metadata
dataset_info:
features:
- name: 'Unnamed: 0'
dtype: int64
- name: query
dtype: string
- name: positive
dtype: string
- name: label
dtype: string
- name: negatives
sequence: string
splits:
- name: train
num_bytes: 425525445
num_examples: 100000
download_size: 227205105
dataset_size: 425525445
configs:
- config_name: default
data_files:
- split: train
path: data/train-*