metadata
dataset_info:
features:
- name: input
dtype: string
- name: target
dtype: string
- name: source_config
dtype: string
- name: model_response
dtype: string
splits:
- name: train
num_bytes: 233103
num_examples: 500
download_size: 86444
dataset_size: 233103
configs:
- config_name: default
data_files:
- split: train
path: data/train-*