metadata
dataset_info:
features:
- name: id
dtype: int64
- name: source
dtype: string
- name: orig_id
dtype: int64
- name: text
dtype: string
- name: score
dtype: float32
splits:
- name: train
num_bytes: 133284577721
num_examples: 52462533
download_size: 81773254662
dataset_size: 133284577721
configs:
- config_name: default
data_files:
- split: train
path: data/train-*