Datasets:
metadata
dataset_info:
features:
- name: instruction
dtype: string
- name: context
dtype: string
- name: response
dtype: string
splits:
- name: train
num_bytes: 10483730
num_examples: 15011
download_size: 7492947
dataset_size: 10483730
configs:
- config_name: default
data_files:
- split: train
path: data/train-*