metadata
license: mit
dataset_info:
features:
- name: id
dtype: int64
- name: inner_id
dtype: int64
- name: vectors
list: float64
- name: sparse
list:
- name: indices
list: int64
- name: values
list: float64
splits:
- name: training
num_bytes: 406218384
num_examples: 45031
download_size: 244227992
dataset_size: 406218384
configs:
- config_name: default
data_files:
- split: training
path: data/training-*
A dataset that represents a vectorized version of all the scraped textual content. The encoding of the content is utilized by a multi-linguage embedding model.