andy_ppp

9 ms·

1. 151.▲
   by andy_ppp 5mo ago

   Fine tuning these models (at least with PPO or equivalent) requires even more VRAM than inference does, potentially 2-3 times more.
2. 152.▲
   by andy_ppp 5mo ago

   It isn’t really slop anymore and it will keep improving.
3. 153.▲
   by andy_ppp 5mo ago

   Great that this solved everyone’s problems isn’t it
4. 154.▲
   by andy_ppp 5mo ago

   This is completely why the rich love it so much
5. 155.▲
   by andy_ppp 5mo ago

   Honestly, I think the need for devs is total copium, the progress made in two years is astounding and in two years time they will be better at programming than 99% of programmers. It’s incredible what they can do now. No it’s not perfect bu
6. 156.▲
   by andy_ppp 5mo ago

   "Difficult decision" says billionaire sacking people, many of whom have families, so he can make even more money.
7. 157.▲
   by andy_ppp 5mo ago

   Frontend has plenty of security considerations.
8. 158.▲
   by andy_ppp 5mo ago

   It'll probably update to that without telling you at some point.
9. 159.▲
   by andy_ppp 5mo ago

   Controversial I know, but EU regulations are largely reasonable and mostly come down to good practices. For example I was part of a team building a crewing application for container ships and largely I agree with almost all of the conclusio
10. 160.▲
    by andy_ppp 5mo ago

    They want to lie about how they found the soldier, and potentially have China spend a few billion trying to copy technology that can never work.
11. 161.▲
    by andy_ppp 5mo ago

    Just, in this case means “at minimum” or “first and foremost, no excuses”. I obviously understand this is a huge undertaking. Nobody said attempting to be competitive with NVIDIA in AI would be a walk in the park.
12. 162.▲
    by andy_ppp 5mo ago

    Please just get everything in PyTorch to work, and work well (and across all graphics cards too). This is the starting point and it doesn't matter how you do it. But the fact you cannot even do some very basic stuff on AMD is going to
13. 163.▲
    by andy_ppp 5mo ago

    Not really - I imagine as with almost everything in life there's a normal distribution, in this case of the quality with which people use AI tools.
14. 164.▲
    by andy_ppp 5mo ago

    I think you mean GIF .
15. 165.▲
    by andy_ppp 5mo ago

    This is a terrible take, he actually helped me get a domain back that I'd accidentally let expire. He's a genuine person just making a business with domains, everything everyone does has some kind of compromise to it if you look c
16. 166.▲
    by andy_ppp 5mo ago

    If Friendster.com was making around $9,000 per year, this would explain why paying $30k + domains returning a similar amount would make sense?
17. 167.▲
    by andy_ppp 5mo ago

    You can buy them in the UK soon, just £450 and I suspect they'll disintegrate quickly... https://www.adidas.co.uk/adizero-adios-pro-evo-3-shoes/KH767...
18. 168.▲
    by andy_ppp 5mo ago

    I don't think what you have said represents William Burns at all: https://www.youtube.com/watch?v=iP8BfC1e0Ug
19. 169.▲
    by andy_ppp 5mo ago

    I think this is the wrong place to debate politics tbh, better luck next time.
20. 170.▲
    by andy_ppp 5mo ago

    You seem to be extremely fond of Russian propaganda.
21. 171.▲
    by andy_ppp 5mo ago

    If China cannot get oil from the middle east what happens to China and China-Russia relations? I didn't say there would be hostilities just Russia would become potentially the more dominant partner. If NATO expansion is the reason for
22. 172.▲
    by andy_ppp 5mo ago

    Have you considered meditation? ;-)
23. 173.▲
    by andy_ppp 5mo ago

    What do you think the war in the gulf is about, the US cannot compete with China so they are destroying the global system that enabled them. There is no plan to have a peace with Iran, only perpetual war and the destruction of the middle ea
24. 174.▲
    by andy_ppp 6mo ago

    Yes, I’ve always thought of AI companies as sentimental. They will definitely do this :-/
25. 175.▲
    by andy_ppp 6mo ago

    So I won’t use stuff by Elon Musk, what is the next best alternative please
26. 176.▲
    by andy_ppp 6mo ago

    I see it completely the opposite way, you use an LLM and correct all its mistakes and it allows you to deliver a rough solution very quickly and then refine it in combination with the AI but it still gets completely lost and stuck on basic
27. 177.▲
    by andy_ppp 6mo ago

    I think it’s clear that Anthropic are leading the industry right now and I believe it’s because they are better at using their own tools to develop software than anyone else. I suspect that trend will accelerate.
28. 178.▲
    by andy_ppp 6mo ago

    Do we know if other models have started detecting and poisoning training/fine tuning that these Chinese models seem to use for alignment, I’d certainly be doing some naughty stuff to keep my moat if I was Anthropic or OpenAI…
29. 179.▲
    by andy_ppp 6mo ago

    I wonder if you could just setup a small model and suggest a load of things and try every file and it might still end up being cheaper and just as good as Mythos at a specific task. Maybe this will be something that holds true for more thin
30. 180.▲
    by andy_ppp 6mo ago

    Fair. Clearly they need the land back GOD gave them 3000 years ago.