{
  "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": 10.593970638360702,
      "learning_rate": 5e-07,
      "loss": 1.6581,
      "step": 1
    },
    {
      "epoch": 0.77,
      "grad_norm": 6.961762957949474,
      "learning_rate": 9.157348061512726e-07,
      "loss": 1.6814,
      "step": 5
    },
    {
      "epoch": 0.92,
      "eval_loss": 2.228210210800171,
      "eval_runtime": 4.1688,
      "eval_samples_per_second": 11.034,
      "eval_steps_per_second": 0.72,
      "step": 6
    },
    {
      "epoch": 1.54,
      "grad_norm": 5.942452001829785,
      "learning_rate": 5e-07,
      "loss": 1.6177,
      "step": 10
    },
    {
      "epoch": 2.0,
      "eval_loss": 2.1941089630126953,
      "eval_runtime": 3.9662,
      "eval_samples_per_second": 11.598,
      "eval_steps_per_second": 0.756,
      "step": 13
    },
    {
      "epoch": 2.31,
      "grad_norm": 4.753177762599499,
      "learning_rate": 8.426519384872732e-08,
      "loss": 1.4862,
      "step": 15
    },
    {
      "epoch": 2.77,
      "eval_loss": 2.1894760131835938,
      "eval_runtime": 3.9745,
      "eval_samples_per_second": 11.574,
      "eval_steps_per_second": 0.755,
      "step": 18
    },
    {
      "epoch": 2.77,
      "step": 18,
      "total_flos": 3716488888320.0,
      "train_loss": 1.5841291414366827,
      "train_runtime": 319.5018,
      "train_samples_per_second": 3.793,
      "train_steps_per_second": 0.056
    }
  ],
  "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
}