News for hackers. Stuff that matters.
Re:Claude 3.5 Sonnet by muzani on Wednesday June 26, 2024 @11:55PM (#40807035)
me: "is 400000 the same as 4 * 100 * 100"

gpt-4o: "Yes, 400,000 is the same as 4 X 100 × 1000."

gpt-4:

    Yes, 400,000 is the same as 4 X 100 × 100.
    Here's the breakdown:
    4 × 100 = 400
    400 × 100 = 40,000
    Oops, it looks like there was an error in my previous calculation. 
    So, 4 × 100 X 100 actually equals 40,000, not 400,000. To make it 400,000, 
    you would need to use 4 × 10 x 10,000 or adjust the numbers similarly.
claude-3.5:

    No, 400000 is not the same as 4 * 100 * 100.
    Let's break it down:
    4 * 100 * 100 = 4 * 10000 = 40000
    400000 is ten times larger than 40000.
Everyone keeps saying gpt-4o beats benchmarks and stuff, but this is consistently my experience with it. The benchmarks fall far off my every day experience.
Re:Claude 3.5 Sonnet by Davidzheng on Thursday June 27, 2024 @01:11AM (#40807461)
The thing is, gpt4(o) is the only model I've talked to that makes me feel like it understands calculus. Not like it's pretending and succeeding
Re:Claude 3.5 Sonnet by muzani on Thursday June 27, 2024 @04:48AM (#40808607)
Ah, I think it's just much better trained. At double the speed and half the price, I'd assume there's less computing power behind it. But likely the training covers some topics really well.
Re:Claude 3.5 Sonnet by Jverse on Thursday June 27, 2024 @06:36AM (#40809058)
My answer from gpt4o was correct and similar to your example from claude. Maybe you were just unlucky?
Re:Claude 3.5 Sonnet by muzani on Friday June 28, 2024 @01:09AM (#40817918)
This kind of thing happens so frequently that I don't think so. My guess is the language auto detection messes it up for multilinguals and it works fine for people who are solely using English? Because it's certainly broken on voice input - I ask it a question in English and it will respond in Russian and such.