metadata
license: mit
dataset_info:
features:
- name: query
dtype: string
- name: pos
sequence: string
- name: neg
sequence: 'null'
- name: relevance
dtype: float64
splits:
- name: train
num_bytes: 89618293101
num_examples: 122146785
download_size: 60471930866
dataset_size: 89618293101
configs:
- config_name: default
data_files:
- split: train
path: reddit/train-*