{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 4.923076923076923,
  "eval_steps": 500,
  "global_step": 40,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.62,
      "grad_norm": 0.1295962631702423,
      "learning_rate": 0.0001923879532511287,
      "loss": 0.0652,
      "step": 5
    },
    {
      "epoch": 1.23,
      "grad_norm": 0.046223945915699005,
      "learning_rate": 0.00017071067811865476,
      "loss": 0.0346,
      "step": 10
    },
    {
      "epoch": 1.85,
      "grad_norm": 0.07042355835437775,
      "learning_rate": 0.000138268343236509,
      "loss": 0.0214,
      "step": 15
    },
    {
      "epoch": 2.46,
      "grad_norm": 0.036413274705410004,
      "learning_rate": 0.0001,
      "loss": 0.0145,
      "step": 20
    },
    {
      "epoch": 3.08,
      "grad_norm": 0.0433778315782547,
      "learning_rate": 6.173165676349103e-05,
      "loss": 0.0117,
      "step": 25
    },
    {
      "epoch": 3.69,
      "grad_norm": 0.02845873311161995,
      "learning_rate": 2.9289321881345254e-05,
      "loss": 0.007,
      "step": 30
    },
    {
      "epoch": 4.31,
      "grad_norm": 0.02572229690849781,
      "learning_rate": 7.612046748871327e-06,
      "loss": 0.0056,
      "step": 35
    },
    {
      "epoch": 4.92,
      "grad_norm": 0.022463396191596985,
      "learning_rate": 0.0,
      "loss": 0.0049,
      "step": 40
    },
    {
      "epoch": 4.92,
      "step": 40,
      "total_flos": 6.324624974954496e+16,
      "train_loss": 0.020610336679965258,
      "train_runtime": 1360.5648,
      "train_samples_per_second": 0.474,
      "train_steps_per_second": 0.029
    }
  ],
  "logging_steps": 5,
  "max_steps": 40,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 100,
  "total_flos": 6.324624974954496e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}