trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.8974358974358974,
  "global_step": 76,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.97,
      "eval_accuracy": 0.6462093862815884,
      "eval_loss": 0.6622310280799866,
      "eval_runtime": 0.8176,
      "eval_samples_per_second": 338.782,
      "eval_steps_per_second": 11.007,
      "step": 19
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6389891696750902,
      "eval_loss": 0.6256995797157288,
      "eval_runtime": 0.8184,
      "eval_samples_per_second": 338.47,
      "eval_steps_per_second": 10.997,
      "step": 39
    },
    {
      "epoch": 2.97,
      "eval_accuracy": 0.6353790613718412,
      "eval_loss": 0.7572575807571411,
      "eval_runtime": 0.8328,
      "eval_samples_per_second": 332.625,
      "eval_steps_per_second": 10.807,
      "step": 58
    },
    {
      "epoch": 3.9,
      "eval_accuracy": 0.6353790613718412,
      "eval_loss": 0.764682948589325,
      "eval_runtime": 0.817,
      "eval_samples_per_second": 339.032,
      "eval_steps_per_second": 11.015,
      "step": 76
    },
    {
      "epoch": 3.9,
      "step": 76,
      "total_flos": 2537191448248320.0,
      "train_loss": 0.5004565590306332,
      "train_runtime": 76.0596,
      "train_samples_per_second": 130.95,
      "train_steps_per_second": 0.999
    }
  ],
  "max_steps": 76,
  "num_train_epochs": 4,
  "total_flos": 2537191448248320.0,
  "trial_name": null,
  "trial_params": null
}