{
  "best_metric": 0.0521879605948925,
  "best_model_checkpoint": "./teapotllm/checkpoint-836",
  "epoch": 4.0,
  "eval_steps": 500,
  "global_step": 836,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 0.16533340513706207,
      "learning_rate": 4e-05,
      "loss": 0.6538,
      "step": 209
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.06449087709188461,
      "eval_runtime": 2.3234,
      "eval_samples_per_second": 45.193,
      "eval_steps_per_second": 6.026,
      "step": 209
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.256564199924469,
      "learning_rate": 3e-05,
      "loss": 0.0551,
      "step": 418
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.05665843188762665,
      "eval_runtime": 2.3233,
      "eval_samples_per_second": 45.195,
      "eval_steps_per_second": 6.026,
      "step": 418
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.4588538706302643,
      "learning_rate": 2e-05,
      "loss": 0.0463,
      "step": 627
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.053252462297677994,
      "eval_runtime": 2.3194,
      "eval_samples_per_second": 45.27,
      "eval_steps_per_second": 6.036,
      "step": 627
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.6458032727241516,
      "learning_rate": 1e-05,
      "loss": 0.0411,
      "step": 836
    },
    {
      "epoch": 4.0,
      "eval_loss": 0.0521879605948925,
      "eval_runtime": 2.3228,
      "eval_samples_per_second": 45.203,
      "eval_steps_per_second": 6.027,
      "step": 836
    }
  ],
  "logging_steps": 500,
  "max_steps": 1045,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1142175263883264.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}