metadata
dataset_info:
features:
- name: output
dtype: string
- name: input
dtype: string
- name: instruction
dtype: string
splits:
- name: train
num_bytes: 778282.5734157651
num_examples: 1000
- name: evaluation
num_bytes: 15565.651468315302
num_examples: 20
download_size: 489186
dataset_size: 793848.2248840804
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: evaluation
path: data/evaluation-*