{
  "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": 0.75,
      "learning_rate": 1.96875e-05,
      "loss": 0.757,
      "step": 9
    },
    {
      "epoch": 1.5,
      "learning_rate": 1.6875e-05,
      "loss": 0.7401,
      "step": 18
    },
    {
      "epoch": 2.25,
      "learning_rate": 1.4062500000000001e-05,
      "loss": 0.714,
      "step": 27
    },
    {
      "epoch": 3.0,
      "learning_rate": 1.125e-05,
      "loss": 0.6912,
      "step": 36
    },
    {
      "epoch": 3.75,
      "learning_rate": 8.4375e-06,
      "loss": 0.6648,
      "step": 45
    },
    {
      "epoch": 4.5,
      "learning_rate": 5.625e-06,
      "loss": 0.6424,
      "step": 54
    },
    {
      "epoch": 5.25,
      "learning_rate": 2.8125e-06,
      "loss": 0.6191,
      "step": 63
    },
    {
      "epoch": 6.0,
      "learning_rate": 0.0,
      "loss": 0.6124,
      "step": 72
    }
  ],
  "logging_steps": 9,
  "max_steps": 72,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 6,
  "save_steps": 500,
  "total_flos": 4.682541225633382e+16,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}