metadata
dataset_info:
features:
- name: id
dtype: string
- name: title
dtype: string
- name: context
dtype: string
- name: question
dtype: string
- name: answers
sequence:
- name: text
dtype: string
- name: answer_start
dtype: int32
splits:
- name: train
num_bytes: 52948607
num_examples: 49881
- name: validation
num_bytes: 5006461
num_examples: 5077
download_size: 29688806
dataset_size: 57955068