Training in progress, step 220

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e1143902ff3564e4c1002e08abff0e3901f7b2864ff1561402b11204e5362411
+oid sha256:3078213dcd15bc291dc37ca3077f9efb155cb015e2ecfc039f1079c016cd9a50
 size 4725640

runs/Apr14_08-02-48_22f82e031f6d/events.out.tfevents.1713081770.22f82e031f6d.90.0

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:89fcfbf6e8d07e1caa1bc23d79d7edf69653bab62411e6172ca4d86ebf90eb8e
-size 23145
+oid sha256:88fcd822406ee94e8874fc7bd858d3212ae4d93b5d3d4d7e675368d5519d6666
+size 23356

wandb/debug-internal.log

@@ -263,3 +263,16 @@ subprocess.TimeoutExpired: Command '['conda', 'env', 'export']' timed out after
 2024-04-14 08:14:06,633 INFO    SenderThread:143 [sender.py:_save_file():1390] saving file wandb-summary.json with policy end
 2024-04-14 08:14:07,589 INFO    Thread-12 :143 [dir_watcher.py:_on_file_modified():288] file/dir modified: /kaggle/working/wandb/run-20240414_080844-fw7v7n5v/files/wandb-summary.json
 2024-04-14 08:14:08,286 DEBUG   HandlerThread:143 [handler.py:handle_request():146] handle_request: status_report
+2024-04-14 08:14:09,590 INFO    Thread-12 :143 [dir_watcher.py:_on_file_modified():288] file/dir modified: /kaggle/working/wandb/run-20240414_080844-fw7v7n5v/files/output.log
+2024-04-14 08:14:13,287 DEBUG   HandlerThread:143 [handler.py:handle_request():146] handle_request: status_report
+2024-04-14 08:14:15,580 DEBUG   SenderThread:143 [sender.py:send():379] send: stats
+2024-04-14 08:14:16,708 DEBUG   HandlerThread:143 [handler.py:handle_request():146] handle_request: stop_status
+2024-04-14 08:14:16,708 DEBUG   SenderThread:143 [sender.py:send_request():406] send_request: stop_status
+2024-04-14 08:14:16,709 DEBUG   HandlerThread:143 [handler.py:handle_request():146] handle_request: internal_messages
+2024-04-14 08:14:18,798 DEBUG   HandlerThread:143 [handler.py:handle_request():146] handle_request: status_report
+2024-04-14 08:14:23,799 DEBUG   HandlerThread:143 [handler.py:handle_request():146] handle_request: status_report
+2024-04-14 08:14:27,910 DEBUG   HandlerThread:143 [handler.py:handle_request():146] handle_request: partial_history
+2024-04-14 08:14:27,913 DEBUG   SenderThread:143 [sender.py:send():379] send: history
+2024-04-14 08:14:27,913 DEBUG   SenderThread:143 [sender.py:send_request():406] send_request: summary_record
+2024-04-14 08:14:27,914 INFO    SenderThread:143 [sender.py:_save_file():1390] saving file wandb-summary.json with policy end
+2024-04-14 08:14:28,597 INFO    Thread-12 :143 [dir_watcher.py:_on_file_modified():288] file/dir modified: /kaggle/working/wandb/run-20240414_080844-fw7v7n5v/files/wandb-summary.json

wandb/run-20240414_080844-fw7v7n5v/files/output.log

@@ -18,4 +18,8 @@
   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-20240414_080844-fw7v7n5v/files/wandb-summary.json

@@ -1 +1 @@
-{"train/loss": 3.0499, "train/grad_norm": 5.890682220458984, "train/learning_rate": 3.316666666666667e-05, "train/epoch": 0.0, "train/global_step": 200, "_timestamp": 1713082446.6311367, "_runtime": 321.87437868118286, "_step": 10}
+{"train/loss": 2.8568, "train/grad_norm": 6.314597129821777, "train/learning_rate": 3.65e-05, "train/epoch": 0.0, "train/global_step": 220, "_timestamp": 1713082467.9102666, "_runtime": 343.1535086631775, "_step": 11}

wandb/run-20240414_080844-fw7v7n5v/logs/debug-internal.log

@@ -263,3 +263,16 @@ subprocess.TimeoutExpired: Command '['conda', 'env', 'export']' timed out after
 2024-04-14 08:14:06,633 INFO    SenderThread:143 [sender.py:_save_file():1390] saving file wandb-summary.json with policy end
 2024-04-14 08:14:07,589 INFO    Thread-12 :143 [dir_watcher.py:_on_file_modified():288] file/dir modified: /kaggle/working/wandb/run-20240414_080844-fw7v7n5v/files/wandb-summary.json
 2024-04-14 08:14:08,286 DEBUG   HandlerThread:143 [handler.py:handle_request():146] handle_request: status_report
+2024-04-14 08:14:09,590 INFO    Thread-12 :143 [dir_watcher.py:_on_file_modified():288] file/dir modified: /kaggle/working/wandb/run-20240414_080844-fw7v7n5v/files/output.log
+2024-04-14 08:14:13,287 DEBUG   HandlerThread:143 [handler.py:handle_request():146] handle_request: status_report
+2024-04-14 08:14:15,580 DEBUG   SenderThread:143 [sender.py:send():379] send: stats
+2024-04-14 08:14:16,708 DEBUG   HandlerThread:143 [handler.py:handle_request():146] handle_request: stop_status
+2024-04-14 08:14:16,708 DEBUG   SenderThread:143 [sender.py:send_request():406] send_request: stop_status
+2024-04-14 08:14:16,709 DEBUG   HandlerThread:143 [handler.py:handle_request():146] handle_request: internal_messages
+2024-04-14 08:14:18,798 DEBUG   HandlerThread:143 [handler.py:handle_request():146] handle_request: status_report
+2024-04-14 08:14:23,799 DEBUG   HandlerThread:143 [handler.py:handle_request():146] handle_request: status_report
+2024-04-14 08:14:27,910 DEBUG   HandlerThread:143 [handler.py:handle_request():146] handle_request: partial_history
+2024-04-14 08:14:27,913 DEBUG   SenderThread:143 [sender.py:send():379] send: history
+2024-04-14 08:14:27,913 DEBUG   SenderThread:143 [sender.py:send_request():406] send_request: summary_record
+2024-04-14 08:14:27,914 INFO    SenderThread:143 [sender.py:_save_file():1390] saving file wandb-summary.json with policy end
+2024-04-14 08:14:28,597 INFO    Thread-12 :143 [dir_watcher.py:_on_file_modified():288] file/dir modified: /kaggle/working/wandb/run-20240414_080844-fw7v7n5v/files/wandb-summary.json