metadata
dataset_info:
features:
- name: instructions
dtype: string
- name: inputs
dtype: string
- name: outputs
dtype: string
splits:
- name: train
num_bytes: 1666560
num_examples: 9516
download_size: 250621
dataset_size: 1666560
configs:
- config_name: default
data_files:
- split: train
path: data/train-*