metadata
dataset_info:
features:
- name: text
dtype: string
- name: mask
sequence: int64
splits:
- name: train
num_bytes: 402732
num_examples: 5000
download_size: 167624
dataset_size: 402732
configs:
- config_name: default
data_files:
- split: train
path: data/train-*