metadata
dataset_info:
features:
- name: 'Unnamed: 0'
dtype: int64
- name: anchor
dtype: string
- name: positive
dtype: string
- name: negative
dtype: string
splits:
- name: train
num_bytes: 148004
num_examples: 1332
download_size: 14594
dataset_size: 148004
configs:
- config_name: default
data_files:
- split: train
path: data/train-*