metadata
dataset_info:
features:
- name: text
dtype: string
splits:
- name: clean
num_bytes: 307909215491
num_examples: 21818123
- name: noisy
num_bytes: 310336391232
num_examples: 23320722
download_size: 323402381156
dataset_size: 618245606723
configs:
- config_name: default
data_files:
- split: clean
path: data/clean-*
- split: noisy
path: data/noisy-*