{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.001965923984272608,
  "eval_steps": 3,
  "global_step": 3,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.000655307994757536,
      "grad_norm": 0.15108448266983032,
      "learning_rate": 2e-05,
      "loss": 0.5443,
      "step": 1
    },
    {
      "epoch": 0.000655307994757536,
      "eval_loss": 0.9055324196815491,
      "eval_runtime": 104.7289,
      "eval_samples_per_second": 24.549,
      "eval_steps_per_second": 3.075,
      "step": 1
    },
    {
      "epoch": 0.001310615989515072,
      "grad_norm": NaN,
      "learning_rate": 2e-05,
      "loss": 0.629,
      "step": 2
    },
    {
      "epoch": 0.001965923984272608,
      "grad_norm": 0.15483306348323822,
      "learning_rate": 4e-05,
      "loss": 0.5365,
      "step": 3
    },
    {
      "epoch": 0.001965923984272608,
      "eval_loss": 0.9036340713500977,
      "eval_runtime": 105.1959,
      "eval_samples_per_second": 24.44,
      "eval_steps_per_second": 3.061,
      "step": 3
    }
  ],
  "logging_steps": 1,
  "max_steps": 10,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 3,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 3.053742363299021e+16,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}