metadata
dataset_info:
features:
- name: instructions
dtype: string
- name: inputs
dtype: string
- name: outputs
dtype: string
splits:
- name: train
num_bytes: 1388074200
num_examples: 599093
download_size: 697907901
dataset_size: 1388074200
configs:
- config_name: default
data_files:
- split: train
path: data/train-*