Training in progress, step 120

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:434e9148a55d7de38db07465532dbaab0cdcfec5676d112c61532c1b6dbdca8d
+oid sha256:27c29bc43948c63a0f226cc87ab0660cea4ab6cd6ae33f4b3a1386449f13e33c
 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 sha256:16ceb75e1d112c8c03c4d54284c8e145bc2bf3178940914e8630516d14a73eb7
-size 22094
+oid sha256:f063564a98aed6f7c51b5d4905e2203dc03cfb732628502334e134f1dd6062a1
+size 22301

wandb/debug-internal.log

@@ -183,3 +183,21 @@ subprocess.TimeoutExpired: Command '['conda', 'env', 'export']' timed out after
 2024-04-14 08:12:12,239 DEBUG   SenderThread:143 [sender.py:send_request():406] send_request: summary_record
 2024-04-14 08:12:12,239 INFO    SenderThread:143 [sender.py:_save_file():1390] saving file wandb-summary.json with policy end
 2024-04-14 08:12:12,542 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 08:12:15,544 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 08:12:15,576 DEBUG   SenderThread:143 [sender.py:send():379] send: stats
+2024-04-14 08:12:16,577 DEBUG   HandlerThread:143 [handler.py:handle_request():146] handle_request: status_report
+2024-04-14 08:12:16,706 DEBUG   HandlerThread:143 [handler.py:handle_request():146] handle_request: stop_status
+2024-04-14 08:12:16,707 DEBUG   SenderThread:143 [sender.py:send_request():406] send_request: stop_status
+2024-04-14 08:12:16,709 DEBUG   HandlerThread:143 [handler.py:handle_request():146] handle_request: internal_messages
+2024-04-14 08:12:21,837 DEBUG   HandlerThread:143 [handler.py:handle_request():146] handle_request: status_report
+2024-04-14 08:12:26,838 DEBUG   HandlerThread:143 [handler.py:handle_request():146] handle_request: status_report
+2024-04-14 08:12:31,707 DEBUG   HandlerThread:143 [handler.py:handle_request():146] handle_request: stop_status
+2024-04-14 08:12:31,707 DEBUG   SenderThread:143 [sender.py:send_request():406] send_request: stop_status
+2024-04-14 08:12:31,747 DEBUG   HandlerThread:143 [handler.py:handle_request():146] handle_request: internal_messages
+2024-04-14 08:12:32,804 DEBUG   HandlerThread:143 [handler.py:handle_request():146] handle_request: status_report
+2024-04-14 08:12:37,569 DEBUG   HandlerThread:143 [handler.py:handle_request():146] handle_request: partial_history
+2024-04-14 08:12:37,570 DEBUG   SenderThread:143 [sender.py:send():379] send: history
+2024-04-14 08:12:37,570 DEBUG   SenderThread:143 [sender.py:send_request():406] send_request: summary_record
+2024-04-14 08:12:37,572 INFO    SenderThread:143 [sender.py:_save_file():1390] saving file wandb-summary.json with policy end
+2024-04-14 08:12:38,260 DEBUG   HandlerThread:143 [handler.py:handle_request():146] handle_request: status_report
+2024-04-14 08:12:38,553 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

@@ -8,4 +8,6 @@
   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-20240414_080844-fw7v7n5v/files/wandb-summary.json

@@ -1 +1 @@
-{"train/loss": 3.3337, "train/grad_norm": 1.9482054710388184, "train/learning_rate": 1.65e-05, "train/epoch": 0.0, "train/global_step": 100, "_timestamp": 1713082332.2370584, "_runtime": 207.48030042648315, "_step": 5}
+{"train/loss": 3.1903, "train/grad_norm": 4.047863006591797, "train/learning_rate": 1.9833333333333335e-05, "train/epoch": 0.0, "train/global_step": 120, "_timestamp": 1713082357.5686247, "_runtime": 232.8118667602539, "_step": 6}

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

@@ -183,3 +183,21 @@ subprocess.TimeoutExpired: Command '['conda', 'env', 'export']' timed out after
 2024-04-14 08:12:12,239 DEBUG   SenderThread:143 [sender.py:send_request():406] send_request: summary_record
 2024-04-14 08:12:12,239 INFO    SenderThread:143 [sender.py:_save_file():1390] saving file wandb-summary.json with policy end
 2024-04-14 08:12:12,542 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 08:12:15,544 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 08:12:15,576 DEBUG   SenderThread:143 [sender.py:send():379] send: stats
+2024-04-14 08:12:16,577 DEBUG   HandlerThread:143 [handler.py:handle_request():146] handle_request: status_report
+2024-04-14 08:12:16,706 DEBUG   HandlerThread:143 [handler.py:handle_request():146] handle_request: stop_status
+2024-04-14 08:12:16,707 DEBUG   SenderThread:143 [sender.py:send_request():406] send_request: stop_status
+2024-04-14 08:12:16,709 DEBUG   HandlerThread:143 [handler.py:handle_request():146] handle_request: internal_messages
+2024-04-14 08:12:21,837 DEBUG   HandlerThread:143 [handler.py:handle_request():146] handle_request: status_report
+2024-04-14 08:12:26,838 DEBUG   HandlerThread:143 [handler.py:handle_request():146] handle_request: status_report
+2024-04-14 08:12:31,707 DEBUG   HandlerThread:143 [handler.py:handle_request():146] handle_request: stop_status
+2024-04-14 08:12:31,707 DEBUG   SenderThread:143 [sender.py:send_request():406] send_request: stop_status
+2024-04-14 08:12:31,747 DEBUG   HandlerThread:143 [handler.py:handle_request():146] handle_request: internal_messages
+2024-04-14 08:12:32,804 DEBUG   HandlerThread:143 [handler.py:handle_request():146] handle_request: status_report
+2024-04-14 08:12:37,569 DEBUG   HandlerThread:143 [handler.py:handle_request():146] handle_request: partial_history
+2024-04-14 08:12:37,570 DEBUG   SenderThread:143 [sender.py:send():379] send: history
+2024-04-14 08:12:37,570 DEBUG   SenderThread:143 [sender.py:send_request():406] send_request: summary_record
+2024-04-14 08:12:37,572 INFO    SenderThread:143 [sender.py:_save_file():1390] saving file wandb-summary.json with policy end
+2024-04-14 08:12:38,260 DEBUG   HandlerThread:143 [handler.py:handle_request():146] handle_request: status_report
+2024-04-14 08:12:38,553 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