metadata
dataset_info:
features:
- name: input_text
dtype: string
- name: target_text
dtype: string
- name: language
sequence: string
splits:
- name: train
num_bytes: 78992628
num_examples: 208137
- name: validation
num_bytes: 1891444
num_examples: 5000
- name: test
num_bytes: 1887850
num_examples: 5000
- name: ood_test
num_bytes: 135451
num_examples: 500
download_size: 36387774
dataset_size: 82907373
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*
- split: ood_test
path: data/ood_test-*