run-8/checkpoint-321/trainer_state.json

{
  "best_metric": 0.3051965941207153,
  "best_model_checkpoint": "distilbert-base-uncased-finetuned-cola/run-8/checkpoint-321",
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 321,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_loss": 0.5813318490982056,
      "eval_matthews_correlation": 0.0,
      "eval_runtime": 0.4888,
      "eval_samples_per_second": 2133.724,
      "eval_steps_per_second": 135.02,
      "step": 107
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.6579855680465698,
      "eval_matthews_correlation": 0.30246066235214725,
      "eval_runtime": 0.4813,
      "eval_samples_per_second": 2167.13,
      "eval_steps_per_second": 137.134,
      "step": 214
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.8824570775032043,
      "eval_matthews_correlation": 0.3051965941207153,
      "eval_runtime": 0.4826,
      "eval_samples_per_second": 2161.247,
      "eval_steps_per_second": 136.762,
      "step": 321
    }
  ],
  "logging_steps": 500,
  "max_steps": 321,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "total_flos": 0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": {
    "learning_rate": 3.438273396233882e-05,
    "num_train_epochs": 3,
    "per_device_train_batch_size": 8,
    "seed": 36
  }
}