Training in progress, step 3760

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:70fc65989648f7dc64734e8244e0e6879cfab6125b857d5586fade7a4f45ce0c
+oid sha256:d2eb339fca1743ce5981d75d94271aea8c49805aa8787cc062a2e480f02e91a8
 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 sha256:97031564a94f3bd32ea2530c94cae8e4830c3fb56456fccbd281bf0d5b2686ac
-size 44715
+oid sha256:4d719a359dfcdd5bb4294d13485b01aee8e81cb078e3b25fbbd8b53008f87a6d
+size 44926

wandb/debug-internal.log

@@ -3122,3 +3122,22 @@ subprocess.TimeoutExpired: Command '['conda', 'env', 'export']' timed out after
 2024-04-12 08:57:02,157 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: stop_status
 2024-04-12 08:57:02,157 DEBUG   SenderThread:485 [sender.py:send_request():406] send_request: stop_status
 2024-04-12 08:57:02,845 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 08:57:03,859 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 08:57:04,847 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 08:57:08,860 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 08:57:13,861 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 08:57:15,185 DEBUG   SenderThread:485 [sender.py:send():379] send: stats
+2024-04-12 08:57:16,916 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: internal_messages
+2024-04-12 08:57:17,157 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: stop_status
+2024-04-12 08:57:17,158 DEBUG   SenderThread:485 [sender.py:send_request():406] send_request: stop_status
+2024-04-12 08:57:19,265 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 08:57:24,266 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 08:57:29,267 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 08:57:30,441 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: partial_history
+2024-04-12 08:57:30,443 DEBUG   SenderThread:485 [sender.py:send():379] send: history
+2024-04-12 08:57:30,443 DEBUG   SenderThread:485 [sender.py:send_request():406] send_request: summary_record
+2024-04-12 08:57:30,444 INFO    SenderThread:485 [sender.py:_save_file():1390] saving file wandb-summary.json with policy end
+2024-04-12 08:57:30,859 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 08:57:31,916 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: internal_messages
+2024-04-12 08:57:32,158 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: stop_status
+2024-04-12 08:57:32,159 DEBUG   SenderThread:485 [sender.py:send_request():406] send_request: stop_status

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

@@ -370,4 +370,6 @@
   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": 3.5619, "train/grad_norm": 25.85703468322754, "train/learning_rate": 2.8102249451758162e-05, "train/epoch": 0.3, "train/global_step": 3740, "_timestamp": 1712912222.1007948, "_runtime": 4577.593076705933, "_step": 186}
+{"train/loss": 3.5173, "train/grad_norm": 32.1122932434082, "train/learning_rate": 2.8078528754588207e-05, "train/epoch": 0.3, "train/global_step": 3760, "_timestamp": 1712912250.4413297, "_runtime": 4605.933611631393, "_step": 187}

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

@@ -3122,3 +3122,22 @@ subprocess.TimeoutExpired: Command '['conda', 'env', 'export']' timed out after
 2024-04-12 08:57:02,157 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: stop_status
 2024-04-12 08:57:02,157 DEBUG   SenderThread:485 [sender.py:send_request():406] send_request: stop_status
 2024-04-12 08:57:02,845 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 08:57:03,859 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 08:57:04,847 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 08:57:08,860 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 08:57:13,861 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 08:57:15,185 DEBUG   SenderThread:485 [sender.py:send():379] send: stats
+2024-04-12 08:57:16,916 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: internal_messages
+2024-04-12 08:57:17,157 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: stop_status
+2024-04-12 08:57:17,158 DEBUG   SenderThread:485 [sender.py:send_request():406] send_request: stop_status
+2024-04-12 08:57:19,265 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 08:57:24,266 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 08:57:29,267 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 08:57:30,441 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: partial_history
+2024-04-12 08:57:30,443 DEBUG   SenderThread:485 [sender.py:send():379] send: history
+2024-04-12 08:57:30,443 DEBUG   SenderThread:485 [sender.py:send_request():406] send_request: summary_record
+2024-04-12 08:57:30,444 INFO    SenderThread:485 [sender.py:_save_file():1390] saving file wandb-summary.json with policy end
+2024-04-12 08:57:30,859 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 08:57:31,916 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: internal_messages
+2024-04-12 08:57:32,158 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: stop_status
+2024-04-12 08:57:32,159 DEBUG   SenderThread:485 [sender.py:send_request():406] send_request: stop_status