metadata
dataset_info:
features:
- name: target
dtype: string
- name: source
dtype: string
splits:
- name: train
num_bytes: 67057726
num_examples: 77580
- name: validation
num_bytes: 5486609
num_examples: 5928
- name: test
num_bytes: 7941840
num_examples: 9241
download_size: 17681227
dataset_size: 80486175
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*