p1esk

19 ms·

1. 361.▲
   by p1esk 2y ago

   I personally cannot “figure out quaternions as a solution to gimbal locking or the theory of relativity”. I’m just not as smart as Einstein. Does it mean I’m not capable of reasoning? Because it seems that’s what you are implying. If you tr
2. 362.▲
   by p1esk 2y ago

   Are you saying “reasoning” means making scientific breakthroughs requiring genius level human intelligence? Something that 99.9999% of humans are not smart enough to do, right?
3. 363.▲
   by p1esk 2y ago

   Ok, so how about an example?
4. 364.▲
   by p1esk 2y ago

   Can you give an example of the reasoning you’re talking about?
5. 365.▲
   by p1esk 2y ago

   I’m curious if Microsoft lets OpenAI train on GH private repos.
6. 366.▲
   by p1esk 2y ago

   So, which new profession are you learning?
7. 367.▲
   by p1esk 2y ago

   “Just do better” is the key metric here. 99% of all human jobs are “finding something relevant with already solved problems”.
8. 368.▲
   by p1esk 2y ago

   fairly clear what they can and can't do It’s not at all clear what the next gen models will do (e.g. gpt5). Might be enough to trigger mass unemployment. Or not.
9. 369.▲
   by p1esk 2y ago

   I just traded my 13 mini for 16 Pro Max. The reason is simple: I want a bigger screen because I read a lot on my phone.
10. 370.▲
    by p1esk 2y ago

    If a student can use a smart friend to effectively function as someone with a higher IQ, then why should that not be allowed?
11. 371.▲
    by p1esk 2y ago

    4x seq length expansion doesn’t sound that bad.
12. 372.▲
    by p1esk 2y ago

    Let’s say we want to use 6-grams and build an embedding vector for the word “because”: we add integer vectors for “becaus” and “ecause”, right? For example: [1,2,3,4,5,6] + [2,3,4,5,6,2] = [3,5,7,9,11,8]. Obviously we cannot use this result
13. 373.▲
    by p1esk 2y ago

    The other 500 values in the skew/askew vectors will be similar though. The 12 character values don’t need to be aligned, their function is to provide spelling. Adding such info will probably help LLM answer questions requiring characte
14. 374.▲
    by p1esk 2y ago

    IIRC, overlapping ngram vectors are summed to form the token embedding - doesn’t it effectively destroy any character level representation of the token? Doesn’t really make sense to me.
15. 375.▲
    by p1esk 2y ago

    Yes
16. 376.▲
    by p1esk 2y ago

    Has anyone tried to combine a token embedding with some representation of the characters in the (sub)word? For example, use a 512 long vector to represent a token, and reserve the last 12 values to spell out the word.
17. 377.▲
    by p1esk 2y ago

    What exactly is left when we remove coding from software engineering? Could it be handled by a manager? Or perhaps by a single senior SWE who could now perform the work of an entire team using these rapidly advancing AI coders?
18. 378.▲
    by p1esk 2y ago

    It sucks that we’re still at “best practices” phase. We’ve been in this phase for the last three decades [1], and I really hope we enter “good theory” phase soon. [1] https://cseweb.ucsd.edu/classes/wi08/cse253&#
19. 379.▲
    by p1esk 2y ago

    Nvidia is a software company that also happens to make decent hardware. People buy their hardware because of their software. And it’s not just CUDA. Nvidia is building a whole bunch of potentially groundbreaking software products. Listen to
20. 380.▲
    by p1esk 2y ago

    Try to get an internship.
21. 381.▲
    by p1esk 2y ago

    This is a strange comment. Majority of SWEs in FAANGs are Asian: https://www.statista.com/statistics/311815/google-employee-e... most of the software you like was probably written by Asians.
22. 382.▲
    by p1esk 2y ago

    Do you use them? How do they compare to professional grade hearing aids?
23. 383.▲
    by p1esk 2y ago

    If I understand correctly the hearing aid feature is not available yet.
24. 384.▲
    by p1esk 2y ago

    The Cold War was just one of many factors leading to the Soviet Union collapse. It was a lot more complex than “stopped catching up to the West”. Source: I lived through it.
25. 385.▲
    by p1esk 2y ago

    The Soviet Union lost the Cold War long after Brezhnev.
26. 386.▲
    by p1esk 2y ago

    Anduril ad?
27. 387.▲
    by p1esk 2y ago

    LLMs are too expensive for this
28. 388.▲
    by p1esk 2y ago

    8xA6000
29. 389.▲
    by p1esk 2y ago

    I provided my opinion based on what they wrote. I did not intend to generalize it. I’m not sure the OP fully understands the serious risks of alcohol consumption. I have observed enough lives damaged by alcoholism to speak from experience.
30. 390.▲
    by p1esk 2y ago

    How long have you been having a single drink a day? If it’s been a year or longer - you’re unique in not forming an addiction. If not - you’re playing with fire.