metadata
dataset_info:
features:
- name: text
dtype: string
- name: step_labels
sequence: string
splits:
- name: train
num_bytes: 3840
num_examples: 5
download_size: 8290
dataset_size: 3840
configs:
- config_name: default
data_files:
- split: train
path: data/train-*