Training in progress, step 320

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7033db835b59eeb254ec6852c415ba6a6a376c54a02d3c672ca0f3404967542e
+oid sha256:364f6da3a4ebddec454e8c37aa50aa9c92120309c0c99ba6b58ce7277c88d74e
 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 sha256:89e58559a2b7961fd68753a1d41f479fe7e0a7e0eb94382aedec29cf53731434
-size 8423
+oid sha256:336f0fec34190fab0f3e6660a6578e58cfdffb17d0039d3a103bb83a98c9c2f9
+size 8634

wandb/debug-internal.log

@@ -314,3 +314,20 @@ subprocess.TimeoutExpired: Command '['conda', 'env', 'export']' timed out after
 2024-04-12 07:47:11,147 DEBUG   SenderThread:485 [sender.py:send_request():406] send_request: summary_record
 2024-04-12 07:47:11,147 INFO    SenderThread:485 [sender.py:_save_file():1390] saving file wandb-summary.json with policy end
 2024-04-12 07:47:12,063 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 07:47:14,064 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 07:47:15,028 DEBUG   SenderThread:485 [sender.py:send():379] send: stats
+2024-04-12 07:47:16,028 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 07:47:16,041 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: internal_messages
+2024-04-12 07:47:16,041 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: stop_status
+2024-04-12 07:47:16,042 DEBUG   SenderThread:485 [sender.py:send_request():406] send_request: stop_status
+2024-04-12 07:47:21,127 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 07:47:26,128 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 07:47:31,041 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: internal_messages
+2024-04-12 07:47:31,041 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: stop_status
+2024-04-12 07:47:31,042 DEBUG   SenderThread:485 [sender.py:send_request():406] send_request: stop_status
+2024-04-12 07:47:32,121 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 07:47:35,137 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: partial_history
+2024-04-12 07:47:35,138 DEBUG   SenderThread:485 [sender.py:send():379] send: history
+2024-04-12 07:47:35,138 DEBUG   SenderThread:485 [sender.py:send_request():406] send_request: summary_record
+2024-04-12 07:47:35,139 INFO    SenderThread:485 [sender.py:_save_file():1390] saving file wandb-summary.json with policy end
+2024-04-12 07:47:36,075 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

@@ -26,4 +26,6 @@
   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": 4.2198, "train/grad_norm": 28.304485321044922, "train/learning_rate": 1.5e-05, "train/epoch": 0.02, "train/global_step": 300, "_timestamp": 1712908031.1453025, "_runtime": 386.6375844478607, "_step": 14}
+{"train/loss": 4.0958, "train/grad_norm": 34.03230285644531, "train/learning_rate": 1.6e-05, "train/epoch": 0.03, "train/global_step": 320, "_timestamp": 1712908055.1366203, "_runtime": 410.62890219688416, "_step": 15}

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

@@ -314,3 +314,20 @@ subprocess.TimeoutExpired: Command '['conda', 'env', 'export']' timed out after
 2024-04-12 07:47:11,147 DEBUG   SenderThread:485 [sender.py:send_request():406] send_request: summary_record
 2024-04-12 07:47:11,147 INFO    SenderThread:485 [sender.py:_save_file():1390] saving file wandb-summary.json with policy end
 2024-04-12 07:47:12,063 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 07:47:14,064 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 07:47:15,028 DEBUG   SenderThread:485 [sender.py:send():379] send: stats
+2024-04-12 07:47:16,028 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 07:47:16,041 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: internal_messages
+2024-04-12 07:47:16,041 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: stop_status
+2024-04-12 07:47:16,042 DEBUG   SenderThread:485 [sender.py:send_request():406] send_request: stop_status
+2024-04-12 07:47:21,127 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 07:47:26,128 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 07:47:31,041 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: internal_messages
+2024-04-12 07:47:31,041 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: stop_status
+2024-04-12 07:47:31,042 DEBUG   SenderThread:485 [sender.py:send_request():406] send_request: stop_status
+2024-04-12 07:47:32,121 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 07:47:35,137 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: partial_history
+2024-04-12 07:47:35,138 DEBUG   SenderThread:485 [sender.py:send():379] send: history
+2024-04-12 07:47:35,138 DEBUG   SenderThread:485 [sender.py:send_request():406] send_request: summary_record
+2024-04-12 07:47:35,139 INFO    SenderThread:485 [sender.py:_save_file():1390] saving file wandb-summary.json with policy end
+2024-04-12 07:47:36,075 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