metadata
dataset_info:
features:
- name: 'Unnamed: 0'
dtype: int64
- name: query
dtype: string
- name: positive
dtype: string
- name: negative
dtype: string
- name: dataset
dtype: string
- name: pos_id
dtype: int64
- name: neg_id
dtype: int64
splits:
- name: train
num_bytes: 2386995256.4969964
num_examples: 2370851
download_size: 2058380738
dataset_size: 2386995256.4969964
configs:
- config_name: default
data_files:
- split: train
path: data/train-*