{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 18750,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "learning_rate": 4.999733333333334e-05,
      "loss": 6.0454,
      "step": 1
    },
    {
      "epoch": 1.0,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 3.0043,
      "step": 6250
    },
    {
      "epoch": 1.0,
      "eval_HasAns_exact": 8.816473410635746,
      "eval_HasAns_f1": 10.888242983453603,
      "eval_HasAns_total": 5002,
      "eval_NoAns_exact": 87.01480592236895,
      "eval_NoAns_f1": 87.01480592236895,
      "eval_NoAns_total": 4998,
      "eval_best_exact": 49.98,
      "eval_best_exact_thresh": 0.0,
      "eval_best_f1": 49.990334360084354,
      "eval_best_f1_thresh": 0.0,
      "eval_exact": 47.9,
      "eval_f1": 48.93629914032349,
      "eval_runtime": 188.2959,
      "eval_samples_per_second": 53.108,
      "eval_steps_per_second": 6.638,
      "eval_total": 10000,
      "step": 6250
    },
    {
      "epoch": 2.0,
      "learning_rate": 1.6666666666666667e-05,
      "loss": 2.4513,
      "step": 12500
    },
    {
      "epoch": 2.0,
      "eval_HasAns_exact": 15.25389844062375,
      "eval_HasAns_f1": 18.855802501363748,
      "eval_HasAns_total": 5002,
      "eval_NoAns_exact": 77.47098839535815,
      "eval_NoAns_f1": 77.47098839535815,
      "eval_NoAns_total": 4998,
      "eval_best_exact": 50.0,
      "eval_best_exact_thresh": 0.0,
      "eval_best_f1": 50.0,
      "eval_best_f1_thresh": 0.0,
      "eval_exact": 46.35,
      "eval_f1": 48.15167241118216,
      "eval_runtime": 184.7185,
      "eval_samples_per_second": 54.136,
      "eval_steps_per_second": 6.767,
      "eval_total": 10000,
      "step": 12500
    },
    {
      "epoch": 3.0,
      "learning_rate": 0.0,
      "loss": 2.2686,
      "step": 18750
    },
    {
      "epoch": 3.0,
      "eval_HasAns_exact": 20.33186725309876,
      "eval_HasAns_f1": 25.51719340946248,
      "eval_HasAns_total": 5002,
      "eval_NoAns_exact": 69.22769107643057,
      "eval_NoAns_f1": 69.22769107643057,
      "eval_NoAns_total": 4998,
      "eval_best_exact": 50.0,
      "eval_best_exact_thresh": 0.0,
      "eval_best_f1": 50.0,
      "eval_best_f1_thresh": 0.0,
      "eval_exact": 44.77,
      "eval_f1": 47.363700143413155,
      "eval_runtime": 187.3148,
      "eval_samples_per_second": 53.386,
      "eval_steps_per_second": 6.673,
      "eval_total": 10000,
      "step": 18750
    },
    {
      "epoch": 3.0,
      "step": 18750,
      "total_flos": 137223244800000.0,
      "train_loss": 2.574889142405192,
      "train_runtime": 9834.5137,
      "train_samples_per_second": 15.252,
      "train_steps_per_second": 1.907
    }
  ],
  "logging_steps": 500,
  "max_steps": 18750,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 200,
  "total_flos": 137223244800000.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}