metadata
dataset_info:
features:
- name: input
dtype: string
- name: output
dtype: int64
- name: instruction
dtype: string
splits:
- name: train
num_bytes: 118902
num_examples: 500
- name: validation
num_bytes: 142187
num_examples: 500
download_size: 132225
dataset_size: 261089
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*