metadata
dataset_info:
features:
- name: background
dtype: string
- name: context
dtype: string
- name: answer
dtype: string
splits:
- name: train
num_bytes: 22366884
num_examples: 12277
- name: validation
num_bytes: 4836352
num_examples: 2653
- name: test
num_bytes: 5198019
num_examples: 2810
download_size: 5144868
dataset_size: 32401255
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*