Training in progress, step 380

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:be1c47e7c45521d38d74b6126649ceb2df1743106d5e52fc59729f4ddcf09e52
+oid sha256:ac7c7b067ce43a0be0be0f92ea01066fe8ed60f6f5f80359ee347257fbaa14a6
 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 sha256:9d3779c73e056dddcffe20a7ac5abfddb0b9e0e93c2792b362874e939c548d52
-size 24833
+oid sha256:fd026ddf9da167843074a9c6242bfc15e90c41bae0ccb4d5957ca813cd0186e1
+size 25044

wandb/debug-internal.log

@@ -381,3 +381,16 @@ subprocess.TimeoutExpired: Command '['conda', 'env', 'export']' timed out after
 2024-04-14 08:16:53,410 DEBUG   SenderThread:143 [sender.py:send_request():406] send_request: summary_record
 2024-04-14 08:16:53,412 INFO    SenderThread:143 [sender.py:_save_file():1390] saving file wandb-summary.json with policy end
 2024-04-14 08:16:53,652 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 08:16:55,653 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 08:16:56,109 DEBUG   HandlerThread:143 [handler.py:handle_request():146] handle_request: status_report
+2024-04-14 08:17:01,110 DEBUG   HandlerThread:143 [handler.py:handle_request():146] handle_request: status_report
+2024-04-14 08:17:01,710 DEBUG   HandlerThread:143 [handler.py:handle_request():146] handle_request: stop_status
+2024-04-14 08:17:01,710 DEBUG   SenderThread:143 [sender.py:send_request():406] send_request: stop_status
+2024-04-14 08:17:01,714 DEBUG   HandlerThread:143 [handler.py:handle_request():146] handle_request: internal_messages
+2024-04-14 08:17:06,816 DEBUG   HandlerThread:143 [handler.py:handle_request():146] handle_request: status_report
+2024-04-14 08:17:11,818 DEBUG   HandlerThread:143 [handler.py:handle_request():146] handle_request: status_report
+2024-04-14 08:17:14,160 DEBUG   HandlerThread:143 [handler.py:handle_request():146] handle_request: partial_history
+2024-04-14 08:17:14,162 DEBUG   SenderThread:143 [sender.py:send():379] send: history
+2024-04-14 08:17:14,162 DEBUG   SenderThread:143 [sender.py:send_request():406] send_request: summary_record
+2024-04-14 08:17:14,162 INFO    SenderThread:143 [sender.py:_save_file():1390] saving file wandb-summary.json with policy end
+2024-04-14 08:17:14,660 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

@@ -34,4 +34,8 @@
   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": 2.7863, "train/grad_norm": 4.9223952293396, "train/learning_rate": 4.9995435879539254e-05, "train/epoch": 0.0, "train/global_step": 360, "_timestamp": 1713082613.4089081, "_runtime": 488.65215015411377, "_step": 18}
+{"train/loss": 2.6197, "train/grad_norm": 7.647037506103516, "train/learning_rate": 4.999181729716214e-05, "train/epoch": 0.0, "train/global_step": 380, "_timestamp": 1713082634.160211, "_runtime": 509.40345311164856, "_step": 19}

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

@@ -381,3 +381,16 @@ subprocess.TimeoutExpired: Command '['conda', 'env', 'export']' timed out after
 2024-04-14 08:16:53,410 DEBUG   SenderThread:143 [sender.py:send_request():406] send_request: summary_record
 2024-04-14 08:16:53,412 INFO    SenderThread:143 [sender.py:_save_file():1390] saving file wandb-summary.json with policy end
 2024-04-14 08:16:53,652 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 08:16:55,653 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 08:16:56,109 DEBUG   HandlerThread:143 [handler.py:handle_request():146] handle_request: status_report
+2024-04-14 08:17:01,110 DEBUG   HandlerThread:143 [handler.py:handle_request():146] handle_request: status_report
+2024-04-14 08:17:01,710 DEBUG   HandlerThread:143 [handler.py:handle_request():146] handle_request: stop_status
+2024-04-14 08:17:01,710 DEBUG   SenderThread:143 [sender.py:send_request():406] send_request: stop_status
+2024-04-14 08:17:01,714 DEBUG   HandlerThread:143 [handler.py:handle_request():146] handle_request: internal_messages
+2024-04-14 08:17:06,816 DEBUG   HandlerThread:143 [handler.py:handle_request():146] handle_request: status_report
+2024-04-14 08:17:11,818 DEBUG   HandlerThread:143 [handler.py:handle_request():146] handle_request: status_report
+2024-04-14 08:17:14,160 DEBUG   HandlerThread:143 [handler.py:handle_request():146] handle_request: partial_history
+2024-04-14 08:17:14,162 DEBUG   SenderThread:143 [sender.py:send():379] send: history
+2024-04-14 08:17:14,162 DEBUG   SenderThread:143 [sender.py:send_request():406] send_request: summary_record
+2024-04-14 08:17:14,162 INFO    SenderThread:143 [sender.py:_save_file():1390] saving file wandb-summary.json with policy end
+2024-04-14 08:17:14,660 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