{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 48,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.6349206349206349,
      "grad_norm": 25.672040939331055,
      "learning_rate": 4.894001499771015e-05,
      "loss": 8.6936,
      "step": 10
    },
    {
      "epoch": 1.253968253968254,
      "grad_norm": 5.61480712890625,
      "learning_rate": 3.8023508512198256e-05,
      "loss": 0.4693,
      "step": 20
    },
    {
      "epoch": 1.8888888888888888,
      "grad_norm": 3.9594666957855225,
      "learning_rate": 2.0459078725514092e-05,
      "loss": 0.1413,
      "step": 30
    },
    {
      "epoch": 2.507936507936508,
      "grad_norm": 0.6755509376525879,
      "learning_rate": 5.2125877892686496e-06,
      "loss": 0.0773,
      "step": 40
    },
    {
      "epoch": 3.0,
      "step": 48,
      "total_flos": 1.1974639068643328e+16,
      "train_loss": 1.9668061497310798,
      "train_runtime": 112.3578,
      "train_samples_per_second": 53.401,
      "train_steps_per_second": 0.427
    }
  ],
  "logging_steps": 10,
  "max_steps": 48,
  "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": 1.1974639068643328e+16,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}