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": 1.165,
      "step": 122
    },
    {
      "epoch": 1.0,
      "eval_exact_match": 68.60884809734969,
      "eval_f1": 71.57514961558589,
      "eval_runtime": 12.6849,
      "eval_samples_per_second": 764.454,
      "eval_steps_per_second": 2.444,
      "step": 122
    },
    {
      "epoch": 2.0,
      "learning_rate": 3e-05,
      "loss": 0.5445,
      "step": 244
    },
    {
      "epoch": 2.0,
      "eval_exact_match": 69.99071877900381,
      "eval_f1": 73.06771046827892,
      "eval_runtime": 8.5294,
      "eval_samples_per_second": 1136.896,
      "eval_steps_per_second": 3.635,
      "step": 244
    },
    {
      "epoch": 3.0,
      "learning_rate": 3e-05,
      "loss": 0.4736,
      "step": 366
    },
    {
      "epoch": 3.0,
      "eval_exact_match": 71.20758997628133,
      "eval_f1": 73.91150191975164,
      "eval_runtime": 8.5195,
      "eval_samples_per_second": 1138.209,
      "eval_steps_per_second": 3.639,
      "step": 366
    }
  ],
  "max_steps": 1220,
  "num_train_epochs": 10,
  "total_flos": 115390042603520.0,
  "trial_name": null,
  "trial_params": null
}