airoboros-gpt4-1.3

## Overview

A continuation of [gpt4-1.2](https://huggingface.co/datasets/jondurbin/airoboros-gpt4-1.2), with:
* all coding instructions now have an equivalent "PLAINFORMAT" version
* several thousand new orca style prompts, this time with reasoning first, then response
* several examples of conversational/character interactions, with asterisk'd actions and quoted dialog

_*Note: I did not filter by token length for this dataset, some are well over 2048 so use carefully.*_

### Usage and License Notices

All airoboros models and datasets are intended and licensed for research use only. I've used the 'cc-nc-4.0' license, but really it is subject to a custom/special license because:

- the base model is LLaMa, which has it's own special research license
- the dataset(s) were generated with OpenAI (gpt-4 and/or gpt-3.5-turbo), which has a clausing saying the data can't be used to create models to compete with openai

So, to reiterate: this model (and datasets) cannot be used commercially.

概述
本数据集是[gpt4-1.2](https://huggingface.co/datasets/jondurbin/airoboros-gpt4-1.2)的延续版本，具体更新如下：
* 所有代码类指令现已提供对应的"PLAINFORMAT"格式版本
* 新增数千条Orca风格提示词，本次采用先呈现推理过程、再输出回复的结构
* 新增多组对话/角色交互示例，包含以星号标注的动作内容与带引号的对话文本

*注意：本数据集未按Token长度进行筛选，部分样本长度远超2048，使用时请谨慎。

## 使用与许可声明
所有Airoboros模型与数据集仅用于研究用途，并受对应许可协议约束。本次采用的是CC-NC-4.0许可，但实际上该数据集受定制专属许可约束，原因如下：
- 基础模型为LLaMa，其本身带有专属研究许可
- 本数据集由OpenAI的GPT-4及/或GPT-3.5-turbo生成，OpenAI相关条款规定，该数据集不得用于开发可与OpenAI竞争的模型

因此再次强调：本模型（及相关数据集）不得用于商业用途。