In the rapidly evolving world of artificial intelligence, two types of AI have captured our imagination: large language models (LLMs) that generate text, and AI image generators that create visual art from descriptions. While these technologies might seem similar on the surface, they actually work in fundamentally different ways. Let’s break down how they differ and how each processes information.
Continue reading “Understanding AI: How Image Generators Differ From Language Models”Guide to Prompt Engineering
Welcome to The Complete Guide to Prompt Engineering. This comprehensive book will take you on a journey from the fundamentals to advanced applications of prompt engineering—the art and science of effectively communicating with artificial intelligence systems. I asked Chatgpt to write this book so I could learn about prompt engineering. I was surprised at how good a job it did. I would be interested in what you think.
Start reading the book at The Complete Guide to Prompt Engineering
Unlock Your Productivity: How to Use Large Language Models (LLMs) for Everyday Tasks
In our increasingly digital world, information is abundant. However, finding efficient ways to process it can be challenging. Utilizing this information effectively is also a challenge. Large language models (LLMs), such as GPT-3, are a subset of artificial intelligence. They are designed to understand human-like text. These models also generate text based on the input they’re given. LLMs are built on advanced statistical patterns derived from vast datasets. They can perform a variety of language-related tasks. This makes them incredibly versatile tools for enhancing productivity.
Continue reading “Unlock Your Productivity: How to Use Large Language Models (LLMs) for Everyday Tasks”What Is a “Few-Shot Prompt”? (And Why It Matters for AI)
If you’ve been exploring the world of AI or language models like ChatGPT, you might’ve come across terms like “few-shot prompt”, “zero-shot learning”, or “prompt engineering.” These phrases may sound technical, but don’t worry—they’re easier to understand than you think!
Continue reading “What Is a “Few-Shot Prompt”? (And Why It Matters for AI)”The Color Coincidence Card Trick
This is based on the Gilbreath Principle. Here’s the complete routine from beginning to end:
Continue reading “The Color Coincidence Card Trick”Card Divination
Performance and Patter
Setup: Ask the spectator to shuffle a deck of cards and remove any 10 cards.
Begin: “Today I’m going to demonstrate a little piece of card magic that works with the power of numbers. Before we begin, I want you to look at these 10 cards you’ve selected and think of any number between 1 and 10. Don’t tell me what it is.”
Continue reading “Card Divination”Jon Racherbaumer’s Freebies
Jon Racherbaumer has passed and his website is no longer live (at this time). I used the wayback machine to find the tricks he published as “Freebies” on his web site.
Continue reading “Jon Racherbaumer’s Freebies”Yet another overhand stack
I was playing around with some of my favorite overhand stacks. I realized that I can do an easy stack. It started with two aces on the bottom of the deck and two aces on the top.
Continue reading “Yet another overhand stack”The Illusion of Understanding: How LLMs Solve Problems and What It Means for AI Intelligence
An article written by Claude.ai on how LLMs solve problems.
When asked how I solved the equation 32 + 112 = 144, I provided a neat, step-by-step mathematical explanation:
- Line up the numbers by place value
- Add the digits in the ones place: 2 + 2 = 4
- Add the digits in the tens place: 3 + 1 = 4
- Add the digits in the hundreds place: 0 + 1 = 1
- Therefore, x = 144
It looked like I was demonstrating mathematical understanding. But was I being truthful about my process? Not exactly.
Continue reading “The Illusion of Understanding: How LLMs Solve Problems and What It Means for AI Intelligence”Predicting the Reversed Cards
Effect:
The performer makes a prediction and writes a number on a piece of paper, which is given to an audience member for safekeeping. A deck of cards is shuffled, and a volunteer assists in the experiment. After some seemingly random card reversals, the number of reversed cards in the deck matches the prediction exactly. This is a Bob Hummer effect and is self working.
Continue reading “Predicting the Reversed Cards”