metadata
language:
- ja
dataset_info:
features:
- name: text
dtype: string
- name: features
dtype: float64
splits:
- name: train
num_bytes: 5783063135
num_examples: 12772871
download_size: 3507421243
dataset_size: 5783063135
configs:
- config_name: default
data_files:
- split: train
path: data/train-*