metadata
dataset_info:
features:
- name: labels
dtype: int64
- name: correct
dtype: int64
- name: features
sequence: float32
splits:
- name: train
num_bytes: 47823964
num_examples: 15467
download_size: 57522356
dataset_size: 47823964
configs:
- config_name: default
data_files:
- split: train
path: data/train-*