metadata
dataset_info:
features:
- name: instructions
dtype: string
- name: input
sequence: string
- name: output
sequence: string
splits:
- name: train
num_bytes: 508992874
num_examples: 463634
- name: validation
num_bytes: 2649412
num_examples: 2440
- name: test
num_bytes: 1081794
num_examples: 1093
download_size: 43641764
dataset_size: 512724080
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*