README.md

---
dataset_info:
  features:
  - name: prompt
    dtype: string
  - name: chosen
    dtype: string
  - name: chosen_score
    dtype: float64
  - name: rejected
    dtype: string
  - name: rejected_score
    dtype: float64
  splits:
  - name: train
    num_bytes: 26507779
    num_examples: 3410
  - name: validation
    num_bytes: 1398384
    num_examples: 183
  download_size: 18743233
  dataset_size: 27906163
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: validation
    path: data/validation-*
---