trainer_state.json

{
  "best_metric": 0.40592309832572937,
  "best_model_checkpoint": "output/checkpoint-100",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 100,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5189873417721519,
      "eval_loss": 0.6217403411865234,
      "eval_runtime": 7.9638,
      "eval_samples_per_second": 9.92,
      "eval_steps_per_second": 0.628,
      "step": 20
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7468354430379747,
      "eval_loss": 0.49112987518310547,
      "eval_runtime": 5.4261,
      "eval_samples_per_second": 14.559,
      "eval_steps_per_second": 0.921,
      "step": 40
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8354430379746836,
      "eval_loss": 0.40680351853370667,
      "eval_runtime": 5.7332,
      "eval_samples_per_second": 13.779,
      "eval_steps_per_second": 0.872,
      "step": 60
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.810126582278481,
      "eval_loss": 0.42049866914749146,
      "eval_runtime": 4.7735,
      "eval_samples_per_second": 16.55,
      "eval_steps_per_second": 1.047,
      "step": 80
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.810126582278481,
      "eval_loss": 0.40592309832572937,
      "eval_runtime": 6.5454,
      "eval_samples_per_second": 12.07,
      "eval_steps_per_second": 0.764,
      "step": 100
    }
  ],
  "logging_steps": 500,
  "max_steps": 100,
  "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": 153307987384788.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}