metadata
dataset_info:
features:
- name: instruction
dtype: string
- name: context
dtype: string
- name: response
dtype: string
splits:
- name: train
num_bytes: 20148292.53244219
num_examples: 6559
- name: validation
num_bytes: 5037841.09669236
num_examples: 1640
download_size: 15361995
dataset_size: 25186133.62913455
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*