metadata
dataset_info:
features:
- name: anchor
dtype: string
- name: logical
dtype: string
- name: illogical
dtype: string
splits:
- name: train
num_bytes: 973741
num_examples: 2000
download_size: 439695
dataset_size: 973741
configs:
- config_name: default
data_files:
- split: train
path: data/train-*