From the course: Introduction to Large Language Models (LLMs) and Prompt Engineering by Pearson
Join today to access over 25,200 courses taught by industry experts.
Chain-of-thought prompting
From the course: Introduction to Large Language Models (LLMs) and Prompt Engineering by Pearson
Chain-of-thought prompting
Moving on to chain-of-thought prompting. Now, this isn't actually the first time we've talked about chain-of-thought prompting. We actually brought it up in an earlier section when we were looking at our RAG and our agent prompt a few lessons ago. But to define the concept a bit more formally, chain-of-thought prompting forces an LLM to generate a reasoning for an answer alongside the actual answer itself. And the goal of TANF on prompting, like most prompting techniques to be fair, is to lead to a more actionable and frankly just better, more accurate result. So to see this in action, we'll take a bit of a history lesson because this is a model that you can no longer actually use on OpenAI's Playground, but because I took the screenshot over a year ago, we can at least see the differences between then And now. So we're looking at a version of GPT-3 called DaVinci. There are similar models to this that you can use that OpenAI claim are just a little bit better. But just to show you…
Download courses and learn on the go
Watch courses on your mobile device without an internet connection. Download courses using your iOS or Android LinkedIn Learning app.