metadata
dataset_info:
features:
- name: chunk
dtype: string
- name: url
dtype: string
- name: embedding
sequence: float32
splits:
- name: train
num_bytes: 2337565832
num_examples: 1209287
download_size: 2561219117
dataset_size: 2337565832
configs:
- config_name: default
data_files:
- split: train
path: data/train-*