metadata
dataset_info:
features:
- name: title
dtype: string
- name: text
dtype: string
- name: wiki_id
dtype: int32
- name: views
dtype: float32
- name: paragraph_id
dtype: int32
- name: langs
dtype: int32
- name: emb
sequence: float32
splits:
- name: train
num_bytes: 1160521080
num_examples: 250000
download_size: 1258146076
dataset_size: 1160521080
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
language:
- fr
250K most consulted paragraphs of the French part of Wikipedia embeded with Solon-embeddings-large-0.1.