{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.769230769230769,
  "eval_steps": 500,
  "global_step": 18,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.15,
      "grad_norm": 7.469101605181307,
      "learning_rate": 1e-05,
      "loss": 1.1372,
      "step": 1
    },
    {
      "epoch": 0.77,
      "grad_norm": 14.904814418948556,
      "learning_rate": 1.8314696123025456e-05,
      "loss": 1.6461,
      "step": 5
    },
    {
      "epoch": 0.92,
      "eval_loss": 1.7000776529312134,
      "eval_runtime": 1.5213,
      "eval_samples_per_second": 4.601,
      "eval_steps_per_second": 0.657,
      "step": 6
    },
    {
      "epoch": 1.54,
      "grad_norm": 5.728000234517732,
      "learning_rate": 1e-05,
      "loss": 1.1299,
      "step": 10
    },
    {
      "epoch": 2.0,
      "eval_loss": 1.6488012075424194,
      "eval_runtime": 1.3181,
      "eval_samples_per_second": 5.311,
      "eval_steps_per_second": 0.759,
      "step": 13
    },
    {
      "epoch": 2.31,
      "grad_norm": 3.997696868462948,
      "learning_rate": 1.6853038769745466e-06,
      "loss": 0.9123,
      "step": 15
    },
    {
      "epoch": 2.77,
      "eval_loss": 1.6476205587387085,
      "eval_runtime": 1.3248,
      "eval_samples_per_second": 5.284,
      "eval_steps_per_second": 0.755,
      "step": 18
    },
    {
      "epoch": 2.77,
      "step": 18,
      "total_flos": 3716488888320.0,
      "train_loss": 1.1346278654204474,
      "train_runtime": 313.3964,
      "train_samples_per_second": 3.906,
      "train_steps_per_second": 0.057
    }
  ],
  "logging_steps": 5,
  "max_steps": 18,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "total_flos": 3716488888320.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}