Training in progress, step 6300

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4952d07d35404125ac82a69ff1d2b691b555ce36f8f5e2491332621aa761003d
+oid sha256:44d9867527ada8f06b3d5b4a3b1a084c5587bf5d24b8f6abb3243ac9e5ecc1bb
 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:eae5cf9b7f19e4d620f566f3c220b67be60268ea84f3ad533257770b5c7395fc
-size 71512
+oid sha256:f65e4880300145361cff158d3d75198f6769e43165410c92a22994d5a8111dd3
+size 71723

wandb/debug-internal.log

@@ -5201,3 +5201,18 @@ subprocess.TimeoutExpired: Command '['conda', 'env', 'export']' timed out after
 2024-04-12 09:48:50,376 DEBUG   SenderThread:485 [sender.py:send_request():406] send_request: summary_record
 2024-04-12 09:48:50,376 INFO    SenderThread:485 [sender.py:_save_file():1390] saving file wandb-summary.json with policy end
 2024-04-12 09:48:51,098 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 09:48:53,099 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 09:48:55,194 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 09:49:00,195 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 09:49:02,134 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: internal_messages
+2024-04-12 09:49:02,407 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: stop_status
+2024-04-12 09:49:02,407 DEBUG   SenderThread:485 [sender.py:send_request():406] send_request: stop_status
+2024-04-12 09:49:05,486 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 09:49:10,487 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 09:49:14,179 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: partial_history
+2024-04-12 09:49:14,180 DEBUG   SenderThread:485 [sender.py:send():379] send: history
+2024-04-12 09:49:14,181 DEBUG   SenderThread:485 [sender.py:send_request():406] send_request: summary_record
+2024-04-12 09:49:14,181 INFO    SenderThread:485 [sender.py:_save_file():1390] saving file wandb-summary.json with policy end
+2024-04-12 09:49:15,108 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 09:49:15,294 DEBUG   SenderThread:485 [sender.py:send():379] send: stats
+2024-04-12 09:49:16,295 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: status_report

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

@@ -626,4 +626,6 @@
   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": 3.4637, "train/grad_norm": 44.352596282958984, "train/learning_rate": 2.408961130718405e-05, "train/epoch": 0.5, "train/global_step": 6280, "_timestamp": 1712915330.374364, "_runtime": 7685.866645812988, "_step": 313}
+{"train/loss": 3.4403, "train/grad_norm": 35.380767822265625, "train/learning_rate": 2.405091801274387e-05, "train/epoch": 0.5, "train/global_step": 6300, "_timestamp": 1712915354.179023, "_runtime": 7709.671304941177, "_step": 314}

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

@@ -5201,3 +5201,18 @@ subprocess.TimeoutExpired: Command '['conda', 'env', 'export']' timed out after
 2024-04-12 09:48:50,376 DEBUG   SenderThread:485 [sender.py:send_request():406] send_request: summary_record
 2024-04-12 09:48:50,376 INFO    SenderThread:485 [sender.py:_save_file():1390] saving file wandb-summary.json with policy end
 2024-04-12 09:48:51,098 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 09:48:53,099 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 09:48:55,194 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 09:49:00,195 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 09:49:02,134 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: internal_messages
+2024-04-12 09:49:02,407 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: stop_status
+2024-04-12 09:49:02,407 DEBUG   SenderThread:485 [sender.py:send_request():406] send_request: stop_status
+2024-04-12 09:49:05,486 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 09:49:10,487 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 09:49:14,179 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: partial_history
+2024-04-12 09:49:14,180 DEBUG   SenderThread:485 [sender.py:send():379] send: history
+2024-04-12 09:49:14,181 DEBUG   SenderThread:485 [sender.py:send_request():406] send_request: summary_record
+2024-04-12 09:49:14,181 INFO    SenderThread:485 [sender.py:_save_file():1390] saving file wandb-summary.json with policy end
+2024-04-12 09:49:15,108 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 09:49:15,294 DEBUG   SenderThread:485 [sender.py:send():379] send: stats
+2024-04-12 09:49:16,295 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: status_report