metadata
dataset_info:
features:
- name: labels
dtype: string
- name: mask_text
dtype: string
splits:
- name: train
num_bytes: 1280772471
num_examples: 1168190
download_size: 498464131
dataset_size: 1280772471
configs:
- config_name: default
data_files:
- split: train
path: data/train-*