metadata
dataset_info:
features:
- name: text
dtype: string
- name: next_text
dtype: string
splits:
- name: train
num_bytes: 21000861132
num_examples: 21022314
- name: test
num_bytes: 1056152
num_examples: 1052
download_size: 14372479490
dataset_size: 21001917284
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*