metadata
dataset_info:
features:
- name: query
dtype: string
- name: positives
list:
- name: author
dtype: string
- name: date
dtype: string
- name: html
dtype: string
- name: text
dtype: string
- name: title
dtype: string
- name: url
dtype: string
- name: negatives
list:
- name: author
dtype: string
- name: date
dtype: string
- name: html
dtype: string
- name: text
dtype: string
- name: title
dtype: string
- name: url
dtype: string
- name: metadata
struct:
- name: ability
dtype: string
- name: dataset_type
dtype: string
- name: format
dtype: string
- name: tags
sequence: string
- name: topic
dtype: string
splits:
- name: train
num_bytes: 226667672
num_examples: 496
download_size: 82547504
dataset_size: 226667672
configs:
- config_name: default
data_files:
- split: train
path: data/train-*