{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0666666666666667,
  "eval_steps": 500,
  "global_step": 120,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.09,
      "grad_norm": 1.0839761544276303,
      "learning_rate": 1.8e-05,
      "loss": 0.4354,
      "step": 10
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.5839985259836128,
      "learning_rate": 3.8e-05,
      "loss": 0.2119,
      "step": 20
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.4201387856386283,
      "learning_rate": 3.991999431715994e-05,
      "loss": 0.1912,
      "step": 30
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.36274725233758565,
      "learning_rate": 3.964425314282723e-05,
      "loss": 0.1801,
      "step": 40
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.4211938897484086,
      "learning_rate": 3.9174511238257296e-05,
      "loss": 0.1766,
      "step": 50
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.3292859661705254,
      "learning_rate": 3.851540763308629e-05,
      "loss": 0.1697,
      "step": 60
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.3400702213716709,
      "learning_rate": 3.7673451436099125e-05,
      "loss": 0.1686,
      "step": 70
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.32704307193400634,
      "learning_rate": 3.665695755323067e-05,
      "loss": 0.1635,
      "step": 80
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.37275253301881667,
      "learning_rate": 3.5475964572057876e-05,
      "loss": 0.1622,
      "step": 90
    },
    {
      "epoch": 0.89,
      "grad_norm": 0.3274514135221438,
      "learning_rate": 3.4142135623730954e-05,
      "loss": 0.1616,
      "step": 100
    },
    {
      "epoch": 0.98,
      "grad_norm": 0.3147797609272402,
      "learning_rate": 3.2668643201401225e-05,
      "loss": 0.1594,
      "step": 110
    },
    {
      "epoch": 1.07,
      "grad_norm": 0.3189184124018789,
      "learning_rate": 3.1070039072644316e-05,
      "loss": 0.1111,
      "step": 120
    }
  ],
  "logging_steps": 10,
  "max_steps": 336,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 120,
  "total_flos": 157833104457728.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}