---
dataset_info:
  features:
  - name: question
    dtype: string
  - name: ground_truth
    dtype: float64
  splits:
  - name: train
    num_bytes: 17311347
    num_examples: 66454
  download_size: 8957845
  dataset_size: 17311347
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---