metadata
dataset_info:
features:
- name: id
dtype: string
- name: system_prompt
dtype: string
- name: input
dtype: string
- name: response
dtype: string
- name: category
dtype: string
- name: language
dtype: string
- name: source
dtype: string
splits:
- name: train
num_bytes: 36573010
num_examples: 21391
download_size: 20925273
dataset_size: 36573010
configs:
- config_name: default
data_files:
- split: train
path: data/train-*