Training in progress, step 7380

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d43dac100a478777d7327b0006c122763d2abb80886baee0035eb613a9d90058
+oid sha256:0c291b58c13f93a64f1b6f82d56e113a222f9938bd3dcf1a4e14b09d3ad485a2
 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:ce3a453e9a3070bc2949837988e1a22a912779f3e26d12c35f62775828b2588a
-size 82906
+oid sha256:815382879fd03330b42542c6a1a2fd32be100c1c6332b124ad3e27f6e4850eb9
+size 83117

wandb/debug-internal.log

@@ -6086,3 +6086,19 @@ subprocess.TimeoutExpired: Command '['conda', 'env', 'export']' timed out after
 2024-04-12 10:10:58,472 INFO    SenderThread:485 [sender.py:_save_file():1390] saving file wandb-summary.json with policy end
 2024-04-12 10:10:58,699 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 10:10:59,702 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 10:11:02,148 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: internal_messages
+2024-04-12 10:11:02,723 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: stop_status
+2024-04-12 10:11:02,723 DEBUG   SenderThread:485 [sender.py:send_request():406] send_request: stop_status
+2024-04-12 10:11:03,831 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 10:11:08,832 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 10:11:13,833 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 10:11:15,338 DEBUG   SenderThread:485 [sender.py:send():379] send: stats
+2024-04-12 10:11:17,148 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: internal_messages
+2024-04-12 10:11:17,725 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: stop_status
+2024-04-12 10:11:17,726 DEBUG   SenderThread:485 [sender.py:send_request():406] send_request: stop_status
+2024-04-12 10:11:19,751 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 10:11:22,649 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: partial_history
+2024-04-12 10:11:22,654 DEBUG   SenderThread:485 [sender.py:send():379] send: history
+2024-04-12 10:11:22,655 DEBUG   SenderThread:485 [sender.py:send_request():406] send_request: summary_record
+2024-04-12 10:11:22,655 INFO    SenderThread:485 [sender.py:_save_file():1390] saving file wandb-summary.json with policy end
+2024-04-12 10:11:22,710 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

@@ -734,4 +734,6 @@
   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": 3.4692, "train/grad_norm": 25.45415687561035, "train/learning_rate": 2.187470630213845e-05, "train/epoch": 0.59, "train/global_step": 7360, "_timestamp": 1712916658.4701746, "_runtime": 9013.962456464767, "_step": 367}
+{"train/loss": 3.5714, "train/grad_norm": 30.72098159790039, "train/learning_rate": 2.1831491698460923e-05, "train/epoch": 0.59, "train/global_step": 7380, "_timestamp": 1712916682.6488101, "_runtime": 9038.141092061996, "_step": 368}

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

@@ -6086,3 +6086,19 @@ subprocess.TimeoutExpired: Command '['conda', 'env', 'export']' timed out after
 2024-04-12 10:10:58,472 INFO    SenderThread:485 [sender.py:_save_file():1390] saving file wandb-summary.json with policy end
 2024-04-12 10:10:58,699 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 10:10:59,702 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 10:11:02,148 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: internal_messages
+2024-04-12 10:11:02,723 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: stop_status
+2024-04-12 10:11:02,723 DEBUG   SenderThread:485 [sender.py:send_request():406] send_request: stop_status
+2024-04-12 10:11:03,831 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 10:11:08,832 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 10:11:13,833 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 10:11:15,338 DEBUG   SenderThread:485 [sender.py:send():379] send: stats
+2024-04-12 10:11:17,148 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: internal_messages
+2024-04-12 10:11:17,725 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: stop_status
+2024-04-12 10:11:17,726 DEBUG   SenderThread:485 [sender.py:send_request():406] send_request: stop_status
+2024-04-12 10:11:19,751 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 10:11:22,649 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: partial_history
+2024-04-12 10:11:22,654 DEBUG   SenderThread:485 [sender.py:send():379] send: history
+2024-04-12 10:11:22,655 DEBUG   SenderThread:485 [sender.py:send_request():406] send_request: summary_record
+2024-04-12 10:11:22,655 INFO    SenderThread:485 [sender.py:_save_file():1390] saving file wandb-summary.json with policy end
+2024-04-12 10:11:22,710 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