metadata
dataset_info:
features:
- name: question
dtype: string
- name: ground_truth
dtype: string
- name: context
dtype: string
splits:
- name: train
num_bytes: 45578423
num_examples: 4719
download_size: 24726626
dataset_size: 45578423
configs:
- config_name: default
data_files:
- split: train
path: data/train-*