metadata
dataset_info:
features:
- name: instruction
dtype: string
- name: response
dtype: string
- name: text
dtype: string
splits:
- name: train
num_bytes: 188088535
num_examples: 66826
- name: validation
num_bytes: 23713219
num_examples: 9535
download_size: 27382625
dataset_size: 211801754
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*