---
dataset_info:
  features:
  - name: split
    dtype: string
  - name: document_id
    dtype: string
  - name: questions
    sequence: string
  - name: options
    sequence:
      sequence: string
  - name: gold_label
    sequence: int64
  - name: difficult
    sequence: int64
  - name: text
    dtype: string
  splits:
  - name: train
    num_bytes: 20953508
    num_examples: 762
  download_size: 7158872
  dataset_size: 20953508
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---

Preprocessed QuALITY dataset by: https://github.com/nyu-mll/quality