metadata
dataset_info:
features:
- name: question
dtype: string
- name: answer
sequence: string
- name: context
sequence: string
splits:
- name: train
num_bytes: 261632248
num_examples: 79168
- name: validation
num_bytes: 28956240
num_examples: 8757
- name: test
num_bytes: 12008906
num_examples: 3610
download_size: 167269853
dataset_size: 302597394
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*