{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 93,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.16260162601626016,
      "grad_norm": 5.873813152313232,
      "learning_rate": 9.99695413509548e-06,
      "loss": 0.6874,
      "step": 5
    },
    {
      "epoch": 0.3252032520325203,
      "grad_norm": 4.466193199157715,
      "learning_rate": 9.890738003669029e-06,
      "loss": 0.5756,
      "step": 10
    },
    {
      "epoch": 0.4878048780487805,
      "grad_norm": 4.869799613952637,
      "learning_rate": 9.635919272833938e-06,
      "loss": 0.466,
      "step": 15
    },
    {
      "epoch": 0.6504065040650406,
      "grad_norm": 2.685967445373535,
      "learning_rate": 9.24024048078213e-06,
      "loss": 0.3592,
      "step": 20
    },
    {
      "epoch": 0.8130081300813008,
      "grad_norm": 2.5385570526123047,
      "learning_rate": 8.715724127386971e-06,
      "loss": 0.3222,
      "step": 25
    },
    {
      "epoch": 0.975609756097561,
      "grad_norm": 2.7402403354644775,
      "learning_rate": 8.078307376628292e-06,
      "loss": 0.3074,
      "step": 30
    },
    {
      "epoch": 1.1300813008130082,
      "grad_norm": 1.7978644371032715,
      "learning_rate": 7.347357813929455e-06,
      "loss": 0.1739,
      "step": 35
    },
    {
      "epoch": 1.2926829268292683,
      "grad_norm": 1.8208461999893188,
      "learning_rate": 6.545084971874738e-06,
      "loss": 0.1577,
      "step": 40
    },
    {
      "epoch": 1.4552845528455285,
      "grad_norm": 2.2071800231933594,
      "learning_rate": 5.695865504800328e-06,
      "loss": 0.145,
      "step": 45
    },
    {
      "epoch": 1.6178861788617886,
      "grad_norm": 2.227810859680176,
      "learning_rate": 4.825502516487497e-06,
      "loss": 0.1559,
      "step": 50
    },
    {
      "epoch": 1.7804878048780488,
      "grad_norm": 2.263707399368286,
      "learning_rate": 3.960441545911205e-06,
      "loss": 0.1392,
      "step": 55
    },
    {
      "epoch": 1.943089430894309,
      "grad_norm": 2.166546106338501,
      "learning_rate": 3.12696703292044e-06,
      "loss": 0.1364,
      "step": 60
    },
    {
      "epoch": 2.097560975609756,
      "grad_norm": 1.3402647972106934,
      "learning_rate": 2.3504036788339763e-06,
      "loss": 0.0884,
      "step": 65
    },
    {
      "epoch": 2.2601626016260163,
      "grad_norm": 1.2159967422485352,
      "learning_rate": 1.6543469682057105e-06,
      "loss": 0.0711,
      "step": 70
    },
    {
      "epoch": 2.4227642276422765,
      "grad_norm": 1.5177407264709473,
      "learning_rate": 1.0599462319663906e-06,
      "loss": 0.0677,
      "step": 75
    },
    {
      "epoch": 2.5853658536585367,
      "grad_norm": 1.4129310846328735,
      "learning_rate": 5.852620357053651e-07,
      "loss": 0.069,
      "step": 80
    },
    {
      "epoch": 2.747967479674797,
      "grad_norm": 1.3288525342941284,
      "learning_rate": 2.447174185242324e-07,
      "loss": 0.0681,
      "step": 85
    },
    {
      "epoch": 2.910569105691057,
      "grad_norm": 1.3459076881408691,
      "learning_rate": 4.865965629214819e-08,
      "loss": 0.067,
      "step": 90
    }
  ],
  "logging_steps": 5,
  "max_steps": 93,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.2533230053045043e+17,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}