metadata
dataset_info:
features:
- name: answer
dtype: string
- name: question
dtype: string
- name: tasktype
dtype: string
splits:
- name: training
num_bytes: 39869724
num_examples: 8688
download_size: 5824180
dataset_size: 39869724
configs:
- config_name: default
data_files:
- split: training
path: data/training-*