metadata
dataset_info:
features:
- name: type
dtype: string
- name: lang
dtype: string
- name: text
dtype: string
splits:
- name: train
num_bytes: 3614079390
num_examples: 5322920
download_size: 1392599348
dataset_size: 3614079390
configs:
- config_name: default
data_files:
- split: train
path: data/train-*