metadata
dataset_info:
features:
- name: clean
dtype: string
- name: erroneous
dtype: string
- name: model
dtype: string
splits:
- name: test
num_bytes: 96910
num_examples: 100
download_size: 48009
dataset_size: 96910
configs:
- config_name: default
data_files:
- split: test
path: data/test-*