metadata
dataset_info:
features:
- name: title
dtype: string
- name: context
dtype: string
- name: question
dtype: string
- name: answer
dtype: string
- name: paragraph_id
dtype: string
splits:
- name: train
num_bytes: 73536399
num_examples: 75722
- name: validation
num_bytes: 10455240
num_examples: 10570
- name: test
num_bytes: 10735398
num_examples: 11877
download_size: 16965943
dataset_size: 94727037
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*