trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 4.0,
  "global_step": 1096,
  "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": 1.179,
      "step": 274
    },
    {
      "epoch": 1.0,
      "eval_exact_match": 73.3006878862073,
      "eval_f1": 76.23985813650026,
      "eval_runtime": 8.4625,
      "eval_samples_per_second": 1013.533,
      "eval_steps_per_second": 3.191,
      "step": 274
    },
    {
      "epoch": 2.0,
      "learning_rate": 3e-05,
      "loss": 0.6306,
      "step": 548
    },
    {
      "epoch": 2.0,
      "eval_exact_match": 73.49889238661537,
      "eval_f1": 76.3130007817605,
      "eval_runtime": 7.5902,
      "eval_samples_per_second": 1130.01,
      "eval_steps_per_second": 3.557,
      "step": 548
    },
    {
      "epoch": 3.0,
      "learning_rate": 3e-05,
      "loss": 0.4746,
      "step": 822
    },
    {
      "epoch": 3.0,
      "eval_exact_match": 72.0648245307217,
      "eval_f1": 76.0894037408996,
      "eval_runtime": 7.7748,
      "eval_samples_per_second": 1103.185,
      "eval_steps_per_second": 3.473,
      "step": 822
    },
    {
      "epoch": 4.0,
      "learning_rate": 3e-05,
      "loss": 0.3721,
      "step": 1096
    },
    {
      "epoch": 4.0,
      "eval_exact_match": 72.91593797365046,
      "eval_f1": 76.32239576178218,
      "eval_runtime": 7.5706,
      "eval_samples_per_second": 1132.933,
      "eval_steps_per_second": 3.566,
      "step": 1096
    }
  ],
  "max_steps": 2740,
  "num_train_epochs": 10,
  "total_flos": 348317284302848.0,
  "trial_name": null,
  "trial_params": null
}