Training in progress, step 14200

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:01e9cbaceb5831adbecae784c9a136002a89e2a397f162dc5d34feccdcfc1987
+oid sha256:c223a46d77a0541dcf420db07f84a2df91cb6efadc0cbaf5a4f31fc8b2d4bda2
 size 13982248

runs/Apr12_11-58-09_f044666cc877/events.out.tfevents.1712923163.f044666cc877.110.0

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:11cf228bdd224c67609f551dd34b0f6eb3ccc42e7c0c7fe3aceaca37bc0aade0
-size 154857
+oid sha256:b8deed49f9be29bb061508eaf1c01033fb07729d17a61a5f003782e262b0edd0
+size 155068

wandb/debug-internal.log

@@ -10624,3 +10624,15 @@ subprocess.TimeoutExpired: Command '['conda', 'env', 'export']' timed out after
 2024-04-12 16:08:52,054 INFO    Thread-12 :151 [dir_watcher.py:_on_file_modified():288] file/dir modified: /kaggle/working/wandb/run-20240412_115951-o0z5y86z/files/wandb-summary.json
 2024-04-12 16:08:52,790 DEBUG   HandlerThread:151 [handler.py:handle_request():146] handle_request: status_report
 2024-04-12 16:08:52,850 DEBUG   SenderThread:151 [sender.py:send():379] send: stats
+2024-04-12 16:08:55,055 INFO    Thread-12 :151 [dir_watcher.py:_on_file_modified():288] file/dir modified: /kaggle/working/wandb/run-20240412_115951-o0z5y86z/files/output.log
+2024-04-12 16:08:57,158 DEBUG   HandlerThread:151 [handler.py:handle_request():146] handle_request: internal_messages
+2024-04-12 16:08:57,161 DEBUG   HandlerThread:151 [handler.py:handle_request():146] handle_request: stop_status
+2024-04-12 16:08:57,161 DEBUG   SenderThread:151 [sender.py:send_request():406] send_request: stop_status
+2024-04-12 16:08:58,241 DEBUG   HandlerThread:151 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 16:09:03,242 DEBUG   HandlerThread:151 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 16:09:08,243 DEBUG   HandlerThread:151 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 16:09:10,332 DEBUG   HandlerThread:151 [handler.py:handle_request():146] handle_request: partial_history
+2024-04-12 16:09:10,333 DEBUG   SenderThread:151 [sender.py:send():379] send: history
+2024-04-12 16:09:10,333 DEBUG   SenderThread:151 [sender.py:send_request():406] send_request: summary_record
+2024-04-12 16:09:10,334 INFO    SenderThread:151 [sender.py:_save_file():1390] saving file wandb-summary.json with policy end
+2024-04-12 16:09:11,061 INFO    Thread-12 :151 [dir_watcher.py:_on_file_modified():288] file/dir modified: /kaggle/working/wandb/run-20240412_115951-o0z5y86z/files/wandb-summary.json

wandb/run-20240412_115951-o0z5y86z/files/output.log

@@ -1418,4 +1418,8 @@
   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-20240412_115951-o0z5y86z/files/wandb-summary.json

@@ -1 +1 @@
-{"train/loss": 3.4774, "train/grad_norm": 35.55162048339844, "train/learning_rate": 1.0305368692688174e-05, "train/epoch": 0.2, "train/global_step": 14180, "_timestamp": 1712938132.011107, "_runtime": 14940.420136928558, "_step": 708}
+{"train/loss": 3.5437, "train/grad_norm": 23.29683494567871, "train/learning_rate": 1.0239940674851941e-05, "train/epoch": 0.2, "train/global_step": 14200, "_timestamp": 1712938150.3315828, "_runtime": 14958.740612745285, "_step": 709}

wandb/run-20240412_115951-o0z5y86z/logs/debug-internal.log

@@ -10624,3 +10624,15 @@ subprocess.TimeoutExpired: Command '['conda', 'env', 'export']' timed out after
 2024-04-12 16:08:52,054 INFO    Thread-12 :151 [dir_watcher.py:_on_file_modified():288] file/dir modified: /kaggle/working/wandb/run-20240412_115951-o0z5y86z/files/wandb-summary.json
 2024-04-12 16:08:52,790 DEBUG   HandlerThread:151 [handler.py:handle_request():146] handle_request: status_report
 2024-04-12 16:08:52,850 DEBUG   SenderThread:151 [sender.py:send():379] send: stats
+2024-04-12 16:08:55,055 INFO    Thread-12 :151 [dir_watcher.py:_on_file_modified():288] file/dir modified: /kaggle/working/wandb/run-20240412_115951-o0z5y86z/files/output.log
+2024-04-12 16:08:57,158 DEBUG   HandlerThread:151 [handler.py:handle_request():146] handle_request: internal_messages
+2024-04-12 16:08:57,161 DEBUG   HandlerThread:151 [handler.py:handle_request():146] handle_request: stop_status
+2024-04-12 16:08:57,161 DEBUG   SenderThread:151 [sender.py:send_request():406] send_request: stop_status
+2024-04-12 16:08:58,241 DEBUG   HandlerThread:151 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 16:09:03,242 DEBUG   HandlerThread:151 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 16:09:08,243 DEBUG   HandlerThread:151 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 16:09:10,332 DEBUG   HandlerThread:151 [handler.py:handle_request():146] handle_request: partial_history
+2024-04-12 16:09:10,333 DEBUG   SenderThread:151 [sender.py:send():379] send: history
+2024-04-12 16:09:10,333 DEBUG   SenderThread:151 [sender.py:send_request():406] send_request: summary_record
+2024-04-12 16:09:10,334 INFO    SenderThread:151 [sender.py:_save_file():1390] saving file wandb-summary.json with policy end
+2024-04-12 16:09:11,061 INFO    Thread-12 :151 [dir_watcher.py:_on_file_modified():288] file/dir modified: /kaggle/working/wandb/run-20240412_115951-o0z5y86z/files/wandb-summary.json