metadata
dataset_info:
features:
- name: instruction
dtype: string
- name: response
dtype: string
- name: prompt_label
dtype: string
- name: response_label
dtype: string
splits:
- name: train
num_bytes: 299144592
num_examples: 33540
download_size: 84847399
dataset_size: 299144592
configs:
- config_name: default
data_files:
- split: train
path: data/train-*