metadata
dataset_info:
features:
- name: prompt
dtype: string
- name: model_response
dtype: string
- name: generation_index
dtype: int64
splits:
- name: train
num_bytes: 694634
num_examples: 32
download_size: 688701
dataset_size: 694634
configs:
- config_name: default
data_files:
- split: train
path: data/train-*