《吴恩达 x OpenAI Prompt课程》【专业翻译，配套代码笔记】01.课程介绍

UP主: 二次元的Datawhale · 时长: 6:30 · 🔗 B站原视频

发布: 2023-05-04 · 收录: 2024-04-18

标签: 吴恩达 · Prompt Engineering · ChatGPT · 大语言模型 · AI开发

课程欢迎与讲师介绍

Andrew： Welcome to this course on ChatGPT Prompt Engineering for Developers. I’m thrilled to have with me Isa Fulford to teach this along with me. She is a member of the technical staff of OpenAI, and has built the popular ChatGPT Retrieval Plugin. A large part of her work has been teaching people how to use LLM (Large Language Model) technology in products and applications. She’s also a contributor to the OpenAI Cookbook that teaches people prompting. I’m thrilled to have you here.

Isa： I’m thrilled to be here and share some prompting best practices with you all.

为什么做这门课：从 Web 界面到开发者 API

Andrew： There’s been a lot of material on the internet for prompting, with articles like “30 prompts everyone has to know”. A lot of that has been focused on the ChatGPT web user interface, which many people are using to do specific and often one-off tasks.

But I think the power of large language models, as a developer using API calls to quickly build software applications, is still very underappreciated. My team at AI Fund, which is a sister company to DeepLearning.AI, has been working with many startups on applying these technologies to many different applications. It’s been exciting to see what APIs can enable developers to do.

课程内容概览：从最佳实践到搭建 Chatbot

Andrew： There’s a lot of material to cover. First, you’ll learn some prompting best practices for software development. Then we’ll cover some common use cases: Summarizing, Inferring, Transforming, Expanding. Then you’ll build a chatbot using an LLM. We hope that this will spark your imagination about new applications that you can build.

两类大模型：Base LLM 与 Instruction-tuned LLM

Andrew： In the development of large language models (LLMs), there have been broadly two types of LLMs, which I’m going to refer to as base LLMs and instruction-tuned LLMs.

A base LLM has been trained to predict the next word based on text training data, often trained on large amounts of data from the internet and other sources to figure out what’s the next most likely word to follow. For example, if you were to prompt this: “Once upon a time there was a unicorn”, it may complete this, that is, it may predict the next several words are: “that lived in a magical forest with all unicorn friends.”

Instruction-tuned LLMs have been trained to follow instructions. So if you were to ask it, “What is the capital of France?”, it’s much more likely to output something like: “The capital of France is Paris.”

Instruction-tuned 的训练方式：微调 + RLHF

Andrew： The way that instruction-tuned LLMs are typically trained is: you start off with a base LLM that’s been trained on a huge amount of text data, and further fine-tune it with inputs and outputs that are instructions and good attempts to follow those instructions, and then often further refine using a technique called RLHF (reinforcement learning from human) to make the system better able to be helpful and follow instructions.

Because instruction-tuned LLMs have been trained to be helpful, honest, and harmless, they are less likely to output problematic text, such as toxic outputs, compared to base LLMs. A lot of the practical usage scenarios have been shifting toward instruction-tuned LLMs.

Some of the best practices you find on the internet may be more suited for a base LLM. In this course, we’ll generally recommend using instruction-tuned LLMs.

致谢：OpenAI 与 DeepLearning.AI 团队

Andrew： I’m grateful to the team from OpenAI and DeepLearning.AI that contributed to the materials that Isa and I will be presenting.

I’m very grateful to Andrew Mayne, Joe Palermo, Boris Power, Tad Sanders, and Lillian Wang from OpenAI, who were very involved with us brainstorming materials and vetting the materials to put together the curriculum for this short course.

And I’m also grateful on the DeepLearning.AI side for the work of Jeff Ladwick, Eddie Xu, and Tommy Nelson.

提示工程两条关键原则：清晰具体 + 给模型时间思考

Andrew： When you use an instruction-tuned LLM, think of giving instructions to another person—someone that’s smart, but doesn’t know the specifics of your task.

When an LLM doesn’t work sometimes, it’s because the instructions weren’t clear enough. For example, if you were to say, “Please write me something about Alan Turing.” It can be helpful to be clear about whether you want the text to focus on his scientific work, his personal life, his role in history, or something else. And if you specify what you want the tone of the text to be—should it take on the tone like a professional journalist would write?

You’ll learn how to be clear and specific, which is an important principle of prompting LLMs. And you’ll also learn a second principle of prompting: giving the LLM time to think.

结尾

Andrew： With that, let’s go on to the next video.