metadata
dataset_info:
features:
- name: id
dtype: string
- name: lang
dtype: string
- name: instruction
dtype: string
- name: response
dtype: string
splits:
- name: test
num_bytes: 322616872
num_examples: 96744
- name: train
num_bytes: 1917633328
num_examples: 652511
- name: validation
num_bytes: 306880892
num_examples: 94758
download_size: 2121109391
dataset_size: 2547131092
configs:
- config_name: default
data_files:
- split: test
path: data/test-*
- split: train
path: data/train-*
- split: validation
path: data/validation-*