Training in progress, step 5600

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4e51450f5e4b92cb6050a5a712137dfb440701725d7829a72cff434a6a6aaa41
+oid sha256:5fd06c9f2222239e592302c7e30afefc517d5cb1a680437c4ab7fea503265580
 size 13982248

runs/Apr12_07-40-41_e5a48bec8248/events.out.tfevents.1712907642.e5a48bec8248.443.0

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:eef85c95fbde3570115b4926112e5e23cb740522c25499dfd2acedeb6ba5d608
-size 64127
+oid sha256:35f007d9dfc376822099ffc1b24263280726b5c38f2b9648a30c3f06e82f0194
+size 64338

wandb/debug-internal.log

@@ -4631,3 +4631,20 @@ subprocess.TimeoutExpired: Command '['conda', 'env', 'export']' timed out after
 2024-04-12 09:34:42,158 INFO    SenderThread:485 [sender.py:_save_file():1390] saving file wandb-summary.json with policy end
 2024-04-12 09:34:42,760 INFO    Thread-12 :485 [dir_watcher.py:_on_file_modified():288] file/dir modified: /kaggle/working/wandb/run-20240412_074044-widovcn3/files/wandb-summary.json
 2024-04-12 09:34:43,760 INFO    Thread-12 :485 [dir_watcher.py:_on_file_modified():288] file/dir modified: /kaggle/working/wandb/run-20240412_074044-widovcn3/files/output.log
+2024-04-12 09:34:44,897 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 09:34:45,264 DEBUG   SenderThread:485 [sender.py:send():379] send: stats
+2024-04-12 09:34:47,021 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: internal_messages
+2024-04-12 09:34:47,390 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: stop_status
+2024-04-12 09:34:47,391 DEBUG   SenderThread:485 [sender.py:send_request():406] send_request: stop_status
+2024-04-12 09:34:50,508 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 09:34:55,509 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 09:35:00,510 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 09:35:02,021 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: internal_messages
+2024-04-12 09:35:02,390 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: stop_status
+2024-04-12 09:35:02,391 DEBUG   SenderThread:485 [sender.py:send_request():406] send_request: stop_status
+2024-04-12 09:35:06,432 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 09:35:07,355 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: partial_history
+2024-04-12 09:35:07,356 DEBUG   SenderThread:485 [sender.py:send():379] send: history
+2024-04-12 09:35:07,356 DEBUG   SenderThread:485 [sender.py:send_request():406] send_request: summary_record
+2024-04-12 09:35:07,357 INFO    SenderThread:485 [sender.py:_save_file():1390] saving file wandb-summary.json with policy end
+2024-04-12 09:35:07,774 INFO    Thread-12 :485 [dir_watcher.py:_on_file_modified():288] file/dir modified: /kaggle/working/wandb/run-20240412_074044-widovcn3/files/wandb-summary.json

wandb/run-20240412_074044-widovcn3/files/output.log

@@ -556,4 +556,6 @@
   warnings.warn(
 /opt/conda/lib/python3.10/site-packages/torch/utils/checkpoint.py:429: UserWarning: torch.utils.checkpoint: please pass in use_reentrant=True or use_reentrant=False explicitly. The default value of use_reentrant will be updated to be False in the future. To maintain current behavior, pass use_reentrant=True. It is recommended that you use use_reentrant=False. Refer to docs for more details on the differences between the two variants.
   warnings.warn(
+/opt/conda/lib/python3.10/site-packages/torch/utils/checkpoint.py:429: UserWarning: torch.utils.checkpoint: please pass in use_reentrant=True or use_reentrant=False explicitly. The default value of use_reentrant will be updated to be False in the future. To maintain current behavior, pass use_reentrant=True. It is recommended that you use use_reentrant=False. Refer to docs for more details on the differences between the two variants.
+  warnings.warn(
 /opt/conda/lib/python3.10/site-packages/torch/utils/checkpoint.py:429: UserWarning: torch.utils.checkpoint: please pass in use_reentrant=True or use_reentrant=False explicitly. The default value of use_reentrant will be updated to be False in the future. To maintain current behavior, pass use_reentrant=True. It is recommended that you use use_reentrant=False. Refer to docs for more details on the differences between the two variants.

wandb/run-20240412_074044-widovcn3/files/wandb-summary.json

@@ -1 +1 @@
-{"train/loss": 3.4894, "train/grad_norm": 35.5067138671875, "train/learning_rate": 2.5380978745169473e-05, "train/epoch": 0.45, "train/global_step": 5580, "_timestamp": 1712914482.1552558, "_runtime": 6837.6475377082825, "_step": 278}
+{"train/loss": 3.4033, "train/grad_norm": 27.13253402709961, "train/learning_rate": 2.5345856659171567e-05, "train/epoch": 0.45, "train/global_step": 5600, "_timestamp": 1712914507.3545787, "_runtime": 6862.8468606472015, "_step": 279}

wandb/run-20240412_074044-widovcn3/logs/debug-internal.log

@@ -4631,3 +4631,20 @@ subprocess.TimeoutExpired: Command '['conda', 'env', 'export']' timed out after
 2024-04-12 09:34:42,158 INFO    SenderThread:485 [sender.py:_save_file():1390] saving file wandb-summary.json with policy end
 2024-04-12 09:34:42,760 INFO    Thread-12 :485 [dir_watcher.py:_on_file_modified():288] file/dir modified: /kaggle/working/wandb/run-20240412_074044-widovcn3/files/wandb-summary.json
 2024-04-12 09:34:43,760 INFO    Thread-12 :485 [dir_watcher.py:_on_file_modified():288] file/dir modified: /kaggle/working/wandb/run-20240412_074044-widovcn3/files/output.log
+2024-04-12 09:34:44,897 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 09:34:45,264 DEBUG   SenderThread:485 [sender.py:send():379] send: stats
+2024-04-12 09:34:47,021 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: internal_messages
+2024-04-12 09:34:47,390 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: stop_status
+2024-04-12 09:34:47,391 DEBUG   SenderThread:485 [sender.py:send_request():406] send_request: stop_status
+2024-04-12 09:34:50,508 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 09:34:55,509 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 09:35:00,510 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 09:35:02,021 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: internal_messages
+2024-04-12 09:35:02,390 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: stop_status
+2024-04-12 09:35:02,391 DEBUG   SenderThread:485 [sender.py:send_request():406] send_request: stop_status
+2024-04-12 09:35:06,432 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 09:35:07,355 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: partial_history
+2024-04-12 09:35:07,356 DEBUG   SenderThread:485 [sender.py:send():379] send: history
+2024-04-12 09:35:07,356 DEBUG   SenderThread:485 [sender.py:send_request():406] send_request: summary_record
+2024-04-12 09:35:07,357 INFO    SenderThread:485 [sender.py:_save_file():1390] saving file wandb-summary.json with policy end
+2024-04-12 09:35:07,774 INFO    Thread-12 :485 [dir_watcher.py:_on_file_modified():288] file/dir modified: /kaggle/working/wandb/run-20240412_074044-widovcn3/files/wandb-summary.json