metadata
dataset_info:
features:
- name: question
dtype: string
- name: answer_start
dtype: int64
- name: answer
dtype: string
- name: context
dtype: string
- name: title
dtype: string
splits:
- name: train
num_bytes: 105138642
num_examples: 32458
download_size: 7320201
dataset_size: 105138642