metadata
dataset_info:
features:
- name: title
dtype: string
- name: context
dtype: string
- name: question
dtype: string
- name: id
dtype: string
- name: answers
struct:
- name: text
sequence: string
- name: answer_start
sequence: int64
splits:
- name: train
num_bytes: 83769368
num_examples: 57923
- name: validation
num_bytes: 9244735
num_examples: 6436
download_size: 57373216
dataset_size: 93014103