Training in progress, step 3080

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bb7001aeafada07c6cd57f862de0b34bee4825e37ea589867f484c999c59a079
+oid sha256:b844370e8571eb2f7a300a69ee5cc0b2b0716d9ce46e702df5730009ba58458c
 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 sha256:47881ab54c1e7922ac010e1ea7f53e33d494e89a6839cbb7ecacd2d990a27b8e
-size 37541
+oid sha256:247524ed1047f5c8303abb5081f983ce3308b03a861f81137bbd4da5613f6bda
+size 37752

wandb/debug-internal.log

@@ -2570,3 +2570,21 @@ subprocess.TimeoutExpired: Command '['conda', 'env', 'export']' timed out after
 2024-04-12 08:43:23,586 DEBUG   SenderThread:485 [sender.py:send_request():406] send_request: summary_record
 2024-04-12 08:43:23,587 INFO    SenderThread:485 [sender.py:_save_file():1390] saving file wandb-summary.json with policy end
 2024-04-12 08:43:24,484 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 08:43:25,340 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 08:43:26,485 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 08:43:30,341 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 08:43:31,769 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: stop_status
+2024-04-12 08:43:31,770 DEBUG   SenderThread:485 [sender.py:send_request():406] send_request: stop_status
+2024-04-12 08:43:31,782 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: internal_messages
+2024-04-12 08:43:35,842 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 08:43:40,843 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 08:43:45,157 DEBUG   SenderThread:485 [sender.py:send():379] send: stats
+2024-04-12 08:43:46,158 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 08:43:46,767 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: stop_status
+2024-04-12 08:43:46,767 DEBUG   SenderThread:485 [sender.py:send_request():406] send_request: stop_status
+2024-04-12 08:43:46,808 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: internal_messages
+2024-04-12 08:43:48,739 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: partial_history
+2024-04-12 08:43:48,743 DEBUG   SenderThread:485 [sender.py:send():379] send: history
+2024-04-12 08:43:48,743 DEBUG   SenderThread:485 [sender.py:send_request():406] send_request: summary_record
+2024-04-12 08:43:48,744 INFO    SenderThread:485 [sender.py:_save_file():1390] saving file wandb-summary.json with policy end
+2024-04-12 08:43:49,494 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

@@ -302,4 +302,6 @@
   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": 3.5578, "train/grad_norm": 57.92033004760742, "train/learning_rate": 2.882543584825435e-05, "train/epoch": 0.24, "train/global_step": 3060, "_timestamp": 1712911403.5848355, "_runtime": 3759.0771174430847, "_step": 152}
+{"train/loss": 3.577, "train/grad_norm": 99.40699005126953, "train/learning_rate": 2.880651800951616e-05, "train/epoch": 0.25, "train/global_step": 3080, "_timestamp": 1712911428.7393134, "_runtime": 3784.2315952777863, "_step": 153}

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

@@ -2570,3 +2570,21 @@ subprocess.TimeoutExpired: Command '['conda', 'env', 'export']' timed out after
 2024-04-12 08:43:23,586 DEBUG   SenderThread:485 [sender.py:send_request():406] send_request: summary_record
 2024-04-12 08:43:23,587 INFO    SenderThread:485 [sender.py:_save_file():1390] saving file wandb-summary.json with policy end
 2024-04-12 08:43:24,484 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 08:43:25,340 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 08:43:26,485 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 08:43:30,341 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 08:43:31,769 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: stop_status
+2024-04-12 08:43:31,770 DEBUG   SenderThread:485 [sender.py:send_request():406] send_request: stop_status
+2024-04-12 08:43:31,782 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: internal_messages
+2024-04-12 08:43:35,842 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 08:43:40,843 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 08:43:45,157 DEBUG   SenderThread:485 [sender.py:send():379] send: stats
+2024-04-12 08:43:46,158 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 08:43:46,767 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: stop_status
+2024-04-12 08:43:46,767 DEBUG   SenderThread:485 [sender.py:send_request():406] send_request: stop_status
+2024-04-12 08:43:46,808 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: internal_messages
+2024-04-12 08:43:48,739 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: partial_history
+2024-04-12 08:43:48,743 DEBUG   SenderThread:485 [sender.py:send():379] send: history
+2024-04-12 08:43:48,743 DEBUG   SenderThread:485 [sender.py:send_request():406] send_request: summary_record
+2024-04-12 08:43:48,744 INFO    SenderThread:485 [sender.py:_save_file():1390] saving file wandb-summary.json with policy end
+2024-04-12 08:43:49,494 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