Training in progress, step 60

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0b2c5bc1a76f4fbdd072da55811ba8b559e7f35d8cd10d44b30780d70158a606
+oid sha256:cdf108f78e81ea41efe0d9e72baa776fbb4ece073643445854f1a3f5ad51a122
 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 sha256:6b3c6c5927141a58e694b25c804dfd428f23852decfed3cb639bb97399b4f0dc
-size 5696
+oid sha256:e078a64a21b5147a8b2b46c6cf552c924986960c683df0512d935ad3a902e668
+size 5903

wandb/debug-internal.log

@@ -93,3 +93,19 @@ subprocess.TimeoutExpired: Command '['conda', 'env', 'export']' timed out after
 2024-04-12 12:00:47,247 DEBUG   SenderThread:151 [sender.py:send_request():406] send_request: summary_record
 2024-04-12 12:00:47,249 INFO    SenderThread:151 [sender.py:_save_file():1390] saving file wandb-summary.json with policy end
 2024-04-12 12:00:47,278 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 12:00:49,969 DEBUG   HandlerThread:151 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 12:00:50,279 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 12:00:52,372 DEBUG   SystemMonitor:151 [system_monitor.py:_start():172] Starting system metrics aggregation loop
+2024-04-12 12:00:52,373 DEBUG   SenderThread:151 [sender.py:send():379] send: stats
+2024-04-12 12:00:53,393 DEBUG   HandlerThread:151 [handler.py:handle_request():146] handle_request: stop_status
+2024-04-12 12:00:53,394 DEBUG   SenderThread:151 [sender.py:send_request():406] send_request: stop_status
+2024-04-12 12:00:53,397 DEBUG   HandlerThread:151 [handler.py:handle_request():146] handle_request: internal_messages
+2024-04-12 12:00:55,467 DEBUG   HandlerThread:151 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 12:00:56,281 INFO    Thread-12 :151 [dir_watcher.py:_on_file_modified():288] file/dir modified: /kaggle/working/wandb/run-20240412_115951-o0z5y86z/files/config.yaml
+2024-04-12 12:01:00,558 DEBUG   HandlerThread:151 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 12:01:05,385 DEBUG   HandlerThread:151 [handler.py:handle_request():146] handle_request: partial_history
+2024-04-12 12:01:05,387 DEBUG   SenderThread:151 [sender.py:send():379] send: history
+2024-04-12 12:01:05,387 DEBUG   SenderThread:151 [sender.py:send_request():406] send_request: summary_record
+2024-04-12 12:01:05,387 INFO    SenderThread:151 [sender.py:_save_file():1390] saving file wandb-summary.json with policy end
+2024-04-12 12:01:06,097 DEBUG   HandlerThread:151 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 12:01:06,286 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/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_115951-o0z5y86z/files/output.log

@@ -1,3 +1,5 @@
+/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": 4.6693, "train/grad_norm": 51.19196319580078, "train/learning_rate": 3.3333333333333333e-06, "train/epoch": 0.0, "train/global_step": 40, "_timestamp": 1712923247.2448928, "_runtime": 55.65392279624939, "_step": 1}
+{"train/loss": 4.6065, "train/grad_norm": 45.4248161315918, "train/learning_rate": 5e-06, "train/epoch": 0.0, "train/global_step": 60, "_timestamp": 1712923265.3853402, "_runtime": 73.79437017440796, "_step": 2}

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

@@ -93,3 +93,19 @@ subprocess.TimeoutExpired: Command '['conda', 'env', 'export']' timed out after
 2024-04-12 12:00:47,247 DEBUG   SenderThread:151 [sender.py:send_request():406] send_request: summary_record
 2024-04-12 12:00:47,249 INFO    SenderThread:151 [sender.py:_save_file():1390] saving file wandb-summary.json with policy end
 2024-04-12 12:00:47,278 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 12:00:49,969 DEBUG   HandlerThread:151 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 12:00:50,279 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 12:00:52,372 DEBUG   SystemMonitor:151 [system_monitor.py:_start():172] Starting system metrics aggregation loop
+2024-04-12 12:00:52,373 DEBUG   SenderThread:151 [sender.py:send():379] send: stats
+2024-04-12 12:00:53,393 DEBUG   HandlerThread:151 [handler.py:handle_request():146] handle_request: stop_status
+2024-04-12 12:00:53,394 DEBUG   SenderThread:151 [sender.py:send_request():406] send_request: stop_status
+2024-04-12 12:00:53,397 DEBUG   HandlerThread:151 [handler.py:handle_request():146] handle_request: internal_messages
+2024-04-12 12:00:55,467 DEBUG   HandlerThread:151 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 12:00:56,281 INFO    Thread-12 :151 [dir_watcher.py:_on_file_modified():288] file/dir modified: /kaggle/working/wandb/run-20240412_115951-o0z5y86z/files/config.yaml
+2024-04-12 12:01:00,558 DEBUG   HandlerThread:151 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 12:01:05,385 DEBUG   HandlerThread:151 [handler.py:handle_request():146] handle_request: partial_history
+2024-04-12 12:01:05,387 DEBUG   SenderThread:151 [sender.py:send():379] send: history
+2024-04-12 12:01:05,387 DEBUG   SenderThread:151 [sender.py:send_request():406] send_request: summary_record
+2024-04-12 12:01:05,387 INFO    SenderThread:151 [sender.py:_save_file():1390] saving file wandb-summary.json with policy end
+2024-04-12 12:01:06,097 DEBUG   HandlerThread:151 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 12:01:06,286 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