trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "global_step": 366,
  "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": 1.3121,
      "step": 122
    },
    {
      "epoch": 1.0,
      "eval_exact_match": 69.45447045477982,
      "eval_f1": 72.38400325486641,
      "eval_runtime": 3.9219,
      "eval_samples_per_second": 2472.511,
      "eval_steps_per_second": 7.904,
      "step": 122
    },
    {
      "epoch": 2.0,
      "learning_rate": 3e-05,
      "loss": 0.5522,
      "step": 244
    },
    {
      "epoch": 2.0,
      "eval_exact_match": 69.85665669794781,
      "eval_f1": 72.8585664154011,
      "eval_runtime": 3.5475,
      "eval_samples_per_second": 2733.459,
      "eval_steps_per_second": 8.738,
      "step": 244
    },
    {
      "epoch": 3.0,
      "learning_rate": 3e-05,
      "loss": 0.4868,
      "step": 366
    },
    {
      "epoch": 3.0,
      "eval_exact_match": 70.46509229658658,
      "eval_f1": 73.408366600223,
      "eval_runtime": 3.5301,
      "eval_samples_per_second": 2746.956,
      "eval_steps_per_second": 8.782,
      "step": 366
    }
  ],
  "max_steps": 1220,
  "num_train_epochs": 10,
  "total_flos": 43819446829056.0,
  "trial_name": null,
  "trial_params": null
}