run-3/checkpoint-162/trainer_state.json

{
  "best_metric": 0.346176833033057,
  "best_model_checkpoint": "distilbert-base-uncased-finetuned-cola/run-3/checkpoint-162",
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 162,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_loss": 0.5624595880508423,
      "eval_matthews_correlation": 0.10385466804655791,
      "eval_runtime": 1.2797,
      "eval_samples_per_second": 815.034,
      "eval_steps_per_second": 102.368,
      "step": 54
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.6093302965164185,
      "eval_matthews_correlation": 0.3349942831439903,
      "eval_runtime": 1.2473,
      "eval_samples_per_second": 836.218,
      "eval_steps_per_second": 105.028,
      "step": 108
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.7864214777946472,
      "eval_matthews_correlation": 0.346176833033057,
      "eval_runtime": 1.4034,
      "eval_samples_per_second": 743.189,
      "eval_steps_per_second": 93.344,
      "step": 162
    }
  ],
  "logging_steps": 500,
  "max_steps": 162,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "total_flos": 0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": {
    "learning_rate": 5.503532571253603e-05,
    "num_train_epochs": 3,
    "per_device_train_batch_size": 16,
    "seed": 28
  }
}