{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 35.08771929824562,
  "eval_steps": 500,
  "global_step": 2000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 8.771929824561404,
      "grad_norm": 5.547207832336426,
      "learning_rate": 3.9035087719298244e-05,
      "loss": 2.3503,
      "step": 500
    },
    {
      "epoch": 17.54385964912281,
      "grad_norm": 4.723608493804932,
      "learning_rate": 2.8070175438596492e-05,
      "loss": 1.6049,
      "step": 1000
    },
    {
      "epoch": 26.31578947368421,
      "grad_norm": 3.2716429233551025,
      "learning_rate": 1.7105263157894737e-05,
      "loss": 1.1304,
      "step": 1500
    },
    {
      "epoch": 35.08771929824562,
      "grad_norm": 3.1184732913970947,
      "learning_rate": 6.140350877192982e-06,
      "loss": 0.8743,
      "step": 2000
    }
  ],
  "logging_steps": 500,
  "max_steps": 2280,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 40,
  "save_steps": 500,
  "total_flos": 2072045813760000.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}