metadata
dataset_info:
features:
- name: instruction
dtype: string
- name: input
dtype: string
- name: id
dtype: string
- name: output
dtype: string
splits:
- name: train
num_bytes: 140797599
num_examples: 67017
download_size: 50922704
dataset_size: 140797599
configs:
- config_name: default
data_files:
- split: train
path: data/train-*