metadata
dataset_info:
features:
- name: text
dtype: string
- name: language_score
dtype: float64
- name: token_count
dtype: int64
splits:
- name: train
num_bytes: 2750096788
num_examples: 783464
download_size: 1658651314
dataset_size: 2750096788
configs:
- config_name: default
data_files:
- split: train
path: data/train-*