metadata
dataset_info:
features:
- name: anchor
dtype: string
- name: positive
dtype: string
splits:
- name: train
num_bytes: 5555
num_examples: 5
download_size: 9911
dataset_size: 5555
configs:
- config_name: default
data_files:
- split: train
path: data/train-*