metadata
dataset_info:
features:
- name: docid
dtype: string
- name: text
dtype: string
- name: lang
dtype: string
splits:
- name: train
num_bytes: 5032156751
num_examples: 268022
download_size: 2820467982
dataset_size: 5032156751
configs:
- config_name: default
data_files:
- split: train
path: data/train-*