metadata
dataset_info:
features:
- name: id
dtype: int64
- name: question
dtype: string
- name: answers
sequence: string
- name: meta_data
list:
- name: answer
dtype: string
- name: answer_alias
sequence: string
- name: cloze
dtype: string
- name: question
dtype: string
- name: document
dtype: string
splits:
- name: train
num_bytes: 839301
num_examples: 729
download_size: 116745
dataset_size: 839301
configs:
- config_name: default
data_files:
- split: train
path: data/train-*