metadata
dataset_info:
features:
- name: act
dtype: string
- name: prompt
dtype: string
splits:
- name: train
num_bytes: 10223
num_examples: 170
download_size: 9017
dataset_size: 10223
configs:
- config_name: default
data_files:
- split: train
path: data/train-*