metadata
dataset_info:
features:
- name: text
dtype: string
- name: tokenized_text
sequence: string
- name: mask
sequence: int64
- name: processed_text
dtype: string
splits:
- name: test
num_bytes: 21509861
num_examples: 19633
download_size: 6020792
dataset_size: 21509861
configs:
- config_name: default
data_files:
- split: test
path: data/test-*