metadata
dataset_info:
features:
- name: corpus
dtype: string
splits:
- name: train
num_bytes: 13599706.146208838
num_examples: 69927
download_size: 8145333
dataset_size: 13599706.146208838
configs:
- config_name: default
data_files:
- split: train
path: data/train-*