run-3/checkpoint-285/trainer_state.json

{
  "best_metric": 0.4066666666666667,
  "best_model_checkpoint": "distilbert-base-uncased-finetuned-multiclass-classification/run-3/checkpoint-285",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 285,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.18,
      "eval_loss": 4.173440933227539,
      "eval_runtime": 9.5961,
      "eval_samples_per_second": 93.788,
      "eval_steps_per_second": 23.447,
      "step": 57
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.23777777777777778,
      "eval_loss": 3.8830435276031494,
      "eval_runtime": 9.2156,
      "eval_samples_per_second": 97.66,
      "eval_steps_per_second": 24.415,
      "step": 114
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.30777777777777776,
      "eval_loss": 3.662949323654175,
      "eval_runtime": 9.2811,
      "eval_samples_per_second": 96.971,
      "eval_steps_per_second": 24.243,
      "step": 171
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.37,
      "eval_loss": 3.5301730632781982,
      "eval_runtime": 10.4078,
      "eval_samples_per_second": 86.474,
      "eval_steps_per_second": 21.618,
      "step": 228
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.4066666666666667,
      "eval_loss": 3.486360788345337,
      "eval_runtime": 11.7461,
      "eval_samples_per_second": 76.621,
      "eval_steps_per_second": 19.155,
      "step": 285
    }
  ],
  "logging_steps": 500,
  "max_steps": 285,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 0,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": {
    "learning_rate": 4.935170904021075e-06,
    "num_train_epochs": 5,
    "per_device_train_batch_size": 64,
    "seed": 37
  }
}