metadata
dataset_info:
features:
- name: is_training
dtype: bool
- name: text
dtype: string
splits:
- name: train
num_bytes: 2523644009
num_examples: 2390347
- name: test
num_bytes: 25412390
num_examples: 24119
download_size: 838283668
dataset_size: 2549056399
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*