trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 4.0,
  "global_step": 488,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "learning_rate": 3e-05,
      "loss": 1.0125,
      "step": 122
    },
    {
      "epoch": 1.0,
      "eval_exact_match": 68.61916056512324,
      "eval_f1": 72.65611452136685,
      "eval_runtime": 9.4452,
      "eval_samples_per_second": 1026.659,
      "eval_steps_per_second": 3.282,
      "step": 122
    },
    {
      "epoch": 2.0,
      "learning_rate": 3e-05,
      "loss": 0.5078,
      "step": 244
    },
    {
      "epoch": 2.0,
      "eval_exact_match": 70.45477982881303,
      "eval_f1": 73.58411758552184,
      "eval_runtime": 8.6455,
      "eval_samples_per_second": 1121.628,
      "eval_steps_per_second": 3.586,
      "step": 244
    },
    {
      "epoch": 3.0,
      "learning_rate": 3e-05,
      "loss": 0.4477,
      "step": 366
    },
    {
      "epoch": 3.0,
      "eval_exact_match": 70.56821697432196,
      "eval_f1": 73.54649137829497,
      "eval_runtime": 8.613,
      "eval_samples_per_second": 1125.861,
      "eval_steps_per_second": 3.599,
      "step": 366
    },
    {
      "epoch": 4.0,
      "learning_rate": 3e-05,
      "loss": 0.4095,
      "step": 488
    },
    {
      "epoch": 4.0,
      "eval_exact_match": 71.0941528307724,
      "eval_f1": 73.77666104591952,
      "eval_runtime": 8.627,
      "eval_samples_per_second": 1124.033,
      "eval_steps_per_second": 3.593,
      "step": 488
    }
  ],
  "max_steps": 1220,
  "num_train_epochs": 10,
  "total_flos": 153959423541248.0,
  "trial_name": null,
  "trial_params": null
}