Training in progress, step 60

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:aeda3f74d0c65b865c78f72723c369a102c203856812079274186432e8ae2082
+oid sha256:251f273c1edf14934f3e7228786d8f001549a481dbddc1dae5ec66a5a4f114d2
 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 sha256:4090afe353b724cb1bf6ee84e984cc3099a34ad4ae39ea1d31362b0c34f051b0
-size 5696
+oid sha256:b52cdce23e15a8b110ccfc45aa7707a256cef332f953385d33e778533b4fb662
+size 5903

wandb/debug-internal.log

@@ -99,3 +99,18 @@ subprocess.TimeoutExpired: Command '['conda', 'env', 'export']' timed out after
 2024-04-12 07:41:46,287 DEBUG   SenderThread:485 [sender.py:send_request():406] send_request: summary_record
 2024-04-12 07:41:46,287 INFO    SenderThread:485 [sender.py:_save_file():1390] saving file wandb-summary.json with policy end
 2024-04-12 07:41:46,921 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 07:41:47,922 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 07:41:48,004 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 07:41:48,922 INFO    Thread-12 :485 [dir_watcher.py:_on_file_modified():288] file/dir modified: /kaggle/working/wandb/run-20240412_074044-widovcn3/files/config.yaml
+2024-04-12 07:41:53,106 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 07:41:58,107 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 07:42:01,035 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: stop_status
+2024-04-12 07:42:01,036 DEBUG   SenderThread:485 [sender.py:send_request():406] send_request: stop_status
+2024-04-12 07:42:01,038 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: internal_messages
+2024-04-12 07:42:03,131 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 07:42:08,132 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 07:42:09,300 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: partial_history
+2024-04-12 07:42:09,301 DEBUG   SenderThread:485 [sender.py:send():379] send: history
+2024-04-12 07:42:09,301 DEBUG   SenderThread:485 [sender.py:send_request():406] send_request: summary_record
+2024-04-12 07:42:09,303 INFO    SenderThread:485 [sender.py:_save_file():1390] saving file wandb-summary.json with policy end
+2024-04-12 07:42:09,932 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/config.yaml

@@ -58,6 +58,22 @@ _wandb:
     - 1: train/global_step
       6:
       - 3
+    - 1: train/loss
+      5: 1
+      6:
+      - 1
+    - 1: train/grad_norm
+      5: 1
+      6:
+      - 1
+    - 1: train/learning_rate
+      5: 1
+      6:
+      - 1
+    - 1: train/epoch
+      5: 1
+      6:
+      - 1
 vocab_size:
   desc: null
   value: 250880

wandb/run-20240412_074044-widovcn3/files/output.log

@@ -2,4 +2,6 @@
   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": 4.4077, "train/grad_norm": 39.39016342163086, "train/learning_rate": 2e-06, "train/epoch": 0.0, "train/global_step": 40, "_timestamp": 1712907706.285044, "_runtime": 61.77732586860657, "_step": 1}
+{"train/loss": 4.4807, "train/grad_norm": 54.24020767211914, "train/learning_rate": 3e-06, "train/epoch": 0.0, "train/global_step": 60, "_timestamp": 1712907729.2995105, "_runtime": 84.79179239273071, "_step": 2}

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

@@ -99,3 +99,18 @@ subprocess.TimeoutExpired: Command '['conda', 'env', 'export']' timed out after
 2024-04-12 07:41:46,287 DEBUG   SenderThread:485 [sender.py:send_request():406] send_request: summary_record
 2024-04-12 07:41:46,287 INFO    SenderThread:485 [sender.py:_save_file():1390] saving file wandb-summary.json with policy end
 2024-04-12 07:41:46,921 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 07:41:47,922 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 07:41:48,004 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 07:41:48,922 INFO    Thread-12 :485 [dir_watcher.py:_on_file_modified():288] file/dir modified: /kaggle/working/wandb/run-20240412_074044-widovcn3/files/config.yaml
+2024-04-12 07:41:53,106 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 07:41:58,107 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 07:42:01,035 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: stop_status
+2024-04-12 07:42:01,036 DEBUG   SenderThread:485 [sender.py:send_request():406] send_request: stop_status
+2024-04-12 07:42:01,038 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: internal_messages
+2024-04-12 07:42:03,131 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 07:42:08,132 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 07:42:09,300 DEBUG   HandlerThread:485 [handler.py:handle_request():146] handle_request: partial_history
+2024-04-12 07:42:09,301 DEBUG   SenderThread:485 [sender.py:send():379] send: history
+2024-04-12 07:42:09,301 DEBUG   SenderThread:485 [sender.py:send_request():406] send_request: summary_record
+2024-04-12 07:42:09,303 INFO    SenderThread:485 [sender.py:_save_file():1390] saving file wandb-summary.json with policy end
+2024-04-12 07:42:09,932 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