metadata
dataset_info:
features:
- name: prompt
dtype: string
- name: answer
dtype: float64
splits:
- name: train
num_bytes: 302
num_examples: 4
download_size: 1514
dataset_size: 302
configs:
- config_name: default
data_files:
- split: train
path: data/train-*