metadata
dataset_info:
features:
- name: prefix
dtype: string
- name: suffix
dtype: string
- name: reference
dtype: string
splits:
- name: train
num_bytes: 21975
num_examples: 50
download_size: 13002
dataset_size: 21975
configs:
- config_name: default
data_files:
- split: train
path: data/train-*