metadata
dataset_info:
features:
- name: title_s
dtype: string
- name: title_dl
dtype: string
- name: source_url
dtype: string
- name: authors
list: string
- name: snippet_s
dtype: string
- name: text
dtype: string
- name: date
dtype: timestamp[ns]
- name: publish_date_dl
dtype: string
- name: url
dtype: string
- name: matches
list:
- name: date
dtype: string
- name: position
dtype: int64
- name: query
dtype: string
- name: embeddings
list: float64
splits:
- name: train
num_bytes: 17374667
num_examples: 1000
download_size: 10264011
dataset_size: 17374667
configs:
- config_name: default
data_files:
- split: train
path: data/train-*