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: 46697335
num_examples: 72198
- name: test
num_bytes: 2758929
num_examples: 3799
download_size: 16895191
dataset_size: 49456264
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*