40<1:03:51,  1.55it/s]
 91%|█████████ | 59607/65536 [10:16:41<1:05:24,  1.51it/s]
 91%|█████████ | 59608/65536 [10:16:41<1:04:10,  1.54it/s]
 91%|█████████ | 59609/65536 [10:16:42<1:01:59,  1.59it/s]
 91%|█████████ | 59610/65536 [10:16:42<1:00:39,  1.63it/s]
 91%|█████████ | 59611/65536 [10:16:43<1:00:04,  1.64it/s]
 91%|█████████ | 59612/65536 [10:16:44<59:23,  1.66it/s]  
 91%|█████████ | 59613/65536 [10:16:44<59:16,  1.67it/s]
 91%|█████████ | 59614/65536 [10:16:45<1:00:17,  1.64it/s]
 91%|█████████ | 59615/65536 [10:16:45<59:47,  1.65it/s]  
 91%|█████████ | 59616/65536 [10:16:46<59:07,  1.67it/s]
 91%|█████████ | 59617/65536 [10:16:47<1:01:31,  1.60it/s]
 91%|█████████ | 59618/65536 [10:16:47<1:02:12,  1.59it/s]
 91%|█████████ | 59619/65536 [10:16:48<1:02:14,  1.58it/s]
 91%|█████████ | 59620/65536 [10:16:49<1:00:27,  1.63it/s]
                                                          
{'loss': 1.556, 'learning_rate': 1.8149259213909636e-07, 'epoch': 3680.25}
+
 91%|█████████ | 59620/65536 [10:16:49<1:00:27,  1.63it/s]
 91%|█████████ | 59621/65536 [10:16:49<59:25,  1.66it/s]  
 91%|█████████ | 59622/65536 [10:16:50<1:01:52,  1.59it/s]
 91%|█████████ | 59623/65536 [10:16:50<1:03:02,  1.56it/s]
 91%|█████████ | 59624/65536 [10:16:51<1:01:48,  1.59it/s]
 91%|█████████ | 59625/65536 [10:16:52<1:00:55,  1.62it/s]
 91%|█████████ | 59626/65536 [10:16:52<59:12,  1.66it/s]  
 91%|█████████ | 59627/65536 [10:16:53<1:00:30,  1.63it/s]
 91%|█████████ | 59628/65536 [10:16:53<1:00:31,  1.63it/s]
 91%|█████████ | 59629/65536 [10:16:54<1:00:38,  1.62it/s]
 91%|█████████ | 59630/65536 [10:16:55<59:51,  1.64it/s]  
 91%|█████████ | 59631/65536 [10:16:55<1:00:31,  1.63it/s]
 91%|█████████ | 59632/65536 [10:16:56<1:01:22,  1.60it/s]
 91%|█████████ | 59633/65536 [10:16:57<1:02:30,  1.57it/s]
 91%|█████████ | 59634/65536 [10:16:57<1:01:28,  1.60it/s]
 91%|█████████ | 59635/65536 [10:16:58<1:00:52,  1.62it/s]
 91%|█████████ | 59636/65536 [10:16:58<59:42,  1.65it/s]  
 91%|█████████ | 59637/65536 [10:16:59<59:59,  1.64it/s]
 91%|█████████ | 59638/65536 [10:17:00<59:46,  1.64it/s]
 91%|█████████ | 59639/65536 [10:17:00<59:10,  1.66it/s]
 91%|█████████ | 59640/65536 [10:17:01<1:00:16,  1.63it/s]
                                                          
{'loss': 1.5306, 'learning_rate': 1.8121709317987022e-07, 'epoch': 3681.48}
+
 91%|█████████ | 59640/65536 [10:17:01<1:00:16,  1.63it/s]
 91%|█████████ | 59641/65536 [10:17:02<1:01:48,  1.59it/s]
 91%|█████████ | 59642/65536 [10:17:02<1:01:55,  1.59it/s]
 91%|█████████ | 59643/65536 [10:17:03<1:01:16,  1.60it/s]
 91%|█████████ | 59644/65536 [10:17:03<59:34,  1.65it/s]  
 91%|█████████ | 59645/65536 [10:17:04<59:24,  1.65it/s]
 91%|█████████ | 59646/65536 [10:17:05<58:55,  1.67it/s]
 91%|█████████ | 59647/65536 [10:17:05<58:40,  1.67it/s]
 91%|█████████ | 59648/65536 [10:17:06<1:00:58,  1.61it/s]
 91%|█████████ | 59649/65536 [10:17:06<1:02:32,  1.57it/s]
 91%|█████████ | 59650/65536 [10:17:07<1:01:52,  1.59it/s]
 91%|█████████ | 59651/65536 [10:17:08<1:02:19,  1.57it/s]
 91%|█████████ | 59652/65536 [10:17:08<1:02:46,  1.56it/s]
 91%|█████████ | 59653/65536 [10:17:09<1:01:34,  1.59it/s]
 91%|█████████ | 59654/65536 [10:17:10<1:00:13,  1.63it/s]
 91%|█████████ | 59655/65536 [10:17:10<1:01:01,  1.61it/s]
 91%|█████████ | 59656/65536 [10:17:11<1:00:05,  1.63it/s]
 91%|█████████ | 59657/65536 [10:17:11<1:00:18,  1.62it/s]
 91%|█████████ | 59658/65536 [10:17:12<1:00:09,  1.63it/s]
 91%|█████████ | 59659/65536 [10:17:13<1:01:05,  1.60it/s]
 91%|█████████ | 59660/65536 [10:17:13<1:01:47,  1.59it/s]
                                                          
{'loss': 1.559, 'learning_rate': 1.8094159422064408e-07, 'epoch': 3682.72}
+
 91%|█████████ | 59660/65536 [10:17:13<1:01:47,  1.59it/s]
 91%|█████████ | 59661/65536 [10:17:14<1:00:47,  1.61it/s]
 91%|█████████ | 59662/65536 [10:17:15<1:01:21,  1.60it/s]
 91%|█████████ | 59663/65536 [10:17:15<1:00:38,  1.61it/s]
 91%|█████████ | 59664/65536 [10:17:16<1:00:19,  1.62it/s]
 91%|█████████ | 59665/65536 [10:17:16<1:02:54,  1.56it/s]
 91%|█████████ | 59666/65536 [10:17:17<1:01:12,  1.60it/s]
 91%|█████████ | 59667/65536 [10:17:18<1:01:38,  1.59it/s]
 91%|█████████ | 59668/65536 [10:17:18<1:02:07,  1.57it/s]
 91%|█████████ | 59669/65536 [10:17:19<1:01:12,  1.60it/s]
 91%|█████████ | 59670/65536 [10:17:20<1:00:39,  1.61it/s]
 91%|█████████ | 59671/65536 [10:17:20<58:54,  1.66it/s]  
 91%|█████████ | 59672/65536 [10:17:21<59:47,  1.63it/s]
 91%|█████████ | 59673/65536 [10:17:21<1:00:19,  1.62it/s]
 91%|█████████ | 59674/65536 [10:17:22<59:37,  1.64it/s]  
 91%|█████████ | 59675/65536 [10:17:23<1:00:07,  1.62it/s]
 91%|█████████ | 59676/65536 [10:17:23<1:00:34,  1.61it/s]
 91%|█████████ | 59677/65536 [10:17:24<59:56,  1.63it/s]  
 91%|█████████ | 59678/65536 [10:17:24<59:46,  1.63it/s]
 91%|█████████ | 59679/65536 [10:17:25<59:16,  1.65it/s]
 91%|█████████ | 59680/65536 [10:17:26<59:22,  1.64it/s]
                                                        
{'loss': 1.5591, 'learning_rate': 1.8066609526141794e-07, 'epoch': 3683.95}
+
 91%|█████████ | 59680/65536 [10:17:26<59:22,  1.64it/s]
 91%|█████████ | 59681/65536 [10:17:26<1:03:19,  1.54it/s]
 91%|█████████ | 59682/65536 [10:17:27<1:03:36,  1.53it/s]
 91%|█████████ | 59683/65536 [10:17:28<1:01:05,  1.60it/s]
 91%|█████████ | 59684/65536 [10:17:28<1:01:18,  1.59it/s]
 91%|█████████ | 59685/65536 [10:17:29<1:00:39,  1.61it/s]
 91%|█████████ | 59686/65536 [10:17:29<59:54,  1.63it/s]  
 91%|█████████ | 59687/65536 [10:17:30<1:01:00,  1.60it/s]
 91%|█████████ | 59688/65536 [10:17:31<1:00:44,  1.60it/s]
 91%|█████████ | 59689/65536 [10:17:31<59:04,  1.65it/s]  
 91%|█████████ | 59690/65536 [10:17:32<1:00:45,  1.60it/s]
 91%|█████████ | 59691/65536 [10:17:33<59:56,  1.63it/s]  
 91%|█████████ | 59692/65536 [10:17:33<58:55,  1.65it/s]
 91%|█████████ | 59693/65536 [10:17:34<1:00:41,  1.60it/s]
 91%|█████████ | 59694/65536 [10:17:34<1:00:01,  1.62it/s]
 91%|█████████ | 59695/65536 [10:17:35<59:44,  1.63it/s]  
 91%|█████████ | 59696/65536 [10:17:36<59:46,  1.63it/s]
 91%|█████████ | 59697/65536 [10:17:36<1:00:25,  1.61it/s]
 91%|█████████ | 59698/65536 [10:17:37<1:01:37,  1.58it/s]
 91%|█████████ | 59699/65536 [10:17:38<1:01:29,  1.58it/s]
 91%|█████████ | 59700/65536 [10:17:38<1:00:52,  1.60it/s]
                                                          
{'loss': 1.5382, 'learning_rate': 1.803905963021918e-07, 'epoch': 3685.19}
+
 91%|█████████ | 59700/65536 [10:17:38<1:00:52,  1.60it/s]
 91%|█████████ | 59701/65536 [10:17:39<1:00:06,  1.62it/s]
 91%|█████████ | 59702/65536 [10:17:39<59:56,  1.62it/s]  
 91%|█████████ | 59703/65536 [10:17:40<59:14,  1.64it/s]
 91%|█████████ | 59704/65536 [10:17:41<1:00:06,  1.62it/s]
 91%|█████████ | 59705/65536 [10:17:41<59:45,  1.63it/s]  
 91%|█████████ | 59706/65536 [10:17:42<59:26,  1.63it/s]
 91%|█████████ | 59707/65536 [10:17:42<59:17,  1.64it/s]
 91%|█████████ | 59708/65536 [10:17:43<1:01:35,  1.58it/s]
 91%|█████████ | 59709/65536 [10:17:44<1:01:01,  1.59it/s]
 91%|█████████ | 59710/65536 [10:17:44<1:00:43,  1.60it/s]
 91%|█████████ | 59711/65536 [10:17:45<1:00:02,  1.62it/s]
 91%|█████████ | 59712/65536 [10:17:46<1:00:05,  1.62it/s]
 91%|█████████ | 59713/65536 [10:17:46<1:01:06,  1.59it/s]
 91%|█████████ | 59714/65536 [10:17:47<1:01:48,  1.57it/s]
 91%|█████████ | 59715/65536 [10:17:48<1:02:08,  1.56it/s]
 91%|█████████ | 59716/65536 [10:17:48<1:01:29,  1.58it/s]
 91%|█████████ | 59717/65536 [10:17:49<1:00:53,  1.59it/s]
 91%|█████████ | 59718/65536 [10:17:49<1:01:42,  1.57it/s]
 91%|█████████ | 59719/65536 [10:17:50<1:00:51,  1.59it/s]
 91%|█████████ | 59720/65536 [10:17:51<1:01:55,  1.57it/s]
                                                          
{'loss': 1.5264, 'learning_rate': 1.8011509734296555e-07, 'epoch': 3686.42}
+
 91%|█████████ | 59720/65536 [10:17:51<1:01:55,  1.57it/s]
 91%|█████████ | 59721/65536 [10:17:51<1:01:39,  1.57it/s]
 91%|█████████ | 59722/65536 [10:17:52<1:02:37,  1.55it/s]
 91%|█████████ | 59723/65536 [10:17:53<1:03:04,  1.54it/s]
 91%|█████████ | 59724/65536 [10:17:53<1:03:56,  1.51it/s]
 91%|█████████ | 59725/65536 [10:17:54<1:02:34,  1.55it/s]
 91%|█████████ | 59726/65536 [10:17:55<1:01:37,  1.57it/s]
 91%|█████████ | 59727/65536 [10:17:55<1:01:12,  1.58it/s]
 91%|█████████ | 59728/65536 [10:17:56<59:58,  1.61it/s]  
 91%|█████████ | 59729/65536 [10:17:56<1:00:11,  1.61it/s]
 91%|█████████ | 59730/65536 [10:17:57<1:01:45,  1.57it/s]
 91%|█████████ | 59731/65536 [10:17:58<1:02:19,  1.55it/s]
 91%|█████████ | 59732/65536 [10:17:58<1:02:03,  1.56it/s]
 91%|█████████ | 59733/65536 [10:17:59<1:01:45,  1.57it/s]
 91%|█████████ | 59734/65536 [10:18:00<1:00:34,  1.60it/s]
 91%|█████████ | 59735/65536 [10:18:00<59:11,  1.63it/s]  
 91%|█████████ | 59736/65536 [10:18:01<57:40,  1.68it/s]
 91%|█████████ | 59737/65536 [10:18:01<57:52,  1.67it/s]
 91%|█████████ | 59738/65536 [10:18:02<59:09,  1.63it/s]
 91%|█████████ | 59739/65536 [10:18:03<59:40,  1.62it/s]
 91%|█████████ | 59740/65536 [10:18:03<59:11,  1.63it/s]
                                                        
{'loss': 1.5939, 'learning_rate': 1.798395983837394e-07, 'epoch': 3687.65}
+
 91%|█████████ | 59740/65536 [10:18:03<59:11,  1.63it/s]
 91%|█████████ | 59741/65536 [10:18:04<1:00:13,  1.60it/s]
 91%|█████████ | 59742/65536 [10:18:04<58:04,  1.66it/s]  
 91%|█████████ | 59743/65536 [10:18:05<58:37,  1.65it/s]
 91%|█████████ | 59744/65536 [10:18:06<58:26,  1.65it/s]
 91%|█████████ | 59745/65536 [10:18:06<59:36,  1.62it/s]
 91%|█████████ | 59746/65536 [10:18:07<1:00:37,  1.59it/s]
 91%|█████████ | 59747/65536 [10:18:08<59:18,  1.63it/s]  
 91%|█████████ | 59748/65536 [10:18:08<1:00:09,  1.60it/s]
 91%|█████████ | 59749/65536 [10:18:09<59:06,  1.63it/s]  
 91%|█████████ | 59750/65536 [10:18:09<59:54,  1.61it/s]
 91%|█████████ | 59751/65536 [10:18:10<1:01:23,  1.57it/s]
 91%|█████████ | 59752/65536 [10:18:11<1:01:25,  1.57it/s]
 91%|█████████ | 59753/65536 [10:18:11<59:50,  1.61it/s]  
 91%|█████████ | 59754/65536 [10:18:12<59:50,  1.61it/s]
 91%|█████████ | 59755/65536 [10:18:13<59:15,  1.63it/s]
 91%|█████████ | 59756/65536 [10:18:13<58:48,  1.64it/s]
 91%|█████████ | 59757/65536 [10:18:14<57:54,  1.66it/s]
 91%|█████████ | 59758/65536 [10:18:14<57:45,  1.67it/s]
 91%|█████████ | 59759/65536 [10:18:15<58:39,  1.64it/s]
 91%|█████████ | 59760/65536 [10:18:16<59:10,  1.63it/s]
                                                        
{'loss': 1.55, 'learning_rate': 1.7956409942451327e-07, 'epoch': 3688.89}
+
 91%|█████████ | 59760/65536 [10:18:16<59:10,  1.63it/s]
 91%|█████████ | 59761/65536 [10:18:16<59:36,  1.61it/s]
 91%|█████████ | 59762/65536 [10:18:17<1:01:03,  1.58it/s]
 91%|█████████ | 59763/65536 [10:18:17<59:29,  1.62it/s]  
 91%|█████████ | 59764/65536 [10:18:18<58:55,  1.63it/s]
 91%|█████████ | 59765/65536 [10:18:19<59:17,  1.62it/s]
 91%|█████████ | 59766/65536 [10:18:19<59:10,  1.63it/s]
 91%|█████████ | 59767/65536 [10:18:20<58:17,  1.65it/s]
 91%|█████████ | 59768/65536 [10:18:20<58:20,  1.65it/s]
 91%|█████████ | 59769/65536 [10:18:21<57:49,  1.66it/s]
 91%|█████████ | 59770/65536 [10:18:22<58:44,  1.64it/s]
 91%|█████████ | 59771/65536 [10:18:22<1:00:32,  1.59it/s]
 91%|█████████ | 59772/65536 [10:18:23<1:03:31,  1.51it/s]
 91%|█████████ | 59773/65536 [10:18:24<1:02:55,  1.53it/s]
 91%|█████████ | 59774/65536 [10:18:24<1:05:04,  1.48it/s]
 91%|█████████ | 59775/65536 [10:18:25<1:04:17,  1.49it/s]
 91%|█████████ | 59776/65536 [10:18:26<1:03:38,  1.51it/s]
 91%|█████████ | 59777/65536 [10:18:26<1:02:10,  1.54it/s]
 91%|█████████ | 59778/65536 [10:18:27<1:01:57,  1.55it/s]
 91%|█████████ | 59779/65536 [10:18:28<1:04:02,  1.50it/s]
 91%|█████████ | 59780/65536 [10:18:28<1:02:03,  1.55it/s]
                                                          
{'loss': 1.5009, 'learning_rate': 1.7928860046528713e-07, 'epoch': 3690.12}
+
 91%|█████████ | 59780/65536 [10:18:28<1:02:03,  1.55it/s]
 91%|█████████ | 59781/65536 [10:18:29<1:02:03,  1.55it/s]
 91%|█████████ | 59782/65536 [10:18:30<1:01:09,  1.57it/s]
 91%|█████████ | 59783/65536 [10:18:30<1:00:20,  1.59it/s]
 91%|█████████ | 59784/65536 [10:18:31<59:37,  1.61it/s]  
 91%|█████████ | 59785/65536 [10:18:31<58:57,  1.63it/s]
 91%|█████████ | 59786/65536 [10:18:32<59:10,  1.62it/s]
 91%|█████████ | 59787/65536 [10:18:33<58:26,  1.64it/s]
 91%|█████████ | 59788/65536 [10:18:33<58:11,  1.65it/s]
 91%|█████████ | 59789/65536 [10:18:34<58:30,  1.64it/s]
 91%|█████████ | 59790/65536 [10:18:35<1:00:04,  1.59it/s]
 91%|█████████ | 59791/65536 [10:18:35<59:02,  1.62it/s]  
 91%|█████████ | 59792/65536 [10:18:36<59:08,  1.62it/s]
 91%|█████████ | 59793/65536 [10:18:36<57:47,  1.66it/s]
 91%|█████████ | 59794/65536 [10:18:37<59:06,  1.62it/s]
 91%|█████████ | 59795/65536 [10:18:38<1:01:04,  1.57it/s]
 91%|█████████ | 59796/65536 [10:18:38<59:08,  1.62it/s]  
 91%|█████████ | 59797/65536 [10:18:39<58:53,  1.62it/s]
 91%|█████████ | 59798/65536 [10:18:39<1:00:41,  1.58it/s]
 91%|█████████ | 59799/65536 [10:18:40<58:36,  1.63it/s]  
 91%|█████████ | 59800/65536 [10:18:41<58:09,  1.64it/s]
                                                        
{'loss': 1.5513, 'learning_rate': 1.79013101506061e-07, 'epoch': 3691.36}
+
 91%|█████████ | 59800/65536 [10:18:41<58:09,  1.64it/s]
 91%|█████████ | 59801/65536 [10:18:41<59:36,  1.60it/s]
 91%|█████████▏| 59802/65536 [10:18:42<58:28,  1.63it/s]
 91%|█████████▏| 59803/65536 [10:18:42<57:54,  1.65it/s]
 91%|█████████▏| 59804/65536 [10:18:43<57:44,  1.65it/s]
 91%|█████████▏| 59805/65536 [10:18:44<58:52,  1.62it/s]
 91%|█████████▏| 59806/65536 [10:18:44<59:15,  1.61it/s]
 91%|█████████▏| 59807/65536 [10:18:45<57:47,  1.65it/s]
 91%|█████████▏| 59808/65536 [10:18:46<57:36,  1.66it/s]
 91%|█████████▏| 59809/65536 [10:18:46<57:44,  1.65it/s]
 91%|█████████▏| 59810/65536 [10:18:47<58:48,  1.62it/s]
 91%|█████████▏| 59811/65536 [10:18:47<1:00:38,  1.57it/s]
 91%|█████████▏| 59812/65536 [10:18:48<58:59,  1.62it/s]  
 91%|█████████▏| 59813/65536 [10:18:49<59:42,  1.60it/s]
 91%|█████████▏| 59814/65536 [10:18:49<1:00:08,  1.59it/s]
 91%|█████████▏| 59815/65536 [10:18:50<59:53,  1.59it/s]  
 91%|█████████▏| 59816/65536 [10:18:51<58:36,  1.63it/s]
 91%|█████████▏| 59817/65536 [10:18:51<58:52,  1.62it/s]
 91%|█████████▏| 59818/65536 [10:18:52<59:21,  1.61it/s]
 91%|█████████▏| 59819/65536 [10:18:52<58:50,  1.62it/s]
 91%|█████████▏| 59820/65536 [10:18:53<59:27,  1.60it/s]
                                                        
{'loss': 1.5661, 'learning_rate': 1.7873760254683484e-07, 'epoch': 3692.59}
+
 91%|█████████▏| 59820/65536 [10:18:53<59:27,  1.60it/s]
 91%|█████████▏| 59821/65536 [10:18:54<58:39,  1.62it/s]
 91%|█████████▏| 59822/65536 [10:18:54<59:47,  1.59it/s]
 91%|█████████▏| 59823/65536 [10:18:55<1:00:36,  1.57it/s]
 91%|█████████▏| 59824/65536 [10:18:56<1:00:27,  1.57it/s]
 91%|█████████▏| 59825/65536 [10:18:56<59:21,  1.60it/s]  
 91%|█████████▏| 59826/65536 [10:18:57<58:47,  1.62it/s]
 91%|█████████▏| 59827/65536 [10:18:57<1:00:45,  1.57it/s]
 91%|█████████▏| 59828/65536 [10:18:58<1:01:17,  1.55it/s]
 91%|█████████▏| 59829/65536 [10:18:59<59:40,  1.59it/s]  
 91%|█████████▏| 59830/65536 [10:18:59<59:18,  1.60it/s]
 91%|█████████▏| 59831/65536 [10:19:00<59:01,  1.61it/s]
 91%|█████████▏| 59832/65536 [10:19:01<59:26,  1.60it/s]
 91%|█████████▏| 59833/65536 [10:19:01<58:55,  1.61it/s]
 91%|█████████▏| 59834/65536 [10:19:02<59:28,  1.60it/s]
 91%|█████████▏| 59835/65536 [10:19:02<57:55,  1.64it/s]
 91%|█████████▏| 59836/65536 [10:19:03<57:41,  1.65it/s]
 91%|█████████▏| 59837/65536 [10:19:04<58:13,  1.63it/s]
 91%|█████████▏| 59838/65536 [10:19:04<57:50,  1.64it/s]
 91%|█████████▏| 59839/65536 [10:19:05<58:52,  1.61it/s]
 91%|█████████▏| 59840/65536 [10:19:05<58:37,  1.62it/s]
                                                        
{'loss': 1.5401, 'learning_rate': 1.784621035876086e-07, 'epoch': 3693.83}
+
 91%|█████████▏| 59840/65536 [10:19:05<58:37,  1.62it/s]
 91%|█████████▏| 59841/65536 [10:19:06<58:09,  1.63it/s]
 91%|█████████▏| 59842/65536 [10:19:07<58:51,  1.61it/s]
 91%|█████████▏| 59843/65536 [10:19:07<1:02:26,  1.52it/s]
 91%|█████████▏| 59844/65536 [10:19:08<1:02:20,  1.52it/s]
 91%|█████████▏| 59845/65536 [10:19:09<1:01:31,  1.54it/s]
 91%|█████████▏| 59846/65536 [10:19:09<1:00:45,  1.56it/s]
 91%|█████████▏| 59847/65536 [10:19:10<59:58,  1.58it/s]  
 91%|█████████▏| 59848/65536 [10:19:11<59:05,  1.60it/s]
 91%|█████████▏| 59849/65536 [10:19:11<57:24,  1.65it/s]
 91%|█████████▏| 59850/65536 [10:19:12<58:32,  1.62it/s]
 91%|█████████▏| 59851/65536 [10:19:12<58:20,  1.62it/s]
 91%|█████████▏| 59852/65536 [10:19:13<58:12,  1.63it/s]
 91%|█████████▏| 59853/65536 [10:19:14<58:37,  1.62it/s]
 91%|█████████▏| 59854/65536 [10:19:14<57:56,  1.63it/s]
 91%|█████████▏| 59855/65536 [10:19:15<57:49,  1.64it/s]
 91%|█████████▏| 59856/65536 [10:19:15<58:02,  1.63it/s]
 91%|█████████▏| 59857/65536 [10:19:16<58:02,  1.63it/s]
 91%|█████████▏| 59858/65536 [10:19:17<58:24,  1.62it/s]
 91%|█████████▏| 59859/65536 [10:19:17<57:28,  1.65it/s]
 91%|█████████▏| 59860/65536 [10:19:18<59:15,  1.60it/s]
                                                        
{'loss': 1.5403, 'learning_rate': 1.7818660462838248e-07, 'epoch': 3695.06}
+
 91%|█████████▏| 59860/65536 [10:19:18<59:15,  1.60it/s]
 91%|█████████▏| 59861/65536 [10:19:19<58:44,  1.61it/s]
 91%|█████████▏| 59862/65536 [10:19:19<59:05,  1.60it/s]
 91%|█████████▏| 59863/65536 [10:19:20<59:17,  1.59it/s]
 91%|█████████▏| 59864/65536 [10:19:21<1:00:23,  1.57it/s]
 91%|█████████▏| 59865/65536 [10:19:21<59:11,  1.60it/s]  
 91%|█████████▏| 59866/65536 [10:19:22<58:01,  1.63it/s]
 91%|█████████▏| 59867/65536 [10:19:22<57:36,  1.64it/s]
 91%|█████████▏| 59868/65536 [10:19:23<56:54,  1.66it/s]
 91%|█████████▏| 59869/65536 [10:19:24<57:43,  1.64it/s]
 91%|█████████▏| 59870/65536 [10:19:24<56:57,  1.66it/s]
 91%|█████████▏| 59871/65536 [10:19:25<58:34,  1.61it/s]
 91%|█████████▏| 59872/65536 [10:19:25<57:54,  1.63it/s]
 91%|█████████▏| 59873/65536 [10:19:26<58:39,  1.61it/s]
 91%|█████████▏| 59874/65536 [10:19:27<58:25,  1.62it/s]
 91%|█████████▏| 59875/65536 [10:19:27<58:04,  1.62it/s]
 91%|█████████▏| 59876/65536 [10:19:28<59:37,  1.58it/s]
 91%|█████████▏| 59877/65536 [10:19:28<58:18,  1.62it/s]
 91%|█████████▏| 59878/65536 [10:19:29<58:01,  1.62it/s]
 91%|█████████▏| 59879/65536 [10:19:30<59:14,  1.59it/s]
 91%|█████████▏| 59880/65536 [10:19:30<58:54,  1.60it/s]
                                                        
{'loss': 1.5321, 'learning_rate': 1.7791110566915634e-07, 'epoch': 3696.3}
+
 91%|████████��▏| 59880/65536 [10:19:30<58:54,  1.60it/s]
 91%|█████████▏| 59881/65536 [10:19:31<57:42,  1.63it/s]
 91%|█████████▏| 59882/65536 [10:19:32<57:14,  1.65it/s]
 91%|█████████▏| 59883/65536 [10:19:32<57:58,  1.63it/s]
 91%|█████████▏| 59884/65536 [10:19:33<58:23,  1.61it/s]
 91%|█████████▏| 59885/65536 [10:19:33<59:31,  1.58it/s]
 91%|█████████▏| 59886/65536 [10:19:34<59:08,  1.59it/s]
 91%|█████████▏| 59887/65536 [10:19:35<58:08,  1.62it/s]
 91%|█████████▏| 59888/65536 [10:19:35<57:48,  1.63it/s]
 91%|█████████▏| 59889/65536 [10:19:36<57:10,  1.65it/s]
 91%|█████████▏| 59890/65536 [10:19:36<56:33,  1.66it/s]
 91%|█████████▏| 59891/65536 [10:19:37<56:55,  1.65it/s]
 91%|█████████▏| 59892/65536 [10:19:38<1:01:58,  1.52it/s]
 91%|█████████▏| 59893/65536 [10:19:38<1:01:09,  1.54it/s]
 91%|█████████▏| 59894/65536 [10:19:39<1:01:57,  1.52it/s]
 91%|█████████▏| 59895/65536 [10:19:40<1:01:15,  1.53it/s]
 91%|█████████▏| 59896/65536 [10:19:40<1:00:57,  1.54it/s]
 91%|█████████▏| 59897/65536 [10:19:41<59:11,  1.59it/s]  
 91%|█████████▏| 59898/65536 [10:19:42<58:10,  1.62it/s]
 91%|█████████▏| 59899/65536 [10:19:42<58:02,  1.62it/s]
 91%|█████████▏| 59900/65536 [10:19:43<58:50,  1.60it/s]
                                                        
{'loss': 1.5369, 'learning_rate': 1.776356067099302e-07, 'epoch': 3697.53}
+
 91%|█████████▏| 59900/65536 [10:19:43<58:50,  1.60it/s]
 91%|█████████▏| 59901/65536 [10:19:43<57:31,  1.63it/s]
 91%|█████████▏| 59902/65536 [10:19:44<57:23,  1.64it/s]
 91%|█████████▏| 59903/65536 [10:19:45<56:50,  1.65it/s]
 91%|█████████▏| 59904/65536 [10:19:45<56:54,  1.65it/s]
 91%|█████████▏| 59905/65536 [10:19:46<58:04,  1.62it/s]
 91%|█████████▏| 59906/65536 [10:19:46<56:28,  1.66it/s]
 91%|█████████▏| 59907/65536 [10:19:47<56:18,  1.67it/s]
 91%|█████████▏| 59908/65536 [10:19:48<57:12,  1.64it/s]
 91%|█████████▏| 59909/65536 [10:19:48<56:14,  1.67it/s]
 91%|█████████▏| 59910/65536 [10:19:49<54:51,  1.71it/s]
 91%|█████████▏| 59911/65536 [10:19:49<54:42,  1.71it/s]
 91%|█████████▏| 59912/65536 [10:19:50<55:51,  1.68it/s]
 91%|█████████▏| 59913/65536 [10:19:51<56:58,  1.64it/s]
 91%|█████████▏| 59914/65536 [10:19:51<56:47,  1.65it/s]
 91%|█████████▏| 59915/65536 [10:19:52<56:09,  1.67it/s]
 91%|█████████▏| 59916/65536 [10:19:52<55:36,  1.68it/s]
 91%|█████████▏| 59917/65536 [10:19:53<58:30,  1.60it/s]
 91%|█████████▏| 59918/65536 [10:19:54<58:36,  1.60it/s]
 91%|█████████▏| 59919/65536 [10:19:54<58:17,  1.61it/s]
 91%|█████████▏| 59920/65536 [10:19:55<58:12,  1.61it/s]
                                                        
{'loss': 1.5659, 'learning_rate': 1.7736010775070406e-07, 'epoch': 3698.77}
+
 91%|█████████▏| 59920/65536 [10:19:55<58:12,  1.61it/s]
 91%|█████████▏| 59921/65536 [10:19:56<58:55,  1.59it/s]
 91%|█████████▏| 59922/65536 [10:19:56<59:25,  1.57it/s]
 91%|█████████▏| 59923/65536 [10:19:57<58:12,  1.61it/s]
 91%|█████████▏| 59924/65536 [10:19:58<1:02:21,  1.50it/s]
 91%|█████████▏| 59925/65536 [10:19:58<1:01:08,  1.53it/s]
 91%|█████████▏| 59926/65536 [10:19:59<1:00:23,  1.55it/s]
 91%|█████████▏| 59927/65536 [10:20:00<59:46,  1.56it/s]  
 91%|█████████▏| 59928/65536 [10:20:00<58:55,  1.59it/s]
 91%|█████████▏| 59929/65536 [10:20:01<58:56,  1.59it/s]
 91%|█████████▏| 59930/65536 [10:20:01<57:41,  1.62it/s]
 91%|█████████▏| 59931/65536 [10:20:02<58:32,  1.60it/s]
 91%|█████████▏| 59932/65536 [10:20:03<57:39,  1.62it/s]
 91%|█████████▏| 59933/65536 [10:20:03<57:09,  1.63it/s]
 91%|█████████▏| 59934/65536 [10:20:04<57:06,  1.63it/s]
 91%|█████████▏| 59935/65536 [10:20:04<56:49,  1.64it/s]
 91%|█████████▏| 59936/65536 [10:20:05<59:36,  1.57it/s]
 91%|█████████▏| 59937/65536 [10:20:06<58:46,  1.59it/s]
 91%|█████████▏| 59938/65536 [10:20:06<58:03,  1.61it/s]
 91%|█████████▏| 59939/65536 [10:20:07<57:22,  1.63it/s]
 91%|█████████▏| 59940/65536 [10:20:08<56:46,  1.64it/s]
                                                        
{'loss': 1.547, 'learning_rate': 1.7708460879147792e-07, 'epoch': 3700.0}
+
 91%|█████████▏| 59940/65536 [10:20:08<56:46,  1.64it/s]
 91%|█████████▏| 59941/65536 [10:20:08<57:38,  1.62it/s]
 91%|█████████▏| 59942/65536 [10:20:09<56:18,  1.66it/s]
 91%|█████████▏| 59943/65536 [10:20:09<56:25,  1.65it/s]
 91%|█████████▏| 59944/65536 [10:20:10<56:15,  1.66it/s]
 91%|█████████▏| 59945/65536 [10:20:11<56:17,  1.66it/s]
 91%|█████████▏| 59946/65536 [10:20:11<57:06,  1.63it/s]
 91%|█████████▏| 59947/65536 [10:20:12<57:26,  1.62it/s]
 91%|█████████▏| 59948/65536 [10:20:12<58:18,  1.60it/s]
 91%|█████████▏| 59949/65536 [10:20:13<59:03,  1.58it/s]
 91%|█████████▏| 59950/65536 [10:20:14<1:00:11,  1.55it/s]
 91%|█████████▏| 59951/65536 [10:20:15<1:02:44,  1.48it/s]
 91%|█████████▏| 59952/65536 [10:20:15<1:01:26,  1.51it/s]
 91%|█████████▏| 59953/65536 [10:20:16<1:00:46,  1.53it/s]
 91%|█████████▏| 59954/65536 [10:20:16<59:20,  1.57it/s]  
 91%|█████████▏| 59955/65536 [10:20:17<58:05,  1.60it/s]
 91%|█████████▏| 59956/65536 [10:20:18<57:31,  1.62it/s]
 91%|█████████▏| 59957/65536 [10:20:18<58:27,  1.59it/s]
 91%|█████████▏| 59958/65536 [10:20:19<58:07,  1.60it/s]
 91%|█████████▏| 59959/65536 [10:20:20<59:38,  1.56it/s]
 91%|█████████▏| 59960/65536 [10:20:20<58:23,  1.59it/s]
                                                        
{'loss': 1.5303, 'learning_rate': 1.7680910983225178e-07, 'epoch': 3701.23}
+
 91%|█████████▏| 59960/65536 [10:20:20<58:23,  1.59it/s]
 91%|█████████▏| 59961/65536 [10:20:21<57:50,  1.61it/s]
 91%|█████████▏| 59962/65536 [10:20:21<58:35,  1.59it/s]
 91%|█████████▏| 59963/65536 [10:20:22<57:41,  1.61it/s]
 91%|█████████▏| 59964/65536 [10:20:23<58:01,  1.60it/s]
 91%|█████████▏| 59965/65536 [10:20:23<58:11,  1.60it/s]
 92%|█████████▏| 59966/65536 [10:20:24<59:24,  1.56it/s]
 92%|█████████▏| 59967/65536 [10:20:25<57:40,  1.61it/s]
 92%|█████████▏| 59968/65536 [10:20:25<58:12,  1.59it/s]
 92%|█████████▏| 59969/65536 [10:20:26<57:21,  1.62it/s]
 92%|█████████▏| 59970/65536 [10:20:26<57:03,  1.63it/s]
 92%|█████████▏| 59971/65536 [10:20:27<55:43,  1.66it/s]
 92%|█████████▏| 59972/65536 [10:20:28<55:18,  1.68it/s]
 92%|█████████▏| 59973/65536 [10:20:28<56:09,  1.65it/s]
 92%|█████████▏| 59974/65536 [10:20:29<59:23,  1.56it/s]
 92%|█████████▏| 59975/65536 [10:20:29<57:57,  1.60it/s]
 92%|█████████▏| 59976/65536 [10:20:30<57:20,  1.62it/s]
 92%|█████████▏| 59977/65536 [10:20:31<57:39,  1.61it/s]
 92%|█████████▏| 59978/65536 [10:20:31<56:53,  1.63it/s]
 92%|█████████▏| 59979/65536 [10:20:32<56:37,  1.64it/s]
 92%|█████████▏| 59980/65536 [10:20:32<56:00,  1.65it/s]
                                                        
{'loss': 1.538, 'learning_rate': 1.7653361087302556e-07, 'epoch': 3702.47}
+
 92%|█████████▏| 59980/65536 [10:20:32<56:00,  1.65it/s]
 92%|█████████▏| 59981/65536 [10:20:33<58:05,  1.59it/s]
 92%|█████████▏| 59982/65536 [10:20:34<57:38,  1.61it/s]
 92%|█████████▏| 59983/65536 [10:20:34<57:59,  1.60it/s]
 92%|█████████▏| 59984/65536 [10:20:35<57:17,  1.62it/s]
 92%|█████████▏| 59985/65536 [10:20:36<57:16,  1.62it/s]
 92%|█████████▏| 59986/65536 [10:20:36<58:01,  1.59it/s]
 92%|█████████▏| 59987/65536 [10:20:37<58:02,  1.59it/s]
 92%|█████████▏| 59988/65536 [10:20:37<57:06,  1.62it/s]
 92%|█████████▏| 59989/65536 [10:20:38<58:23,  1.58it/s]
 92%|█████████▏| 59990/65536 [10:20:39<58:06,  1.59it/s]
 92%|█████████▏| 59991/65536 [10:20:39<57:26,  1.61it/s]
 92%|█████████▏| 59992/65536 [10:20:40<56:21,  1.64it/s]
 92%|█████████▏| 59993/65536 [10:20:41<57:19,  1.61it/s]
 92%|█████████▏| 59994/65536 [10:20:41<57:43,  1.60it/s]
 92%|█████████▏| 59995/65536 [10:20:42<56:52,  1.62it/s]
 92%|█████████▏| 59996/65536 [10:20:42<56:38,  1.63it/s]
 92%|█████████▏| 59997/65536 [10:20:43<57:30,  1.61it/s]
 92%|█████████▏| 59998/65536 [10:20:44<56:33,  1.63it/s]
 92%|█████��███▏| 59999/65536 [10:20:44<57:51,  1.59it/s]
 92%|█████████▏| 60000/65536 [10:20:45<57:25,  1.61it/s]
                                                        
{'loss': 1.4937, 'learning_rate': 1.7625811191379941e-07, 'epoch': 3703.7}
+
 92%|█████████▏| 60000/65536 [10:20:45<57:25,  1.61it/s]Some non-default generation parameters are set in the model config. These should go into a GenerationConfig file (https://huggingface.co./docs/transformers/generation_strategies#save-a-custom-decoding-strategy-with-your-model) instead. This warning will be raised to an exception in v4.41.
+Non-default generation parameters: {'max_length': 200, 'early_stopping': True, 'num_beams': 5, 'forced_eos_token_id': 2}
+
 92%|█████████▏| 60001/65536 [10:21:15<14:30:53,  9.44s/it]
 92%|█████████▏| 60002/65536 [10:21:16<10:27:46,  6.81s/it]
 92%|█████████▏| 60003/65536 [10:21:16<7:36:38,  4.95s/it] 
 92%|█████████▏| 60004/65536 [10:21:17<5:37:39,  3.66s/it]/opt/conda/lib/python3.10/multiprocessing/popen_fork.py:66: RuntimeWarning: os.fork() was called. os.fork() is incompatible with multithreaded code, and JAX is multithreaded, so this will likely lead to a deadlock.
+  self.pid = os.fork()
+
 92%|█████████▏| 60005/65536 [10:21:18<4:16:44,  2.79s/it]
 92%|█████████▏| 60006/65536 [10:21:18<3:17:17,  2.14s/it]
 92%|█████████▏| 60007/65536 [10:21:19<2:36:56,  1.70s/it]
 92%|█████████▏| 60008/65536 [10:21:20<2:09:36,  1.41s/it]
 92%|█████████▏| 60009/65536 [10:21:20<1:49:15,  1.19s/it]
 92%|█████████▏| 60010/65536 [10:21:21<1:35:13,  1.03s/it]
 92%|█████████▏| 60011/65536 [10:21:22<1:24:54,  1.08it/s]
 92%|█████████▏| 60012/65536 [10:21:22<1:16:45,  1.20it/s]
 92%|█████████▏| 60013/65536 [10:21:23<1:10:22,  1.31it/s]
 92%|█████████▏| 60014/65536 [10:21:24<1:05:48,  1.40it/s]
 92%|█████████▏| 60015/65536 [10:21:24<1:02:54,  1.46it/s]
 92%|█████████▏| 60016/65536 [10:21:25<1:01:52,  1.49it/s]
 92%|█████████▏| 60017/65536 [10:21:25<1:02:57,  1.46it/s]
 92%|█████████▏| 60018/65536 [10:21:26<1:00:42,  1.51it/s]
 92%|█████████▏| 60019/65536 [10:21:27<1:01:25,  1.50it/s]
 92%|█████████▏| 60020/65536 [10:21:27<1:01:19,  1.50it/s]
                                                          
{'loss': 1.5526, 'learning_rate': 1.7598261295457327e-07, 'epoch': 3704.94}
+
 92%|█████████▏| 60020/65536 [10:21:27<1:01:19,  1.50it/s]
 92%|█████████▏| 60021/65536 [10:21:28<59:49,  1.54it/s]  
 92%|█████████▏| 60022/65536 [10:21:29<1:01:22,  1.50it/s]
 92%|█████████▏| 60023/65536 [10:21:29<1:00:35,  1.52it/s]
 92%|█████████▏| 60024/65536 [10:21:30<1:00:06,  1.53it/s]
 92%|█████████▏| 60025/65536 [10:21:31<1:01:15,  1.50it/s]
 92%|█████████▏| 60026/65536 [10:21:31<1:02:05,  1.48it/s]
 92%|█████████▏| 60027/65536 [10:21:32<1:00:59,  1.51it/s]
 92%|█████████▏| 60028/65536 [10:21:33<59:44,  1.54it/s]  
 92%|█████████▏| 60029/65536 [10:21:33<58:10,  1.58it/s]
 92%|█████████▏| 60030/65536 [10:21:34<57:54,  1.58it/s]
 92%|█████████▏| 60031/65536 [10:21:35<58:48,  1.56it/s]
 92%|█████████▏| 60032/65536 [10:21:35<57:38,  1.59it/s]
 92%|█████████▏| 60033/65536 [10:21:36<57:00,  1.61it/s]
 92%|█████████▏| 60034/65536 [10:21:36<56:58,  1.61it/s]
 92%|█████████▏| 60035/65536 [10:21:37<57:32,  1.59it/s]
 92%|█████████▏| 60036/65536 [10:21:38<58:48,  1.56it/s]
 92%|█████████▏| 60037/65536 [10:21:38<57:47,  1.59it/s]
 92%|█████████▏| 60038/65536 [10:21:39<1:01:03,  1.50it/s]
 92%|█████████▏| 60039/65536 [10:21:40<1:02:08,  1.47it/s]
 92%|█████████▏| 60040/65536 [10:21:40<1:01:35,  1.49it/s]
                                                          
{'loss': 1.5274, 'learning_rate': 1.7570711399534713e-07, 'epoch': 3706.17}
+
 92%|█████████▏| 60040/65536 [10:21:40<1:01:35,  1.49it/s]
 92%|█████████▏| 60041/65536 [10:21:41<1:02:42,  1.46it/s]
 92%|█████████▏| 60042/65536 [10:21:42<1:00:02,  1.53it/s]
 92%|█████████▏| 60043/65536 [10:21:42<1:01:32,  1.49it/s]
 92%|█████████▏| 60044/65536 [10:21:43<1:01:38,  1.48it/s]
 92%|█████████▏| 60045/65536 [10:21:44<58:58,  1.55it/s]  
 92%|█████████▏| 60046/65536 [10:21:44<57:46,  1.58it/s]
 92%|█████████▏| 60047/65536 [10:21:45<56:48,  1.61it/s]
 92%|█████████▏| 60048/65536 [10:21:46<56:26,  1.62it/s]
 92%|█████████▏| 60049/65536 [10:21:46<57:57,  1.58it/s]
 92%|█████████▏| 60050/65536 [10:21:47<57:57,  1.58it/s]
 92%|█████████▏| 60051/65536 [10:21:47<58:13,  1.57it/s]
 92%|█████████▏| 60052/65536 [10:21:48<58:09,  1.57it/s]
 92%|█████████▏| 60053/65536 [10:21:49<59:22,  1.54it/s]
 92%|█████████▏| 60054/65536 [10:21:49<59:37,  1.53it/s]
 92%|█████████▏| 60055/65536 [10:21:50<1:00:17,  1.51it/s]
 92%|█████████▏| 60056/65536 [10:21:51<1:00:43,  1.50it/s]
 92%|█████████▏| 60057/65536 [10:21:51<59:00,  1.55it/s]  
 92%|█████████▏| 60058/65536 [10:21:52<59:52,  1.52it/s]
 92%|█████████▏| 60059/65536 [10:21:53<1:00:37,  1.51it/s]
 92%|█████████▏| 60060/65536 [10:21:53<59:42,  1.53it/s]  
                                                        
{'loss': 1.5919, 'learning_rate': 1.75431615036121e-07, 'epoch': 3707.41}
+
 92%|█████████▏| 60060/65536 [10:21:53<59:42,  1.53it/s]
 92%|█████████▏| 60061/65536 [10:21:54<1:01:02,  1.49it/s]
 92%|█████████▏| 60062/65536 [10:21:55<1:01:14,  1.49it/s]
 92%|█████████▏| 60063/65536 [10:21:55<1:01:00,  1.50it/s]
 92%|█████████▏| 60064/65536 [10:21:56<1:00:19,  1.51it/s]
 92%|█████████▏| 60065/65536 [10:21:57<1:00:20,  1.51it/s]
\ No newline at end of file