Training in progress, step 3420

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3cd865daf8559d59e29839bc441ae79f1e2737ae9485572c595f64db334fed61
+oid sha256:65f8e5f02732fdec9957da384752cf78dcb48b177a3d9870168cb62a3d63c083
 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 sha256:3e9b4d26342b315f76f80e80cfc4f915bef8292a532ed9024aa0f62a95958a9a
-size 41128
+oid sha256:5ee20ddb9331909423f0cb4ed0eb79a4b0483a37b6aaf312b67299c5f2389118
+size 41339

wandb/debug-internal.log

@@ -2851,3 +2851,20 @@ subprocess.TimeoutExpired: Command '['conda', 'env', 'export']' timed out after
 2024-04-12 08:50:25,884 DEBUG   SenderThread:485 [sender.py:send_request():406] send_request: summary_record
 2024-04-12 08:50:25,885 INFO    SenderThread:485 [sender.py:_save_file():1390] saving file wandb-summary.json with policy end
 2024-04-12 08:50:26,658 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 08:50:28,659 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 08:50:29,584 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 08:50:31,771 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: stop_status
+2024-04-12 08:50:31,771 DEBUG   SenderThread:485 [sender.py:send_request():406] send_request: stop_status
+2024-04-12 08:50:31,785 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: internal_messages
+2024-04-12 08:50:34,900 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 08:50:39,901 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 08:50:44,902 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 08:50:45,171 DEBUG   SenderThread:485 [sender.py:send():379] send: stats
+2024-04-12 08:50:46,771 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: stop_status
+2024-04-12 08:50:46,772 DEBUG   SenderThread:485 [sender.py:send_request():406] send_request: stop_status
+2024-04-12 08:50:46,812 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: internal_messages
+2024-04-12 08:50:48,388 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: partial_history
+2024-04-12 08:50:48,390 DEBUG   SenderThread:485 [sender.py:send():379] send: history
+2024-04-12 08:50:48,390 DEBUG   SenderThread:485 [sender.py:send_request():406] send_request: summary_record
+2024-04-12 08:50:48,390 INFO    SenderThread:485 [sender.py:_save_file():1390] saving file wandb-summary.json with policy end
+2024-04-12 08:50:48,666 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

@@ -336,4 +336,6 @@
   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": 3.4868, "train/grad_norm": 35.51163101196289, "train/learning_rate": 2.848427611628093e-05, "train/epoch": 0.27, "train/global_step": 3400, "_timestamp": 1712911825.8830612, "_runtime": 4181.375343084335, "_step": 169}
+{"train/loss": 3.4984, "train/grad_norm": 28.35199546813965, "train/learning_rate": 2.8462924493036168e-05, "train/epoch": 0.27, "train/global_step": 3420, "_timestamp": 1712911848.3883505, "_runtime": 4203.880632400513, "_step": 170}

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

@@ -2851,3 +2851,20 @@ subprocess.TimeoutExpired: Command '['conda', 'env', 'export']' timed out after
 2024-04-12 08:50:25,884 DEBUG   SenderThread:485 [sender.py:send_request():406] send_request: summary_record
 2024-04-12 08:50:25,885 INFO    SenderThread:485 [sender.py:_save_file():1390] saving file wandb-summary.json with policy end
 2024-04-12 08:50:26,658 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 08:50:28,659 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 08:50:29,584 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 08:50:31,771 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: stop_status
+2024-04-12 08:50:31,771 DEBUG   SenderThread:485 [sender.py:send_request():406] send_request: stop_status
+2024-04-12 08:50:31,785 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: internal_messages
+2024-04-12 08:50:34,900 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 08:50:39,901 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 08:50:44,902 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 08:50:45,171 DEBUG   SenderThread:485 [sender.py:send():379] send: stats
+2024-04-12 08:50:46,771 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: stop_status
+2024-04-12 08:50:46,772 DEBUG   SenderThread:485 [sender.py:send_request():406] send_request: stop_status
+2024-04-12 08:50:46,812 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: internal_messages
+2024-04-12 08:50:48,388 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: partial_history
+2024-04-12 08:50:48,390 DEBUG   SenderThread:485 [sender.py:send():379] send: history
+2024-04-12 08:50:48,390 DEBUG   SenderThread:485 [sender.py:send_request():406] send_request: summary_record
+2024-04-12 08:50:48,390 INFO    SenderThread:485 [sender.py:_save_file():1390] saving file wandb-summary.json with policy end
+2024-04-12 08:50:48,666 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