---
dataset_info:
  features:
  - name: question
    dtype: string
  - name: text
    dtype: string
  - name: source
    dtype: string
  splits:
  - name: train
    num_bytes: 6269346
    num_examples: 10481
  download_size: 3430433
  dataset_size: 6269346
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---

This dataset has been created using the "support" part of the dataset: https://huggingface.co/datasets/allenai/sciq