Training in progress, step 320

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:29dd7ebae639bd2d9820d94f55a7d57ed7d7e651db6caee9214dbc394d6d8a52
+oid sha256:dfd12fc64f06bad2fa92746516b8664309c4466b97e18621e506d23fda100aa8
 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:ae5773e554f094ce81367b16b38c0c0ec338753900f3e03ebe23e2677d114d98
-size 24200
+oid sha256:6b0b8ece91c15d46e7b95880c71f9152fe5d1b6c8f4d4efd66b601834caf961a
+size 24411

wandb/debug-internal.log

@@ -337,3 +337,16 @@ subprocess.TimeoutExpired: Command '['conda', 'env', 'export']' timed out after
 2024-04-14 08:15:51,233 DEBUG   SenderThread:143 [sender.py:send_request():406] send_request: summary_record
 2024-04-14 08:15:51,233 INFO    SenderThread:143 [sender.py:_save_file():1390] saving file wandb-summary.json with policy end
 2024-04-14 08:15:51,627 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 08:15:52,983 DEBUG   HandlerThread:143 [handler.py:handle_request():146] handle_request: status_report
+2024-04-14 08:15:53,628 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 08:15:57,983 DEBUG   HandlerThread:143 [handler.py:handle_request():146] handle_request: status_report
+2024-04-14 08:16:01,709 DEBUG   HandlerThread:143 [handler.py:handle_request():146] handle_request: stop_status
+2024-04-14 08:16:01,709 DEBUG   HandlerThread:143 [handler.py:handle_request():146] handle_request: internal_messages
+2024-04-14 08:16:01,710 DEBUG   SenderThread:143 [sender.py:send_request():406] send_request: stop_status
+2024-04-14 08:16:03,740 DEBUG   HandlerThread:143 [handler.py:handle_request():146] handle_request: status_report
+2024-04-14 08:16:08,741 DEBUG   HandlerThread:143 [handler.py:handle_request():146] handle_request: status_report
+2024-04-14 08:16:11,993 DEBUG   HandlerThread:143 [handler.py:handle_request():146] handle_request: partial_history
+2024-04-14 08:16:11,994 DEBUG   SenderThread:143 [sender.py:send():379] send: history
+2024-04-14 08:16:11,994 DEBUG   SenderThread:143 [sender.py:send_request():406] send_request: summary_record
+2024-04-14 08:16:11,994 INFO    SenderThread:143 [sender.py:_save_file():1390] saving file wandb-summary.json with policy end
+2024-04-14 08:16:12,636 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

@@ -28,4 +28,8 @@
   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-20240414_080844-fw7v7n5v/files/wandb-summary.json

@@ -1 +1 @@
-{"train/loss": 2.7932, "train/grad_norm": 8.951983451843262, "train/learning_rate": 4.9833333333333336e-05, "train/epoch": 0.0, "train/global_step": 300, "_timestamp": 1713082551.2315898, "_runtime": 426.4748318195343, "_step": 15}
+{"train/loss": 2.9341, "train/grad_norm": 4.788575172424316, "train/learning_rate": 4.9999526661182696e-05, "train/epoch": 0.0, "train/global_step": 320, "_timestamp": 1713082571.9924576, "_runtime": 447.2356996536255, "_step": 16}

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

@@ -337,3 +337,16 @@ subprocess.TimeoutExpired: Command '['conda', 'env', 'export']' timed out after
 2024-04-14 08:15:51,233 DEBUG   SenderThread:143 [sender.py:send_request():406] send_request: summary_record
 2024-04-14 08:15:51,233 INFO    SenderThread:143 [sender.py:_save_file():1390] saving file wandb-summary.json with policy end
 2024-04-14 08:15:51,627 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 08:15:52,983 DEBUG   HandlerThread:143 [handler.py:handle_request():146] handle_request: status_report
+2024-04-14 08:15:53,628 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 08:15:57,983 DEBUG   HandlerThread:143 [handler.py:handle_request():146] handle_request: status_report
+2024-04-14 08:16:01,709 DEBUG   HandlerThread:143 [handler.py:handle_request():146] handle_request: stop_status
+2024-04-14 08:16:01,709 DEBUG   HandlerThread:143 [handler.py:handle_request():146] handle_request: internal_messages
+2024-04-14 08:16:01,710 DEBUG   SenderThread:143 [sender.py:send_request():406] send_request: stop_status
+2024-04-14 08:16:03,740 DEBUG   HandlerThread:143 [handler.py:handle_request():146] handle_request: status_report
+2024-04-14 08:16:08,741 DEBUG   HandlerThread:143 [handler.py:handle_request():146] handle_request: status_report
+2024-04-14 08:16:11,993 DEBUG   HandlerThread:143 [handler.py:handle_request():146] handle_request: partial_history
+2024-04-14 08:16:11,994 DEBUG   SenderThread:143 [sender.py:send():379] send: history
+2024-04-14 08:16:11,994 DEBUG   SenderThread:143 [sender.py:send_request():406] send_request: summary_record
+2024-04-14 08:16:11,994 INFO    SenderThread:143 [sender.py:_save_file():1390] saving file wandb-summary.json with policy end
+2024-04-14 08:16:12,636 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