---
dataset_info:
  features:
  - name: qid
    dtype: string
  - name: question
    dtype: string
  - name: answer_candidates
    sequence: string
  - name: correct_answer_index
    dtype: int64
  - name: input_ids
    sequence:
      sequence: int32
  - name: token_type_ids
    sequence:
      sequence: int8
  - name: attention_mask
    sequence:
      sequence: int8
  splits:
  - name: train
    num_bytes: 80377569
    num_examples: 13061
  - name: validation
    num_bytes: 6438977
    num_examples: 995
  - name: test
    num_bytes: 6443893
    num_examples: 997
  download_size: 16067303
  dataset_size: 93260439
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: validation
    path: data/validation-*
  - split: test
    path: data/test-*
---