{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 7.5,
  "eval_steps": 500,
  "global_step": 16,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.25,
      "learning_rate": 7.5e-05,
      "loss": 0.8681,
      "step": 1
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.00015,
      "loss": 0.8696,
      "step": 2
    },
    {
      "epoch": 1.25,
      "learning_rate": 0.000225,
      "loss": 0.8552,
      "step": 3
    },
    {
      "epoch": 1.5,
      "learning_rate": 0.0003,
      "loss": 0.7928,
      "step": 4
    },
    {
      "epoch": 2.25,
      "learning_rate": 0.00028928571428571425,
      "loss": 0.7436,
      "step": 5
    },
    {
      "epoch": 2.5,
      "learning_rate": 0.00027857142857142854,
      "loss": 0.6743,
      "step": 6
    },
    {
      "epoch": 3.25,
      "learning_rate": 0.00026785714285714287,
      "loss": 0.6366,
      "step": 7
    },
    {
      "epoch": 3.5,
      "learning_rate": 0.0002571428571428571,
      "loss": 0.5894,
      "step": 8
    },
    {
      "epoch": 4.25,
      "learning_rate": 0.0002464285714285714,
      "loss": 0.5553,
      "step": 9
    },
    {
      "epoch": 4.5,
      "learning_rate": 0.00023571428571428569,
      "loss": 0.5088,
      "step": 10
    },
    {
      "epoch": 5.25,
      "learning_rate": 0.000225,
      "loss": 0.4581,
      "step": 11
    },
    {
      "epoch": 5.5,
      "learning_rate": 0.00021428571428571427,
      "loss": 0.4222,
      "step": 12
    },
    {
      "epoch": 6.25,
      "learning_rate": 0.00020357142857142856,
      "loss": 0.3699,
      "step": 13
    },
    {
      "epoch": 6.5,
      "learning_rate": 0.00019285714285714286,
      "loss": 0.3272,
      "step": 14
    },
    {
      "epoch": 7.25,
      "learning_rate": 0.00018214285714285712,
      "loss": 0.2859,
      "step": 15
    },
    {
      "epoch": 7.5,
      "learning_rate": 0.0001714285714285714,
      "loss": 0.2595,
      "step": 16
    }
  ],
  "logging_steps": 1,
  "max_steps": 32,
  "num_train_epochs": 8,
  "save_steps": 500,
  "total_flos": 3984531891683328.0,
  "trial_name": null,
  "trial_params": null
}