{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.5555555555555556,
  "eval_steps": 25,
  "global_step": 100,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.14,
      "grad_norm": 6.271986484527588,
      "learning_rate": 2.3847695390781562e-05,
      "loss": 0.9018,
      "step": 25
    },
    {
      "epoch": 0.14,
      "eval_loss": 0.6543389558792114,
      "eval_runtime": 33.9351,
      "eval_samples_per_second": 1.798,
      "eval_steps_per_second": 0.236,
      "step": 25
    },
    {
      "epoch": 0.28,
      "grad_norm": 6.715949535369873,
      "learning_rate": 2.2595190380761526e-05,
      "loss": 0.6032,
      "step": 50
    },
    {
      "epoch": 0.28,
      "eval_loss": 0.6594853401184082,
      "eval_runtime": 33.9905,
      "eval_samples_per_second": 1.795,
      "eval_steps_per_second": 0.235,
      "step": 50
    },
    {
      "epoch": 0.42,
      "grad_norm": 5.575837135314941,
      "learning_rate": 2.1342685370741483e-05,
      "loss": 0.5277,
      "step": 75
    },
    {
      "epoch": 0.42,
      "eval_loss": 0.6358431577682495,
      "eval_runtime": 33.9553,
      "eval_samples_per_second": 1.796,
      "eval_steps_per_second": 0.236,
      "step": 75
    },
    {
      "epoch": 0.56,
      "grad_norm": 6.133337020874023,
      "learning_rate": 2.0090180360721444e-05,
      "loss": 0.4387,
      "step": 100
    },
    {
      "epoch": 0.56,
      "eval_loss": 0.6550642251968384,
      "eval_runtime": 33.9823,
      "eval_samples_per_second": 1.795,
      "eval_steps_per_second": 0.235,
      "step": 100
    }
  ],
  "logging_steps": 25,
  "max_steps": 500,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 25,
  "total_flos": 4421038846771200.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}