Training in progress, step 40

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3ea77a2400c32cfb7cf6845fdbd212870169c65e267554bf3ddf8bf631948cdc
+oid sha256:8873e4fdd5279928a04e9a0a09c50f9664887f0cb51b7ea9edff60bc99b21365
 size 8077608

runs/Apr12_07-33-49_e5a48bec8248/events.out.tfevents.1712907230.e5a48bec8248.174.1

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f6636275300a1b9bf7a83951f2cceefe8a6f9b4f3df60c1573079e7b0469582b
-size 5486
+oid sha256:c42462f9ce2b5c465f011bc8dbbfaa053a2fc677d11434902c64e9be8fa49e19
+size 5693

wandb/debug-internal.log

@@ -276,3 +276,10 @@ subprocess.TimeoutExpired: Command '['conda', 'env', 'export']' timed out after
 2024-04-12 07:33:57,224 DEBUG   HandlerThread:216 [handler.py:handle_request():146] handle_request: status_report
 2024-04-12 07:33:57,225 INFO    SenderThread:216 [sender.py:_save_file():1390] saving file wandb-summary.json with policy end
 2024-04-12 07:33:58,175 INFO    Thread-12 :216 [dir_watcher.py:_on_file_modified():288] file/dir modified: /kaggle/working/wandb/run-20240412_072736-gyl3aena/files/wandb-summary.json
+2024-04-12 07:34:00,176 INFO    Thread-12 :216 [dir_watcher.py:_on_file_modified():288] file/dir modified: /kaggle/working/wandb/run-20240412_072736-gyl3aena/files/output.log
+2024-04-12 07:34:03,010 DEBUG   HandlerThread:216 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 07:34:04,112 DEBUG   HandlerThread:216 [handler.py:handle_request():146] handle_request: partial_history
+2024-04-12 07:34:04,114 DEBUG   SenderThread:216 [sender.py:send():379] send: history
+2024-04-12 07:34:04,114 DEBUG   SenderThread:216 [sender.py:send_request():406] send_request: summary_record
+2024-04-12 07:34:04,114 INFO    SenderThread:216 [sender.py:_save_file():1390] saving file wandb-summary.json with policy end
+2024-04-12 07:34:04,178 INFO    Thread-12 :216 [dir_watcher.py:_on_file_modified():288] file/dir modified: /kaggle/working/wandb/run-20240412_072736-gyl3aena/files/wandb-summary.json

wandb/run-20240412_072736-gyl3aena/files/output.log

@@ -8,4 +8,6 @@ trainable params: 2,019,328 || all params: 1,067,333,632 || trainable%: 0.189193
 /opt/conda/lib/python3.10/site-packages/accelerate/accelerator.py:432: FutureWarning: Passing the following arguments to `Accelerator` is deprecated and will be removed in version 1.0 of Accelerate: dict_keys(['dispatch_batches', 'split_batches', 'even_batches', 'use_seedable_sampler']). Please pass an `accelerate.DataLoaderConfiguration` instead:
 dataloader_config = DataLoaderConfiguration(dispatch_batches=None, split_batches=False, even_batches=True, use_seedable_sampler=True)
   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_072736-gyl3aena/files/wandb-summary.json

@@ -1 +1 @@
-{"train/loss": 4.6745, "train/grad_norm": 87.1609115600586, "train/learning_rate": 2.341780598043574e-05, "train/epoch": 0.0, "train/global_step": 20, "_timestamp": 1712907237.222082, "_runtime": 380.59000301361084, "_step": 4, "train_runtime": 39.0078, "train_samples_per_second": 3.076, "train_steps_per_second": 1.538, "total_flos": 19812939669504.0, "train_loss": 4.6811678568522135}
+{"train/loss": 4.5875, "train/grad_norm": 226.922119140625, "train/learning_rate": 7.97387338950315e-06, "train/epoch": 0.0, "train/global_step": 40, "_timestamp": 1712907244.1121442, "_runtime": 387.48006534576416, "_step": 5, "train_runtime": 39.0078, "train_samples_per_second": 3.076, "train_steps_per_second": 1.538, "total_flos": 19812939669504.0, "train_loss": 4.6811678568522135}

wandb/run-20240412_072736-gyl3aena/logs/debug-internal.log

@@ -276,3 +276,10 @@ subprocess.TimeoutExpired: Command '['conda', 'env', 'export']' timed out after
 2024-04-12 07:33:57,224 DEBUG   HandlerThread:216 [handler.py:handle_request():146] handle_request: status_report
 2024-04-12 07:33:57,225 INFO    SenderThread:216 [sender.py:_save_file():1390] saving file wandb-summary.json with policy end
 2024-04-12 07:33:58,175 INFO    Thread-12 :216 [dir_watcher.py:_on_file_modified():288] file/dir modified: /kaggle/working/wandb/run-20240412_072736-gyl3aena/files/wandb-summary.json
+2024-04-12 07:34:00,176 INFO    Thread-12 :216 [dir_watcher.py:_on_file_modified():288] file/dir modified: /kaggle/working/wandb/run-20240412_072736-gyl3aena/files/output.log
+2024-04-12 07:34:03,010 DEBUG   HandlerThread:216 [handler.py:handle_request():146] handle_request: status_report
+2024-04-12 07:34:04,112 DEBUG   HandlerThread:216 [handler.py:handle_request():146] handle_request: partial_history
+2024-04-12 07:34:04,114 DEBUG   SenderThread:216 [sender.py:send():379] send: history
+2024-04-12 07:34:04,114 DEBUG   SenderThread:216 [sender.py:send_request():406] send_request: summary_record
+2024-04-12 07:34:04,114 INFO    SenderThread:216 [sender.py:_save_file():1390] saving file wandb-summary.json with policy end
+2024-04-12 07:34:04,178 INFO    Thread-12 :216 [dir_watcher.py:_on_file_modified():288] file/dir modified: /kaggle/working/wandb/run-20240412_072736-gyl3aena/files/wandb-summary.json