metadata
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
dataset_info:
features:
- name: id
dtype: string
- name: text
dtype: string
splits:
- name: train
num_bytes: 84244312900
num_examples: 54573064
download_size: 21571487893
dataset_size: 84244312900