---
dataset_info:
  features:
  - name: reasoning
    dtype: string
  - name: question
    dtype: string
  - name: options
    dtype: string
  - name: images
    dtype: 'null'
  - name: answer_label
    dtype: string
  - name: answer
    dtype: string
  - name: dataset_name
    dtype: string
  - name: dataset_index
    dtype: int64
  splits:
  - name: train
    num_bytes: 146789396.75137275
    num_examples: 23460
  download_size: 70583796
  dataset_size: 146789396.75137275
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---