---
dataset_info:
  features:
  - name: prompt
    dtype: string
  - name: requirements
    dtype: string
  - name: response_0
    dtype: string
  - name: response_1
    dtype: string
  - name: check
    dtype: string
  - name: importance
    dtype: int64
  - name: response_0_1_judged_preference_mean
    list: float64
  - name: response_0_1_judged_preference_majority
    list: int64
  splits:
  - name: train
    num_bytes: 33754
    num_examples: 10
  download_size: 31567
  dataset_size: 33754
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---