{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.8298755186721992,
  "eval_steps": 500,
  "global_step": 4000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.1,
      "grad_norm": 59.103092193603516,
      "learning_rate": 5e-05,
      "loss": 3.2072,
      "step": 500
    },
    {
      "epoch": 0.21,
      "grad_norm": 4.993757247924805,
      "learning_rate": 4.726477024070022e-05,
      "loss": 1.7111,
      "step": 1000
    },
    {
      "epoch": 0.31,
      "grad_norm": 21.068601608276367,
      "learning_rate": 4.452954048140044e-05,
      "loss": 1.1954,
      "step": 1500
    },
    {
      "epoch": 0.41,
      "grad_norm": 46.41459274291992,
      "learning_rate": 4.179431072210066e-05,
      "loss": 0.8376,
      "step": 2000
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.6010065078735352,
      "learning_rate": 3.9059080962800876e-05,
      "loss": 0.6233,
      "step": 2500
    },
    {
      "epoch": 0.62,
      "grad_norm": 57.68833541870117,
      "learning_rate": 3.6323851203501094e-05,
      "loss": 0.5353,
      "step": 3000
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.1653015911579132,
      "learning_rate": 3.358862144420131e-05,
      "loss": 0.3499,
      "step": 3500
    },
    {
      "epoch": 0.83,
      "grad_norm": 8.111839294433594,
      "learning_rate": 3.085339168490153e-05,
      "loss": 0.3051,
      "step": 4000
    }
  ],
  "logging_steps": 500,
  "max_steps": 9640,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 4000,
  "total_flos": 8668418408448000.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}