Training in progress, step 2920

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:53f013c8a2c8bef97abb794e6a06b77a2e4147632cf2273758eca6f336e2f82f
+oid sha256:6a6c16135c13b23b78a32406104d4f95e87d9791d1a272abfccfbd9fdb510426
 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 sha256:590501ed22a9c966a19f8dbeb4075f43d9937f0d266f8185fa5269eaaa03c05b
-size 35853
+oid sha256:3d7cb1a081228fe53616945e4d3de7484453eeb4373cfb3285f6c29a5b44eb69
+size 36064

wandb/debug-internal.log

@@ -2438,3 +2438,18 @@ subprocess.TimeoutExpired: Command '['conda', 'env', 'export']' timed out after
 2024-04-12 08:40:01,777 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: stop_status
 2024-04-12 08:40:01,778 DEBUG   SenderThread:485 [sender.py:send_request():406] send_request: stop_status
 2024-04-12 08:40:01,781 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: internal_messages
+2024-04-12 08:40:04,403 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 08:40:05,060 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 08:40:10,061 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 08:40:15,061 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 08:40:15,150 DEBUG   SenderThread:485 [sender.py:send():379] send: stats
+2024-04-12 08:40:16,765 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: stop_status
+2024-04-12 08:40:16,765 DEBUG   SenderThread:485 [sender.py:send_request():406] send_request: stop_status
+2024-04-12 08:40:16,780 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: internal_messages
+2024-04-12 08:40:20,854 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 08:40:25,855 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 08:40:26,264 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: partial_history
+2024-04-12 08:40:26,265 DEBUG   SenderThread:485 [sender.py:send():379] send: history
+2024-04-12 08:40:26,266 DEBUG   SenderThread:485 [sender.py:send_request():406] send_request: summary_record
+2024-04-12 08:40:26,266 INFO    SenderThread:485 [sender.py:_save_file():1390] saving file wandb-summary.json with policy end
+2024-04-12 08:40:26,412 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

@@ -286,4 +286,8 @@
   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.
+  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": 3.5707, "train/grad_norm": 41.04441833496094, "train/learning_rate": 2.8971541407162637e-05, "train/epoch": 0.23, "train/global_step": 2900, "_timestamp": 1712911201.1533623, "_runtime": 3556.6456441879272, "_step": 144}
+{"train/loss": 3.5748, "train/grad_norm": 31.303403854370117, "train/learning_rate": 2.8953789174920795e-05, "train/epoch": 0.23, "train/global_step": 2920, "_timestamp": 1712911226.263947, "_runtime": 3581.7562289237976, "_step": 145}

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

@@ -2438,3 +2438,18 @@ subprocess.TimeoutExpired: Command '['conda', 'env', 'export']' timed out after
 2024-04-12 08:40:01,777 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: stop_status
 2024-04-12 08:40:01,778 DEBUG   SenderThread:485 [sender.py:send_request():406] send_request: stop_status
 2024-04-12 08:40:01,781 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: internal_messages
+2024-04-12 08:40:04,403 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 08:40:05,060 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 08:40:10,061 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 08:40:15,061 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 08:40:15,150 DEBUG   SenderThread:485 [sender.py:send():379] send: stats
+2024-04-12 08:40:16,765 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: stop_status
+2024-04-12 08:40:16,765 DEBUG   SenderThread:485 [sender.py:send_request():406] send_request: stop_status
+2024-04-12 08:40:16,780 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: internal_messages
+2024-04-12 08:40:20,854 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 08:40:25,855 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 08:40:26,264 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: partial_history
+2024-04-12 08:40:26,265 DEBUG   SenderThread:485 [sender.py:send():379] send: history
+2024-04-12 08:40:26,266 DEBUG   SenderThread:485 [sender.py:send_request():406] send_request: summary_record
+2024-04-12 08:40:26,266 INFO    SenderThread:485 [sender.py:_save_file():1390] saving file wandb-summary.json with policy end
+2024-04-12 08:40:26,412 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