metadata
dataset_info:
features:
- name: id
dtype: int32
- name: instruction
list:
- name: content
dtype: string
- name: role
dtype: string
splits:
- name: train
num_bytes: 1112312
num_examples: 1210
- name: test
num_bytes: 61326
num_examples: 68
download_size: 321773
dataset_size: 1173638
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*