Writing logs to ./outputs/2024-02-28-00-04-11-804961/train_log.txt.
Wrote original training args to ./outputs/2024-02-28-00-04-11-804961/training_args.json.
***** Running training *****
  Num examples = 238766
  Num epochs = 3
  Num clean epochs = 3
  Instantaneous batch size per device = 8
  Total train batch size (w. parallel, distributed & accumulation) = 8
  Gradient accumulation steps = 1
  Total optimization steps = 89538
==========================================================
Epoch 1
Running clean epoch 1/3
Train accuracy: 90.13%
Eval accuracy: 83.97%
Best score found. Saved model to ./outputs/2024-02-28-00-04-11-804961/best_model/
==========================================================
Epoch 2
Running clean epoch 2/3
Train accuracy: 93.19%
Eval accuracy: 87.71%
Best score found. Saved model to ./outputs/2024-02-28-00-04-11-804961/best_model/
==========================================================
Epoch 3
Running clean epoch 3/3
Train accuracy: 95.65%
Eval accuracy: 88.57%
Best score found. Saved model to ./outputs/2024-02-28-00-04-11-804961/best_model/
Wrote README to ./outputs/2024-02-28-00-04-11-804961/README.md.