metadata
language:
- en
dataset_info:
features:
- name: answer
dtype: string
- name: context
dtype: string
- name: question
dtype: string
splits:
- name: train
num_bytes: 36151581
num_examples: 105000
download_size: 4149229
dataset_size: 36151581
configs:
- config_name: default
data_files:
- split: train
path: data/train-*