metadata
dataset_info:
features:
- name: system
dtype: string
- name: keyword
dtype: string
- name: prompt
dtype: string
- name: response
struct:
- name: content
dtype: string
- name: task
dtype: string
- name: task
dtype: string
- name: response_cv2str
dtype: string
- name: text
dtype: string
splits:
- name: train
num_bytes: 26683439
num_examples: 3240
- name: test
num_bytes: 2943907
num_examples: 360
download_size: 9184319
dataset_size: 29627346
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*