From the course: Advanced Prompt Engineering Techniques
Unlock the full course today
Join today to access over 24,400 courses taught by industry experts.
Chain-of-thought (CoT) prompting
From the course: Advanced Prompt Engineering Techniques
Chain-of-thought (CoT) prompting
- If I gave you a logic puzzle like Miken has five apples and gives chase two of them, how many apples does Miken and have left? You would go through a chain of thoughts to figure out how to solve the puzzle. Now, when we work with AI systems, we can use few shot prompting to provide examples of chains of thought and then have the system match those patterns in its own so-called Reasoning to solve logic puzzles. This is called chain of thought prompting. Let me show you an example of this. So first I'm going to show you what happens if I don't provide any examples. So here we have a logic puzzle. Tracy used a piece of wire four feet long to support tomato plants in her garden. The wire was cut into pieces six inches long. How many pieces did she obtain? So if I run this in ChatGPT, ChatGPT will try to figure this out and it will go through some sort of logic process and it'll probably get the answer right, but not…
Contents
-
-
-
Reference prompting5m 27s
-
(Locked)
Zero-shot and few-shot prompting3m 20s
-
(Locked)
Chain-of-thought (CoT) prompting5m 19s
-
(Locked)
Take a deep breath and work step by step2m 48s
-
(Locked)
Generated knowledge prompting3m 5s
-
(Locked)
Tree-of-thought (ToT) prompting3m 21s
-
(Locked)
Directional stimulus prompting2m 57s
-
(Locked)
Chain-of-density (CoD) prompting5m 40s
-