metadata
dataset_info:
features:
- name: question
dtype: string
- name: actions
dtype: string
splits:
- name: train
num_bytes: 496626
num_examples: 3734
download_size: 107361
dataset_size: 496626
configs:
- config_name: default
data_files:
- split: train
path: data/train-*