{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 6.0,
  "eval_steps": 500,
  "global_step": 72,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.83,
      "learning_rate": 8.611111111111112e-05,
      "loss": 1.7203,
      "step": 10
    },
    {
      "epoch": 1.0,
      "eval_loss": 1.4726248979568481,
      "eval_runtime": 18.228,
      "eval_samples_per_second": 10.369,
      "eval_steps_per_second": 1.317,
      "step": 12
    },
    {
      "epoch": 1.67,
      "learning_rate": 7.222222222222222e-05,
      "loss": 1.4259,
      "step": 20
    },
    {
      "epoch": 2.0,
      "eval_loss": 1.317138910293579,
      "eval_runtime": 18.3382,
      "eval_samples_per_second": 10.306,
      "eval_steps_per_second": 1.309,
      "step": 24
    },
    {
      "epoch": 2.5,
      "learning_rate": 5.833333333333334e-05,
      "loss": 1.32,
      "step": 30
    },
    {
      "epoch": 3.0,
      "eval_loss": 1.2585612535476685,
      "eval_runtime": 18.3327,
      "eval_samples_per_second": 10.309,
      "eval_steps_per_second": 1.309,
      "step": 36
    },
    {
      "epoch": 3.33,
      "learning_rate": 4.4444444444444447e-05,
      "loss": 1.2774,
      "step": 40
    },
    {
      "epoch": 4.0,
      "eval_loss": 1.227800726890564,
      "eval_runtime": 18.3332,
      "eval_samples_per_second": 10.309,
      "eval_steps_per_second": 1.309,
      "step": 48
    },
    {
      "epoch": 4.17,
      "learning_rate": 3.055555555555556e-05,
      "loss": 1.2255,
      "step": 50
    },
    {
      "epoch": 5.0,
      "learning_rate": 1.6666666666666667e-05,
      "loss": 1.2197,
      "step": 60
    },
    {
      "epoch": 5.0,
      "eval_loss": 1.2121710777282715,
      "eval_runtime": 18.3433,
      "eval_samples_per_second": 10.303,
      "eval_steps_per_second": 1.308,
      "step": 60
    },
    {
      "epoch": 5.83,
      "learning_rate": 2.777777777777778e-06,
      "loss": 1.2053,
      "step": 70
    },
    {
      "epoch": 6.0,
      "eval_loss": 1.2067466974258423,
      "eval_runtime": 18.3328,
      "eval_samples_per_second": 10.309,
      "eval_steps_per_second": 1.309,
      "step": 72
    }
  ],
  "logging_steps": 10,
  "max_steps": 72,
  "num_train_epochs": 6,
  "save_steps": 500,
  "total_flos": 3.683597393092608e+16,
  "trial_name": null,
  "trial_params": null
}