metadata
dataset_info:
features:
- name: sentence
dtype: string
splits:
- name: train
num_bytes: 591658727.6825764
num_examples: 4000000
- name: valid
num_bytes: 66561606.86428984
num_examples: 450000
- name: test
num_bytes: 3224360
num_examples: 22210
download_size: 454512275
dataset_size: 661444694.5468663