metadata
dataset_info:
features:
- name: prompt
dtype: string
- name: data_source
dtype: string
- name: ability
dtype: string
- name: instance
dtype: string
splits:
- name: train
num_bytes: 1557159
num_examples: 82
- name: validation
num_bytes: 286723
num_examples: 23
download_size: 556277
dataset_size: 1843882
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*