metadata
dataset_info:
features:
- name: lang
dtype: string
- name: text
dtype: string
- name: source
dtype: string
splits:
- name: train
num_bytes: 9732958260
num_examples: 4932629
- name: test
num_bytes: 2497636
num_examples: 1000
download_size: 5432947755
dataset_size: 9735455896
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*