metadata
dataset_info:
features:
- name: prompt
dtype: string
- name: answer
dtype: string
splits:
- name: train
num_bytes: 2475
num_examples: 99
- name: valid
num_bytes: 50
num_examples: 2
- name: test
num_bytes: 50
num_examples: 2
download_size: 3197
dataset_size: 2575
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: valid
path: data/valid-*
- split: test
path: data/test-*