---
dataset_info:
  features:
  - name: question
    dtype: string
  - name: chosen
    dtype: string
  - name: rejected
    dtype: string
  - name: __index_level_0__
    dtype: int64
  splits:
  - name: train
    num_bytes: 18211311
    num_examples: 19534
  - name: eval
    num_bytes: 2026517
    num_examples: 2171
  download_size: 5458436
  dataset_size: 20237828
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: eval
    path: data/eval-*
---