{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.8182245185533112,
  "eval_steps": 500,
  "global_step": 2000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.7,
      "grad_norm": 2.76055908203125,
      "learning_rate": 0.00016473906911142455,
      "loss": 3.9965,
      "step": 500
    },
    {
      "epoch": 1.41,
      "grad_norm": 2.5890133380889893,
      "learning_rate": 0.00012947813822284908,
      "loss": 3.3018,
      "step": 1000
    },
    {
      "epoch": 2.11,
      "grad_norm": 2.6693315505981445,
      "learning_rate": 9.421720733427363e-05,
      "loss": 2.8988,
      "step": 1500
    },
    {
      "epoch": 2.82,
      "grad_norm": 3.1421236991882324,
      "learning_rate": 5.8956276445698163e-05,
      "loss": 2.3929,
      "step": 2000
    }
  ],
  "logging_steps": 500,
  "max_steps": 2836,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 500,
  "total_flos": 783843434496000.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}