{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 531,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.56,
      "grad_norm": 0.47637438774108887,
      "learning_rate": 0.00019510565162951537,
      "loss": 1.5071,
      "step": 100
    },
    {
      "epoch": 1.13,
      "grad_norm": 0.6510860919952393,
      "learning_rate": 0.00018090169943749476,
      "loss": 1.3602,
      "step": 200
    },
    {
      "epoch": 1.69,
      "grad_norm": 0.7494458556175232,
      "learning_rate": 0.00015877852522924732,
      "loss": 1.2589,
      "step": 300
    },
    {
      "epoch": 2.26,
      "grad_norm": 0.9085477590560913,
      "learning_rate": 0.00013090169943749476,
      "loss": 1.1859,
      "step": 400
    },
    {
      "epoch": 2.82,
      "grad_norm": 1.0386964082717896,
      "learning_rate": 0.0001,
      "loss": 1.093,
      "step": 500
    }
  ],
  "logging_steps": 100,
  "max_steps": 1000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 6,
  "save_steps": 500,
  "total_flos": 7860174334476288.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}