{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 945,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 0.84375,
      "learning_rate": 1.8805243942587e-05,
      "loss": 1.4836,
      "step": 189
    },
    {
      "epoch": 1.0,
      "eval_loss": 1.248240351676941,
      "eval_runtime": 14.7272,
      "eval_samples_per_second": 5.364,
      "eval_steps_per_second": 0.679,
      "step": 189
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.98046875,
      "learning_rate": 1.4032141389483648e-05,
      "loss": 1.2209,
      "step": 378
    },
    {
      "epoch": 2.0,
      "eval_loss": 1.2223888635635376,
      "eval_runtime": 14.6831,
      "eval_samples_per_second": 5.38,
      "eval_steps_per_second": 0.681,
      "step": 378
    },
    {
      "epoch": 3.0,
      "grad_norm": 1.0703125,
      "learning_rate": 7.555868193598188e-06,
      "loss": 1.1941,
      "step": 567
    },
    {
      "epoch": 3.0,
      "eval_loss": 1.2139047384262085,
      "eval_runtime": 14.7087,
      "eval_samples_per_second": 5.371,
      "eval_steps_per_second": 0.68,
      "step": 567
    },
    {
      "epoch": 4.0,
      "grad_norm": 1.0078125,
      "learning_rate": 2.111992708420646e-06,
      "loss": 1.1819,
      "step": 756
    },
    {
      "epoch": 4.0,
      "eval_loss": 1.2119375467300415,
      "eval_runtime": 14.6929,
      "eval_samples_per_second": 5.377,
      "eval_steps_per_second": 0.681,
      "step": 756
    },
    {
      "epoch": 5.0,
      "grad_norm": 1.125,
      "learning_rate": 0.0,
      "loss": 1.1787,
      "step": 945
    },
    {
      "epoch": 5.0,
      "eval_loss": 1.2118842601776123,
      "eval_runtime": 14.7081,
      "eval_samples_per_second": 5.371,
      "eval_steps_per_second": 0.68,
      "step": 945
    },
    {
      "epoch": 5.0,
      "step": 945,
      "total_flos": 6.657716927594496e+17,
      "train_loss": 1.2518653021918402,
      "train_runtime": 4827.6756,
      "train_samples_per_second": 1.564,
      "train_steps_per_second": 0.196
    }
  ],
  "logging_steps": 500,
  "max_steps": 945,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "total_flos": 6.657716927594496e+17,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}