{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.909090909090909,
  "eval_steps": 500,
  "global_step": 48,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.6060606060606061,
      "grad_norm": 2.365993915131312,
      "learning_rate": 0.0001934016108732548,
      "loss": 4.9025,
      "step": 10
    },
    {
      "epoch": 1.2121212121212122,
      "grad_norm": 0.5010752859822769,
      "learning_rate": 0.00014572423233046386,
      "loss": 2.152,
      "step": 20
    },
    {
      "epoch": 1.8181818181818183,
      "grad_norm": 0.32099488413380073,
      "learning_rate": 7.470666176083192e-05,
      "loss": 1.8198,
      "step": 30
    },
    {
      "epoch": 2.4242424242424243,
      "grad_norm": 0.2622629063842015,
      "learning_rate": 1.660021821101222e-05,
      "loss": 1.7642,
      "step": 40
    },
    {
      "epoch": 2.909090909090909,
      "step": 48,
      "total_flos": 88796276195328.0,
      "train_loss": 2.5086214343706765,
      "train_runtime": 154.8962,
      "train_samples_per_second": 4.997,
      "train_steps_per_second": 0.31
    }
  ],
  "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": 88796276195328.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}