metadata
dataset_info:
features:
- name: CATEGORY
dtype: string
- name: TITLE
dtype: string
- name: CONTEXT
dtype: string
- name: INSTRUCTION
dtype: string
- name: RESPONSE
dtype: string
splits:
- name: train
num_bytes: 339949
num_examples: 684
download_size: 14098
dataset_size: 339949
configs:
- config_name: default
data_files:
- split: train
path: data/train-*