Training in progress, step 18520

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e31afd1c461e20344e65e088c22c1fa2b87f37ae229e186c8dd536291662f3e0
+oid sha256:1ee3c331e12cc90d2109f4e1480b06ec8c2e15119b962650d98dc8e9275f0114
 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 sha256:10475b3e0f3b71c120ca0c92e54a9ae16bb2ecfc382ba913c9c073e0796369e8
-size 200857
+oid sha256:6ee9d53a8927d127dcbcdd7adda0c81b8ec82577095669e93f1c060abcd9b58b
+size 201072

wandb/debug-internal.log

@@ -13847,3 +13847,17 @@ subprocess.TimeoutExpired: Command '['conda', 'env', 'export']' timed out after
 2024-04-12 17:24:58,600 DEBUG   SenderThread:151 [sender.py:send_request():406] send_request: summary_record
 2024-04-12 17:24:58,601 INFO    SenderThread:151 [sender.py:_save_file():1390] saving file wandb-summary.json with policy end
 2024-04-12 17:24:58,853 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 17:25:01,854 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 17:25:03,359 DEBUG   HandlerThread:151 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 17:25:08,361 DEBUG   HandlerThread:151 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 17:25:12,510 DEBUG   HandlerThread:151 [handler.py:handle_request():146] handle_request: stop_status
+2024-04-12 17:25:12,510 DEBUG   HandlerThread:151 [handler.py:handle_request():146] handle_request: internal_messages
+2024-04-12 17:25:12,511 DEBUG   SenderThread:151 [sender.py:send_request():406] send_request: stop_status
+2024-04-12 17:25:13,583 DEBUG   HandlerThread:151 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 17:25:18,584 DEBUG   HandlerThread:151 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 17:25:21,636 DEBUG   HandlerThread:151 [handler.py:handle_request():146] handle_request: partial_history
+2024-04-12 17:25:21,638 DEBUG   SenderThread:151 [sender.py:send():379] send: history
+2024-04-12 17:25:21,638 DEBUG   SenderThread:151 [sender.py:send_request():406] send_request: summary_record
+2024-04-12 17:25:21,639 INFO    SenderThread:151 [sender.py:_save_file():1390] saving file wandb-summary.json with policy end
+2024-04-12 17:25:21,861 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 17:25:22,995 DEBUG   SenderThread:151 [sender.py:send():379] send: stats

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

@@ -1850,4 +1850,8 @@
   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": 3.4247, "train/grad_norm": 20.147789001464844, "train/learning_rate": 7.339261343510206e-07, "train/epoch": 0.26, "train/global_step": 18500, "_timestamp": 1712942698.5986316, "_runtime": 19507.00766158104, "_step": 924}
+{"train/loss": 3.5844, "train/grad_norm": 19.394601821899414, "train/learning_rate": 7.145784012061424e-07, "train/epoch": 0.26, "train/global_step": 18520, "_timestamp": 1712942721.6361759, "_runtime": 19530.045205831528, "_step": 925}

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

@@ -13847,3 +13847,17 @@ subprocess.TimeoutExpired: Command '['conda', 'env', 'export']' timed out after
 2024-04-12 17:24:58,600 DEBUG   SenderThread:151 [sender.py:send_request():406] send_request: summary_record
 2024-04-12 17:24:58,601 INFO    SenderThread:151 [sender.py:_save_file():1390] saving file wandb-summary.json with policy end
 2024-04-12 17:24:58,853 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 17:25:01,854 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 17:25:03,359 DEBUG   HandlerThread:151 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 17:25:08,361 DEBUG   HandlerThread:151 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 17:25:12,510 DEBUG   HandlerThread:151 [handler.py:handle_request():146] handle_request: stop_status
+2024-04-12 17:25:12,510 DEBUG   HandlerThread:151 [handler.py:handle_request():146] handle_request: internal_messages
+2024-04-12 17:25:12,511 DEBUG   SenderThread:151 [sender.py:send_request():406] send_request: stop_status
+2024-04-12 17:25:13,583 DEBUG   HandlerThread:151 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 17:25:18,584 DEBUG   HandlerThread:151 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 17:25:21,636 DEBUG   HandlerThread:151 [handler.py:handle_request():146] handle_request: partial_history
+2024-04-12 17:25:21,638 DEBUG   SenderThread:151 [sender.py:send():379] send: history
+2024-04-12 17:25:21,638 DEBUG   SenderThread:151 [sender.py:send_request():406] send_request: summary_record
+2024-04-12 17:25:21,639 INFO    SenderThread:151 [sender.py:_save_file():1390] saving file wandb-summary.json with policy end
+2024-04-12 17:25:21,861 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 17:25:22,995 DEBUG   SenderThread:151 [sender.py:send():379] send: stats

wandb/run-20240412_115951-o0z5y86z/run-o0z5y86z.wandb

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a3fbed7d9f3c59baba307b6aa910f9ad51ebe469259a9213f9a7d6e83a15c6dc
-size 1299986
+oid sha256:b8ef47c775ca6318a932a149b43d758f888d75599b45e13d35bbc99f4a5276ba
+size 1303574