{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.6515506906437321,
  "global_step": 20000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.16,
      "learning_rate": 2.8373729476153247e-05,
      "loss": 1.6126,
      "step": 5000
    },
    {
      "epoch": 0.16,
      "eval_exact_match": 68.15336463223787,
      "eval_f1": 73.98256617857639,
      "eval_loss": 1.189887523651123,
      "eval_runtime": 1542.7266,
      "eval_samples_per_second": 7.463,
      "eval_steps_per_second": 3.732,
      "step": 5000
    },
    {
      "epoch": 0.33,
      "learning_rate": 2.674550430023456e-05,
      "loss": 1.1965,
      "step": 10000
    },
    {
      "epoch": 0.33,
      "eval_exact_match": 75.13475917231786,
      "eval_f1": 79.95184555682803,
      "eval_loss": 0.9056070446968079,
      "eval_runtime": 1533.9546,
      "eval_samples_per_second": 7.505,
      "eval_steps_per_second": 3.753,
      "step": 10000
    },
    {
      "epoch": 0.49,
      "learning_rate": 2.511695334897055e-05,
      "loss": 1.0769,
      "step": 15000
    },
    {
      "epoch": 0.49,
      "eval_exact_match": 77.56042427403929,
      "eval_f1": 82.22503770504048,
      "eval_loss": 0.9495312571525574,
      "eval_runtime": 1533.818,
      "eval_samples_per_second": 7.506,
      "eval_steps_per_second": 3.753,
      "step": 15000
    },
    {
      "epoch": 0.65,
      "learning_rate": 2.3488728173051864e-05,
      "loss": 0.9992,
      "step": 20000
    },
    {
      "epoch": 0.65,
      "eval_exact_match": 81.09024517475221,
      "eval_f1": 85.49992934663945,
      "eval_loss": 0.8323877453804016,
      "eval_runtime": 1541.3611,
      "eval_samples_per_second": 7.469,
      "eval_steps_per_second": 3.735,
      "step": 20000
    }
  ],
  "max_steps": 92088,
  "num_train_epochs": 3,
  "total_flos": 4.412940582912e+16,
  "trial_name": null,
  "trial_params": null
}