metadata
dataset_info:
features:
- name: size
dtype: int64
- name: timestamp
dtype: float64
splits:
- name: train
num_bytes: 10426935088
num_examples: 651683443
download_size: 6946974067
dataset_size: 10426935088
configs:
- config_name: default
data_files:
- split: train
path: data/train-*