metadata
dataset_info:
features:
- name: id
dtype: string
- name: title
dtype: string
- name: context
dtype: string
- name: question
dtype: string
- name: answers
struct:
- name: answer_start
sequence: int64
- name: text
sequence: string
splits:
- name: validation
num_bytes: 2902927
num_examples: 3809
- name: test
num_bytes: 3667235
num_examples: 4152
- name: train
num_bytes: 13859759
num_examples: 17056
download_size: 2662048
dataset_size: 20429921