metadata
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*
- split: validation
path: data/validation-*
dataset_info:
features:
- name: original_sent
dtype: string
- name: masked_sent
dtype: string
splits:
- name: train
num_bytes: 14319276284
num_examples: 48833571
- name: test
num_bytes: 2144523252
num_examples: 6104196
- name: validation
num_bytes: 2169841655
num_examples: 6104197
download_size: 8287754892
dataset_size: 18633641191