{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 4.752376188094047,
  "eval_steps": 500,
  "global_step": 9500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.25,
      "grad_norm": 40.72060012817383,
      "learning_rate": 1.8999499749874938e-05,
      "loss": 0.416,
      "step": 500
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.45640167593955994,
      "learning_rate": 1.7998999499749875e-05,
      "loss": 0.259,
      "step": 1000
    },
    {
      "epoch": 0.75,
      "grad_norm": 24.118961334228516,
      "learning_rate": 1.6998499249624815e-05,
      "loss": 0.2726,
      "step": 1500
    },
    {
      "epoch": 1.0,
      "grad_norm": 43.76200485229492,
      "learning_rate": 1.599799899949975e-05,
      "loss": 0.2471,
      "step": 2000
    },
    {
      "epoch": 1.25,
      "grad_norm": 31.93831443786621,
      "learning_rate": 1.4997498749374688e-05,
      "loss": 0.1518,
      "step": 2500
    },
    {
      "epoch": 1.5,
      "grad_norm": 0.02012985199689865,
      "learning_rate": 1.3996998499249626e-05,
      "loss": 0.1421,
      "step": 3000
    },
    {
      "epoch": 1.75,
      "grad_norm": 0.22942015528678894,
      "learning_rate": 1.2996498249124563e-05,
      "loss": 0.1463,
      "step": 3500
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.18562756478786469,
      "learning_rate": 1.19959979989995e-05,
      "loss": 0.1512,
      "step": 4000
    },
    {
      "epoch": 2.25,
      "grad_norm": 0.14861415326595306,
      "learning_rate": 1.0995497748874438e-05,
      "loss": 0.0811,
      "step": 4500
    },
    {
      "epoch": 2.5,
      "grad_norm": 0.19507120549678802,
      "learning_rate": 9.994997498749376e-06,
      "loss": 0.0634,
      "step": 5000
    },
    {
      "epoch": 2.75,
      "grad_norm": 0.011564640328288078,
      "learning_rate": 8.994497248624313e-06,
      "loss": 0.0776,
      "step": 5500
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.1320619136095047,
      "learning_rate": 7.993996998499251e-06,
      "loss": 0.0583,
      "step": 6000
    },
    {
      "epoch": 3.25,
      "grad_norm": 0.002501310082152486,
      "learning_rate": 6.9934967483741875e-06,
      "loss": 0.0283,
      "step": 6500
    },
    {
      "epoch": 3.5,
      "grad_norm": 0.008065233938395977,
      "learning_rate": 5.992996498249125e-06,
      "loss": 0.0245,
      "step": 7000
    },
    {
      "epoch": 3.75,
      "grad_norm": 0.012150153517723083,
      "learning_rate": 4.992496248124062e-06,
      "loss": 0.0282,
      "step": 7500
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.0018927728524431586,
      "learning_rate": 3.991995997999e-06,
      "loss": 0.0181,
      "step": 8000
    },
    {
      "epoch": 4.25,
      "grad_norm": 0.00044900510692968965,
      "learning_rate": 2.991495747873937e-06,
      "loss": 0.0051,
      "step": 8500
    },
    {
      "epoch": 4.5,
      "grad_norm": 0.0030473130755126476,
      "learning_rate": 1.9909954977488747e-06,
      "loss": 0.0097,
      "step": 9000
    },
    {
      "epoch": 4.75,
      "grad_norm": 0.004492764361202717,
      "learning_rate": 9.90495247623812e-07,
      "loss": 0.0048,
      "step": 9500
    }
  ],
  "logging_steps": 500,
  "max_steps": 9995,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "total_flos": 1.7704832040505344e+16,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}