metadata
dataset_info:
features:
- name: text
dtype: string
splits:
- name: clean
num_bytes: 100325969043
num_examples: 12653878
- name: noisy
num_bytes: 144185494007
num_examples: 19713534
download_size: 139825706722
dataset_size: 244511463050
configs:
- config_name: default
data_files:
- split: clean
path: data/clean-*
- split: noisy
path: data/noisy-*