---
dataset_info:
  features:
  - name: prompt
    dtype: string
  - name: chosen
    dtype: string
  - name: rejected
    dtype: string
  - name: preference
    dtype: string
  - name: reason
    dtype: string
  splits:
  - name: train
    num_bytes: 11790
    num_examples: 10
  download_size: 12249
  dataset_size: 11790
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---