run-2/checkpoint-428/trainer_state.json

{
  "best_metric": 0.663247201911855,
  "best_model_checkpoint": "deberta-v3-large-finetuned-cola-midterm/run-2/checkpoint-321",
  "epoch": 4.0,
  "eval_steps": 500,
  "global_step": 428,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_loss": 0.470249742269516,
      "eval_matthews_correlation": 0.37472354614099185,
      "eval_runtime": 8.4859,
      "eval_samples_per_second": 122.91,
      "eval_steps_per_second": 7.778,
      "step": 107
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.37880179286003113,
      "eval_matthews_correlation": 0.6454833331679224,
      "eval_runtime": 8.7227,
      "eval_samples_per_second": 119.574,
      "eval_steps_per_second": 7.566,
      "step": 214
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.4730988144874573,
      "eval_matthews_correlation": 0.663247201911855,
      "eval_runtime": 8.9367,
      "eval_samples_per_second": 116.71,
      "eval_steps_per_second": 7.385,
      "step": 321
    },
    {
      "epoch": 4.0,
      "eval_loss": 0.5757762789726257,
      "eval_matthews_correlation": 0.6431581256935667,
      "eval_runtime": 8.8924,
      "eval_samples_per_second": 117.291,
      "eval_steps_per_second": 7.422,
      "step": 428
    }
  ],
  "logging_steps": 500,
  "max_steps": 428,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 500,
  "total_flos": 0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": {
    "learning_rate": 5.723917767775046e-06,
    "num_train_epochs": 4,
    "per_device_train_batch_size": 8,
    "seed": 4
  }
}