---
dataset_info:
  features:
  - name: comparison_prompt
    dtype: string
  - name: messages
    dtype: string
  - name: generation
    dtype: string
  - name: model_name
    dtype: string
  splits:
  - name: train
    num_bytes: 17706850
    num_examples: 1083
  download_size: 9845062
  dataset_size: 17706850
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---