metadata
dataset_info:
features:
- name: prompt
dtype: string
- name: target
dtype: string
- name: prompt_type
dtype: string
splits:
- name: train
num_bytes: 904334880
num_examples: 656936
- name: test
num_bytes: 1461136
num_examples: 1000
download_size: 488977188
dataset_size: 905796016
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*