---
dataset_info:
  features:
  - name: question
    dtype: string
  - name: answer
    dtype: string
  - name: steps
    sequence: string
  splits:
  - name: train
    num_bytes: 186991963
    num_examples: 100000
  - name: valid
    num_bytes: 9159418
    num_examples: 5000
  - name: test
    num_bytes: 9629641
    num_examples: 5000
  download_size: 129657345
  dataset_size: 205781022
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: valid
    path: data/valid-*
  - split: test
    path: data/test-*
---