metadata
dataset_info:
features:
- name: question
dtype: string
- name: id
dtype: int64
- name: answers
sequence: string
- name: generated_answer
dtype: string
splits:
- name: train
num_bytes: 8755
num_examples: 100
download_size: 7850
dataset_size: 8755
configs:
- config_name: default
data_files:
- split: train
path: data/train-*