metadata
dataset_info:
features:
- name: text
dtype: string
- name: id
dtype: string
- name: subcorpus
dtype: string
splits:
- name: train
num_bytes: 58722964960
num_examples: 16146079
download_size: 35198695914
dataset_size: 58722964960
configs:
- config_name: default
data_files:
- split: train
path: data/train-*