{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 10.0,
  "global_step": 3820,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "learning_rate": 0.0002,
      "loss": 1.9462,
      "step": 382
    },
    {
      "epoch": 1.0,
      "eval_gen_len": 15.453,
      "eval_loss": 0.2963322103023529,
      "eval_per": 0.3495,
      "eval_runtime": 122.3144,
      "eval_samples_per_second": 56.976,
      "eval_steps_per_second": 1.782,
      "step": 382
    },
    {
      "epoch": 2.0,
      "learning_rate": 0.00017777777777777779,
      "loss": 0.2979,
      "step": 764
    },
    {
      "epoch": 2.0,
      "eval_gen_len": 15.6759,
      "eval_loss": 0.21439459919929504,
      "eval_per": 0.2941,
      "eval_runtime": 121.944,
      "eval_samples_per_second": 57.149,
      "eval_steps_per_second": 1.788,
      "step": 764
    },
    {
      "epoch": 3.0,
      "learning_rate": 0.00015555555555555556,
      "loss": 0.2319,
      "step": 1146
    },
    {
      "epoch": 3.0,
      "eval_gen_len": 15.7219,
      "eval_loss": 0.18984708189964294,
      "eval_per": 0.2765,
      "eval_runtime": 120.5224,
      "eval_samples_per_second": 57.823,
      "eval_steps_per_second": 1.809,
      "step": 1146
    },
    {
      "epoch": 4.0,
      "learning_rate": 0.00013333333333333334,
      "loss": 0.2042,
      "step": 1528
    },
    {
      "epoch": 4.0,
      "eval_gen_len": 15.7235,
      "eval_loss": 0.1799211949110031,
      "eval_per": 0.2719,
      "eval_runtime": 119.8528,
      "eval_samples_per_second": 58.146,
      "eval_steps_per_second": 1.819,
      "step": 1528
    },
    {
      "epoch": 5.0,
      "learning_rate": 0.00011111111111111112,
      "loss": 0.1879,
      "step": 1910
    },
    {
      "epoch": 5.0,
      "eval_gen_len": 15.7054,
      "eval_loss": 0.1743946522474289,
      "eval_per": 0.2657,
      "eval_runtime": 120.4255,
      "eval_samples_per_second": 57.87,
      "eval_steps_per_second": 1.81,
      "step": 1910
    },
    {
      "epoch": 6.0,
      "learning_rate": 8.888888888888889e-05,
      "loss": 0.1763,
      "step": 2292
    },
    {
      "epoch": 6.0,
      "eval_gen_len": 15.7406,
      "eval_loss": 0.1728605031967163,
      "eval_per": 0.2645,
      "eval_runtime": 119.6482,
      "eval_samples_per_second": 58.246,
      "eval_steps_per_second": 1.822,
      "step": 2292
    },
    {
      "epoch": 7.0,
      "learning_rate": 6.666666666666667e-05,
      "loss": 0.1677,
      "step": 2674
    },
    {
      "epoch": 7.0,
      "eval_gen_len": 15.7269,
      "eval_loss": 0.1693088412284851,
      "eval_per": 0.2619,
      "eval_runtime": 119.4883,
      "eval_samples_per_second": 58.324,
      "eval_steps_per_second": 1.824,
      "step": 2674
    },
    {
      "epoch": 8.0,
      "learning_rate": 4.4444444444444447e-05,
      "loss": 0.1613,
      "step": 3056
    },
    {
      "epoch": 8.0,
      "eval_gen_len": 15.7371,
      "eval_loss": 0.1667601615190506,
      "eval_per": 0.2604,
      "eval_runtime": 118.43,
      "eval_samples_per_second": 58.845,
      "eval_steps_per_second": 1.841,
      "step": 3056
    },
    {
      "epoch": 9.0,
      "learning_rate": 2.2222222222222223e-05,
      "loss": 0.1559,
      "step": 3438
    },
    {
      "epoch": 9.0,
      "eval_gen_len": 15.7383,
      "eval_loss": 0.16669157147407532,
      "eval_per": 0.2585,
      "eval_runtime": 118.8757,
      "eval_samples_per_second": 58.624,
      "eval_steps_per_second": 1.834,
      "step": 3438
    },
    {
      "epoch": 10.0,
      "learning_rate": 0.0,
      "loss": 0.1534,
      "step": 3820
    },
    {
      "epoch": 10.0,
      "eval_gen_len": 15.7318,
      "eval_loss": 0.16682308912277222,
      "eval_per": 0.2588,
      "eval_runtime": 118.3076,
      "eval_samples_per_second": 58.906,
      "eval_steps_per_second": 1.843,
      "step": 3820
    },
    {
      "epoch": 10.0,
      "step": 3820,
      "total_flos": 5.601731015983104e+16,
      "train_loss": 0.36827531884478026,
      "train_runtime": 2516.6862,
      "train_samples_per_second": 193.814,
      "train_steps_per_second": 1.518
    }
  ],
  "max_steps": 3820,
  "num_train_epochs": 10,
  "total_flos": 5.601731015983104e+16,
  "trial_name": null,
  "trial_params": null
}