metadata
dataset_info:
features:
- name: origignal_text
dtype: string
- name: masked_text
dtype: string
splits:
- name: train
num_bytes: 105504
num_examples: 503
- name: test
num_bytes: 270062
num_examples: 1247
- name: validation
num_bytes: 832730
num_examples: 3902
download_size: 860612
dataset_size: 1208296
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*
- split: validation
path: data/validation-*