run-3/checkpoint-81/trainer_state.json

{
  "best_metric": 0.2497095186939662,
  "best_model_checkpoint": "distilbert-base-uncased-finetuned-cola/run-3/checkpoint-81",
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 81,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_loss": 0.6095482110977173,
      "eval_matthews_correlation": 0.0,
      "eval_runtime": 0.48,
      "eval_samples_per_second": 2172.844,
      "eval_steps_per_second": 137.495,
      "step": 27
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.5869027376174927,
      "eval_matthews_correlation": 0.058936921430734465,
      "eval_runtime": 0.4832,
      "eval_samples_per_second": 2158.591,
      "eval_steps_per_second": 136.593,
      "step": 54
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.6861404180526733,
      "eval_matthews_correlation": 0.2497095186939662,
      "eval_runtime": 0.4822,
      "eval_samples_per_second": 2162.868,
      "eval_steps_per_second": 136.864,
      "step": 81
    }
  ],
  "logging_steps": 500,
  "max_steps": 108,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 500,
  "total_flos": 0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "learning_rate": 4.3023437109267575e-05,
    "num_train_epochs": 4,
    "per_device_train_batch_size": 32,
    "seed": 33
  }
}