{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.005326569435637286,
  "eval_steps": 63,
  "global_step": 63,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 8.454872120059185e-05,
      "eval_loss": NaN,
      "eval_runtime": 228.2507,
      "eval_samples_per_second": 10.909,
      "eval_steps_per_second": 5.455,
      "step": 1
    },
    {
      "epoch": 0.0008454872120059184,
      "grad_norm": NaN,
      "learning_rate": 0.00019979453927503364,
      "loss": 5.6754,
      "step": 10
    },
    {
      "epoch": 0.0016909744240118367,
      "grad_norm": NaN,
      "learning_rate": 0.00019815591569910654,
      "loss": 6.8971,
      "step": 20
    },
    {
      "epoch": 0.0025364616360177552,
      "grad_norm": NaN,
      "learning_rate": 0.00019490557470106686,
      "loss": 3.6576,
      "step": 30
    },
    {
      "epoch": 0.0033819488480236735,
      "grad_norm": NaN,
      "learning_rate": 0.0001900968867902419,
      "loss": 0.6001,
      "step": 40
    },
    {
      "epoch": 0.004227436060029592,
      "grad_norm": NaN,
      "learning_rate": 0.00018380881048918405,
      "loss": 1.2811,
      "step": 50
    },
    {
      "epoch": 0.0050729232720355105,
      "grad_norm": NaN,
      "learning_rate": 0.00017614459583691346,
      "loss": 7.1498,
      "step": 60
    },
    {
      "epoch": 0.005326569435637286,
      "eval_loss": NaN,
      "eval_runtime": 227.6495,
      "eval_samples_per_second": 10.938,
      "eval_steps_per_second": 5.469,
      "step": 63
    }
  ],
  "logging_steps": 10,
  "max_steps": 250,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 63,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 2.033089127645184e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}