metadata
dataset_info:
features:
- name: source
dtype: string
- name: target
dtype: string
splits:
- name: train
num_bytes: 2838436
num_examples: 42768
download_size: 1633244
dataset_size: 2838436
configs:
- config_name: default
data_files:
- split: train
path: data/train-*