run-1/checkpoint-536/trainer_state.json

{
  "best_metric": 0.44251427534495513,
  "best_model_checkpoint": "distilbert-base-uncased-finetuned-cola/run-1/checkpoint-536",
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 536,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_loss": 0.5035493969917297,
      "eval_matthews_correlation": 0.4141806938515634,
      "eval_runtime": 0.7419,
      "eval_samples_per_second": 1405.832,
      "eval_steps_per_second": 88.96,
      "step": 268
    },
    {
      "epoch": 1.87,
      "grad_norm": 4.896795749664307,
      "learning_rate": 5.453884401804003e-06,
      "loss": 0.4997,
      "step": 500
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.4967799484729767,
      "eval_matthews_correlation": 0.44251427534495513,
      "eval_runtime": 0.8433,
      "eval_samples_per_second": 1236.743,
      "eval_steps_per_second": 78.26,
      "step": 536
    }
  ],
  "logging_steps": 500,
  "max_steps": 1072,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 500,
  "total_flos": 96116481642000.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "learning_rate": 1.0221265871912396e-05,
    "num_train_epochs": 4,
    "per_device_train_batch_size": 32,
    "seed": 8
  }
}