metadata
dataset_info:
features:
- name: id
dtype: int64
- name: prompt
dtype: string
- name: response
dtype: string
- name: instruction
dtype: string
splits:
- name: train
num_bytes: 12000184
num_examples: 1959
- name: test
num_bytes: 2537564
num_examples: 425
download_size: 2056827
dataset_size: 14537748
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*