metadata
dataset_info:
features:
- name: source
dtype: string
- name: target
dtype: string
splits:
- name: train
num_bytes: 7211001
num_examples: 28497
download_size: 3836529
dataset_size: 7211001
configs:
- config_name: default
data_files:
- split: train
path: data/train-*