metadata
dataset_info:
features:
- name: language
dtype: string
- name: question
sequence: string
- name: context
sequence: string
- name: references
struct:
- name: answers
struct:
- name: answer_start
sequence: int64
- name: text
sequence: string
- name: id
dtype: string
- name: id
dtype: string
- name: labels
dtype: bool
splits:
- name: train
num_bytes: 30320669
num_examples: 29800
- name: validation
num_bytes: 3761508
num_examples: 3709
download_size: 17981416
dataset_size: 34082177