metadata
dataset_info:
features:
- name: text
dtype: string
- name: step_labels
sequence: string
splits:
- name: train
num_bytes: 40086
num_examples: 50
download_size: 27850
dataset_size: 40086
configs:
- config_name: default
data_files:
- split: train
path: data/train-*