---
dataset_info:
  features:
  - name: question
    dtype: string
  - name: choices
    sequence: string
  - name: answer
    dtype: string
  - name: rationale
    dtype: string
  - name: split
    dtype: string
  - name: source_dataset
    dtype: string
  - name: subject
    dtype: 'null'
  splits:
  - name: train
    num_bytes: 49514739
    num_examples: 109475
  - name: test
    num_bytes: 706021
    num_examples: 1254
  - name: validation
    num_bytes: 705574
    num_examples: 1254
  download_size: 28975940
  dataset_size: 50926334
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: test
    path: data/test-*
  - split: validation
    path: data/validation-*
---