---
dataset_info:
  features:
  - name: meta_info
    dtype: string
  - name: question
    dtype: string
  - name: answer_idx
    dtype: string
  - name: answer
    dtype: string
  - name: options
    list:
    - name: key
      dtype: string
    - name: value
      dtype: string
  - name: responses
    list:
      list:
      - name: content
        dtype: string
      - name: responses
        sequence: string
      - name: role
        dtype: string
  - name: rewards
    sequence: int64
  splits:
  - name: train
    num_bytes: 92307848.95608175
    num_examples: 8475
  download_size: 38217974
  dataset_size: 92307848.95608175
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---