metadata
dataset_info:
features:
- name: anchor
dtype: string
- name: positive
dtype: string
- name: negative_1
dtype: string
- name: negative_2
dtype: string
- name: negative_3
dtype: string
- name: negative_4
dtype: string
- name: negative_5
dtype: string
splits:
- name: train
num_bytes: 105776614
num_examples: 5813
- name: validation
num_bytes: 26454458
num_examples: 1454
download_size: 70522965
dataset_size: 132231072
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*