Training in progress, step 40

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5bdd0d7af5fcc36cb3a3c639b8d0e10933390cc6ced74159ac35d92a61985aa0
+oid sha256:aeda3f74d0c65b865c78f72723c369a102c203856812079274186432e8ae2082
 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:bcd311551f206d6e178eae36161b33978c1915bacdaed37b693ddfca00e24da3
-size 5489
+oid sha256:4090afe353b724cb1bf6ee84e984cc3099a34ad4ae39ea1d31362b0c34f051b0
+size 5696

wandb/debug-internal.log

@@ -83,3 +83,19 @@ subprocess.TimeoutExpired: Command '['conda', 'env', 'export']' timed out after
 2024-04-12 07:41:22,910 INFO    Thread-12 :485 [dir_watcher.py:_on_file_created():271] file/dir created: /kaggle/working/wandb/run-20240412_074044-widovcn3/files/wandb-summary.json
 2024-04-12 07:41:23,910 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 07:41:26,797 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 07:41:31,036 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: stop_status
+2024-04-12 07:41:31,036 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: internal_messages
+2024-04-12 07:41:31,037 DEBUG   SenderThread:485 [sender.py:send_request():406] send_request: stop_status
+2024-04-12 07:41:32,086 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 07:41:37,087 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 07:41:42,088 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 07:41:45,014 DEBUG   SystemMonitor:485 [system_monitor.py:_start():172] Starting system metrics aggregation loop
+2024-04-12 07:41:45,015 DEBUG   SenderThread:485 [sender.py:send():379] send: stats
+2024-04-12 07:41:46,035 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: stop_status
+2024-04-12 07:41:46,035 DEBUG   SenderThread:485 [sender.py:send_request():406] send_request: stop_status
+2024-04-12 07:41:46,076 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: internal_messages
+2024-04-12 07:41:46,285 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: partial_history
+2024-04-12 07:41:46,286 DEBUG   SenderThread:485 [sender.py:send():379] send: history
+2024-04-12 07:41:46,287 DEBUG   SenderThread:485 [sender.py:send_request():406] send_request: summary_record
+2024-04-12 07:41:46,287 INFO    SenderThread:485 [sender.py:_save_file():1390] saving file wandb-summary.json with policy end
+2024-04-12 07:41:46,921 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

@@ -1,3 +1,5 @@
+/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": 4.5777, "train/grad_norm": 62.10089111328125, "train/learning_rate": 1e-06, "train/epoch": 0.0, "train/global_step": 20, "_timestamp": 1712907682.0631418, "_runtime": 37.555423736572266, "_step": 0}
+{"train/loss": 4.4077, "train/grad_norm": 39.39016342163086, "train/learning_rate": 2e-06, "train/epoch": 0.0, "train/global_step": 40, "_timestamp": 1712907706.285044, "_runtime": 61.77732586860657, "_step": 1}

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

@@ -83,3 +83,19 @@ subprocess.TimeoutExpired: Command '['conda', 'env', 'export']' timed out after
 2024-04-12 07:41:22,910 INFO    Thread-12 :485 [dir_watcher.py:_on_file_created():271] file/dir created: /kaggle/working/wandb/run-20240412_074044-widovcn3/files/wandb-summary.json
 2024-04-12 07:41:23,910 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 07:41:26,797 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 07:41:31,036 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: stop_status
+2024-04-12 07:41:31,036 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: internal_messages
+2024-04-12 07:41:31,037 DEBUG   SenderThread:485 [sender.py:send_request():406] send_request: stop_status
+2024-04-12 07:41:32,086 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 07:41:37,087 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 07:41:42,088 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 07:41:45,014 DEBUG   SystemMonitor:485 [system_monitor.py:_start():172] Starting system metrics aggregation loop
+2024-04-12 07:41:45,015 DEBUG   SenderThread:485 [sender.py:send():379] send: stats
+2024-04-12 07:41:46,035 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: stop_status
+2024-04-12 07:41:46,035 DEBUG   SenderThread:485 [sender.py:send_request():406] send_request: stop_status
+2024-04-12 07:41:46,076 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: internal_messages
+2024-04-12 07:41:46,285 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: partial_history
+2024-04-12 07:41:46,286 DEBUG   SenderThread:485 [sender.py:send():379] send: history
+2024-04-12 07:41:46,287 DEBUG   SenderThread:485 [sender.py:send_request():406] send_request: summary_record
+2024-04-12 07:41:46,287 INFO    SenderThread:485 [sender.py:_save_file():1390] saving file wandb-summary.json with policy end
+2024-04-12 07:41:46,921 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