README.md

---
dataset_info:
  features:
  - name: system
    dtype: string
  - name: question
    dtype: string
  - name: chosen
    dtype: string
  - name: rejected
    dtype: string
  - name: chosen_score
    dtype: int64
  - name: rejected_score
    dtype: int64
  - name: winner
    dtype: string
  - name: topic
    dtype: string
  splits:
  - name: train
    num_bytes: 25488402
    num_examples: 7987
  download_size: 13993628
  dataset_size: 25488402
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---