metadata
dataset_info:
features:
- name: _id
dtype: string
- name: text
dtype: string
- name: query-id
dtype: string
- name: corpus-id
dtype: string
- name: score
dtype: float64
- name: embedding_ada
list: float64
- name: embedding_small
list: float64
- name: embedding_large
list: float64
splits:
- name: train
num_bytes: 53703090
num_examples: 1406
download_size: 40602435
dataset_size: 53703090
configs:
- config_name: default
data_files:
- split: train
path: data/train-*