trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 4.0,
  "global_step": 108,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_bleu": 57.2256,
      "eval_gen_len": 29.225,
      "eval_loss": 0.8249181509017944,
      "eval_runtime": 3.4566,
      "eval_samples_per_second": 23.144,
      "eval_steps_per_second": 2.025,
      "step": 27
    },
    {
      "epoch": 2.0,
      "eval_bleu": 58.125,
      "eval_gen_len": 28.95,
      "eval_loss": 0.7963635325431824,
      "eval_runtime": 3.3627,
      "eval_samples_per_second": 23.79,
      "eval_steps_per_second": 2.082,
      "step": 54
    },
    {
      "epoch": 3.0,
      "eval_bleu": 59.3353,
      "eval_gen_len": 29.35,
      "eval_loss": 0.778590977191925,
      "eval_runtime": 3.3696,
      "eval_samples_per_second": 23.742,
      "eval_steps_per_second": 2.077,
      "step": 81
    },
    {
      "epoch": 4.0,
      "eval_bleu": 59.7829,
      "eval_gen_len": 29.475,
      "eval_loss": 0.7762163877487183,
      "eval_runtime": 3.3687,
      "eval_samples_per_second": 23.748,
      "eval_steps_per_second": 2.078,
      "step": 108
    },
    {
      "epoch": 4.0,
      "step": 108,
      "total_flos": 26710057746432.0,
      "train_loss": 0.678381743254485,
      "train_runtime": 23.2951,
      "train_samples_per_second": 54.776,
      "train_steps_per_second": 4.636
    }
  ],
  "max_steps": 108,
  "num_train_epochs": 4,
  "total_flos": 26710057746432.0,
  "trial_name": null,
  "trial_params": null
}