metadata
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: valid
path: data/valid-*
dataset_info:
features:
- name: tgt
dtype: string
splits:
- name: train
num_bytes: 1571222812.8
num_examples: 6400000
- name: valid
num_bytes: 392805703.2
num_examples: 1600000
download_size: 888047205
dataset_size: 1964028516