metadata
dataset_info:
features:
- name: question
dtype: string
- name: answer
dtype: string
- name: contexts
sequence: string
- name: ground_truth
dtype: string
splits:
- name: train
num_bytes: 178916
num_examples: 50
download_size: 105114
dataset_size: 178916
configs:
- config_name: default
data_files:
- split: train
path: data/train-*