---
dataset_info:
  features:
  - name: data_id
    dtype: string
  - name: chosen_rule_5_yes_prob
    dtype: float64
  - name: chosen_rule_5_no_prob
    dtype: float64
  - name: rejected_rule_5_yes_prob
    dtype: float64
  - name: rejected_rule_5_no_prob
    dtype: float64
  - name: __index_level_0__
    dtype: int64
  splits:
  - name: train
    num_bytes: 4153890
    num_examples: 85000
  download_size: 3457981
  dataset_size: 4153890
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---