metadata
dataset_info:
features:
- name: input
sequence: string
- name: output
sequence: string
- name: instructions
dtype: string
splits:
- name: train
num_bytes: 4349314
num_examples: 3068
- name: validation
num_bytes: 557068
num_examples: 384
- name: test
num_bytes: 556717
num_examples: 384
download_size: 471960
dataset_size: 5463099
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*