{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 33,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.9090909090909091,
      "grad_norm": 0.22085539996623993,
      "learning_rate": 0.00016889669190756868,
      "loss": 0.7953,
      "num_input_tokens_seen": 327680,
      "step": 10
    },
    {
      "epoch": 1.8181818181818183,
      "grad_norm": 0.1958368867635727,
      "learning_rate": 7.493474677412794e-05,
      "loss": 0.6435,
      "num_input_tokens_seen": 655360,
      "step": 20
    },
    {
      "epoch": 2.7272727272727275,
      "grad_norm": 0.1823256015777588,
      "learning_rate": 4.586074359995119e-06,
      "loss": 0.6208,
      "num_input_tokens_seen": 983040,
      "step": 30
    },
    {
      "epoch": 3.0,
      "num_input_tokens_seen": 1081344,
      "step": 33,
      "total_flos": 1.819712064533299e+16,
      "train_loss": 0.6852215998100512,
      "train_runtime": 65.3842,
      "train_samples_per_second": 7.984,
      "train_steps_per_second": 0.505
    }
  ],
  "logging_steps": 10,
  "max_steps": 33,
  "num_input_tokens_seen": 1081344,
  "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": 1.819712064533299e+16,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}