|
--- |
|
dataset_info: |
|
features: |
|
- name: context |
|
dtype: string |
|
- name: question |
|
dtype: string |
|
- name: answer |
|
dtype: string |
|
splits: |
|
- name: train |
|
num_bytes: 19552618 |
|
num_examples: 14755 |
|
download_size: 9802932 |
|
dataset_size: 19552618 |
|
configs: |
|
- config_name: default |
|
data_files: |
|
- split: train |
|
path: data/train-* |
|
license: cc-by-nc-4.0 |
|
language: |
|
- mk |
|
--- |
|
|
|
This is a machine translated version of Databricks's Dolly dataset https://huggingface.co/datasets/databricks/databricks-dolly-15k |
|
|
|
If you use this dataset in your work please cite the original dataset. |
|
``` |
|
@online{DatabricksBlog2023DollyV2, |
|
author = {Mike Conover and Matt Hayes and Ankit Mathur and Jianwei Xie and Jun Wan and Sam Shah and Ali Ghodsi and Patrick Wendell and Matei Zaharia and Reynold Xin}, |
|
title = {Free Dolly: Introducing the World's First Truly Open Instruction-Tuned LLM}, |
|
year = {2023}, |
|
url = {https://www.databricks.com/blog/2023/04/12/dolly-first-open-commercially-viable-instruction-tuned-llm}, |
|
urldate = {2023-06-30} |
|
} |
|
``` |
