{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 75,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 0.89453125,
      "learning_rate": 1.883869132745561e-05,
      "loss": 2.2779,
      "step": 15
    },
    {
      "epoch": 1.0,
      "eval_loss": 1.9970979690551758,
      "eval_runtime": 9.2739,
      "eval_samples_per_second": 6.146,
      "eval_steps_per_second": 0.431,
      "step": 15
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.94140625,
      "learning_rate": 1.408083612243465e-05,
      "loss": 1.8274,
      "step": 30
    },
    {
      "epoch": 2.0,
      "eval_loss": 1.636780023574829,
      "eval_runtime": 9.2796,
      "eval_samples_per_second": 6.143,
      "eval_steps_per_second": 0.431,
      "step": 30
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.62890625,
      "learning_rate": 7.590322975433857e-06,
      "loss": 1.5549,
      "step": 45
    },
    {
      "epoch": 3.0,
      "eval_loss": 1.4507859945297241,
      "eval_runtime": 9.2973,
      "eval_samples_per_second": 6.131,
      "eval_steps_per_second": 0.43,
      "step": 45
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.58203125,
      "learning_rate": 2.1229202668228197e-06,
      "loss": 1.4289,
      "step": 60
    },
    {
      "epoch": 4.0,
      "eval_loss": 1.3850109577178955,
      "eval_runtime": 9.3004,
      "eval_samples_per_second": 6.129,
      "eval_steps_per_second": 0.43,
      "step": 60
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.58203125,
      "learning_rate": 0.0,
      "loss": 1.3983,
      "step": 75
    },
    {
      "epoch": 5.0,
      "eval_loss": 1.3798422813415527,
      "eval_runtime": 9.2908,
      "eval_samples_per_second": 6.135,
      "eval_steps_per_second": 0.431,
      "step": 75
    },
    {
      "epoch": 5.0,
      "step": 75,
      "total_flos": 1.0301359320465408e+17,
      "train_loss": 1.6974857076009116,
      "train_runtime": 651.9512,
      "train_samples_per_second": 1.795,
      "train_steps_per_second": 0.115
    }
  ],
  "logging_steps": 500,
  "max_steps": 75,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "total_flos": 1.0301359320465408e+17,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}