{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.88,
  "eval_steps": 500,
  "global_step": 18,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.16,
      "grad_norm": 10.593756006766313,
      "learning_rate": 5e-07,
      "loss": 1.7036,
      "step": 1
    },
    {
      "epoch": 0.8,
      "grad_norm": 7.23312775238484,
      "learning_rate": 9.157348061512726e-07,
      "loss": 1.7014,
      "step": 5
    },
    {
      "epoch": 0.96,
      "eval_loss": 1.9007213115692139,
      "eval_runtime": 5.4737,
      "eval_samples_per_second": 10.961,
      "eval_steps_per_second": 0.731,
      "step": 6
    },
    {
      "epoch": 1.6,
      "grad_norm": 5.80431534123651,
      "learning_rate": 5e-07,
      "loss": 1.6285,
      "step": 10
    },
    {
      "epoch": 1.92,
      "eval_loss": 1.8675754070281982,
      "eval_runtime": 5.3012,
      "eval_samples_per_second": 11.318,
      "eval_steps_per_second": 0.755,
      "step": 12
    },
    {
      "epoch": 2.4,
      "grad_norm": 4.830213221606365,
      "learning_rate": 8.426519384872732e-08,
      "loss": 1.5573,
      "step": 15
    },
    {
      "epoch": 2.88,
      "eval_loss": 1.859323263168335,
      "eval_runtime": 5.2878,
      "eval_samples_per_second": 11.347,
      "eval_steps_per_second": 0.756,
      "step": 18
    },
    {
      "epoch": 2.88,
      "step": 18,
      "total_flos": 3716488888320.0,
      "train_loss": 1.6118707987997267,
      "train_runtime": 322.7432,
      "train_samples_per_second": 3.625,
      "train_steps_per_second": 0.056
    }
  ],
  "logging_steps": 5,
  "max_steps": 18,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "total_flos": 3716488888320.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}