DavidLanz/chinese-dolly-input-output-15k

---
license: cc-by-sa-3.0
task_categories:
- question-answering
- summarization
- text-generation

language:
- zh
- en

size_categories:
- 10K<n<100K

---
Chinese-Dolly-15k 是繁體中文翻譯的Dolly instruction(Databricks)資料集，並用於 Fine tune 的問答 JSON 格式。

原來的資料集'databricks/databricks-dolly-15k'是由數千名Databricks員工根據InstructGPT論文中概述的幾種行為類別生成的遵循指示記錄的開來源資料集。這幾個行為類別包括頭腦風暴、分類、封閉型問答、生成、資訊擷取、開放類型的問答和摘要。

在知識共用署名-相同方式共用3.0（CC BY-SA 3.0）許可下，此資料集可用於任何學術或商業用途。

如果你也在做這些資料集的籌備，歡迎來聯繫我們，避免重複花錢。

## Citation

Please cite the repo if you use the data or code in this repo.

```
@misc{alpaca,
author = {DavidLanz},
title = {An Instruction-following Chinese Language model, LoRA tuning on LLaMA},
year = {2023},
publisher = {GitHub},
journal = {GitHub repository},
url = {https://www.databricks.com/blog/2023/04/12/dolly-first-open-commercially-viable-instruction-tuned-llm},
urldate = {2023-09-15}
}
```

Chinese-Dolly-15k is a Traditional Chinese translation of the Dolly instruction (Databricks) dataset, used for Fine tune in question-answering JSON format. The original dataset databricks/databricks-dolly-15k was generated by thousands of Databricks employees based on several behavior categories outlined in the InstructGPT paper, including brainstorming, classification, closed-ended question-answering, generation, information extraction, open-ended question-answering, and summarization. This dataset is available under the Creative Commons Attribution-ShareAlike 3.0 (CC BY-SA 3.0) license for any academic or commercial use.

原始信息汇总

数据集概述

数据集名称

• Chinese-Dolly-15k

数据集描述

• 繁體中文翻譯的Dolly instruction(Databricks)資料集，用於Fine tune的問答JSON格式。

原始数据集

• 原始数据集为databricks/databricks-dolly-15k，由数千名Databricks员工生成，遵循InstructGPT论文中概述的几种行为类别，包括头脑风暴、分类、封闭型问答、生成、信息提取、开放类型问答和摘要。

许可

• 在知识共享署名-相同方式共享3.0（CC BY-SA 3.0）许可下，此数据集可用于任何学术或商业用途。

任务类别

• question-answering
• summarization
• text-generation

语言

• zh
• en

数据集大小

• 10K<n<100K

引用

@misc{alpaca, author = {DavidLanz}, title = {An Instruction-following Chinese Language model, LoRA tuning on LLaMA}, year = {2023}, publisher = {GitHub}, journal = {GitHub repository}, url = {https://www.databricks.com/blog/2023/04/12/dolly-first-open-commercially-viable-instruction-tuned-llm}, urldate = {2023-09-15} }