LLMs generate the answer one token at a time and the second token isn't known as long as the first token wasn't generated. Knowing this is crucial for understanding why LLMs generate nonsensical answers trying to explain unexplainable. The below screenshot illustrates the problem perfectly. The user asks whether a number prime or not. The LLM generates the first word, which is "No" in this case. That's it. There's no way back. It will continue generating a logical explanation of a wrong answer, which is impossible. The chance of generating a wrong first token is never zero, so the situations like this are inevitable. And there's nothing to laugh at if you understands how LLMs work.

The history of @Photoshop icons: Which one did you start with, and which is your favorite?

X is officially valued at $19 billion, employees were notified today. The company is granting them shares in the company at that valuation, which is a 55 percent drop from Musk’s $44 billion purchase price theverge.com/2023/10/30/2393…

TONIGHT: A very special ✨impressions✨ episode of Make Some Noise with three of my all-time favorite mimics. Get ready for the magic of “Two Liam Neesons Kidnap Each Other’s Daughters.”

petition for friends to do 5 min rapid fire slide deck updates on all the happenings in their lives and what they learned

Donald Trump & his family have argued all week in his fraud trial that Mar-a-Lago is actually worth a billion dollars, even though its taxable value is listed as $18 million. So I wrote to the Palm Beach County Property Appraiser that it’s time for Trump to pay more in taxes.

Americans *genuinely* believe they have better food than France. They really believe it. It’s truly extraordinary.