Live data from Hacker News

Viewing profile — vladf

vladf

HN member
Joined
Tue, May 16, 2017, 9:50 PM UTC
HN karma
424
Public activity
157 items

About vladf

vladfeinberg.com

Recent public activity

  1. comment
    Comment #47256769

    That still looks like a “converge faster” paper. https://arxiv.org/abs/2006.10732 The above provides a nuanced theoretical view. GD inductive bias is probably better unless your mo…

  2. comment
    Comment #44115702

    This is available for Flash

  3. comment
  4. comment
    Comment #43477143

    That's pretty disappointing, it has been out for a while, and we still get top comments like ( https://news.ycombinator.com/item?id=43475043 ) where people clearly think native ima…

  5. comment
    Comment #42926110

    Yes perhaps one day cities like Tokyo will catch up

  6. comment
    Comment #40094004

    Ah! Finally a real world use for egg drop (with eggs and floors both equal to num commits since init, but maybe fewer eggs for those less patient).

  7. comment
    Comment #39894738

    > You still have a group-size of 64 in 4-bit fyi. Results may vary :) > Again, and I keep repeating this but it seems to be ignored every time: this is experimental work and it's s…

  8. comment
    Comment #39886892

    If you’re willing to pay for the latency cost of per layer cpu fetching/offloading, I don’t see what extreme quant buys you. You could just do a layer-by-layer fetching scheme with…

  9. comment
    Comment #39884800

    I see, so we’re still fetching the metadata to gpu, and rescaling on gpu, just on-demand and discarding metadata when we’re done with that layer? Why not do the same optimization f…

  10. comment
    Comment #39875915

    Thanks for the reply. I’m quite familiar with subchannel quant, but still feel like my questions did not get addressed. 1 Could you post the full memory use of the methods? E.g. yo…

  11. comment
    Comment #39871609

    Err, you are just restating what I’m saying, without addressing the concerns. 1 - is it fair to use ram in two places and report only one of them without any asterisk? (If you thin…

  12. comment
    Comment #39870848

    Really strong binary results. So strong it was fishy. I hope someone can explain my confusion below. > We compared the performance of the Llama2-7B model in three configurations: F…

  13. story
  14. comment
    Comment #39483493

    I’m not sure if you’re being facetious, but this is literary available for early access, but not ga yet. https://simonwillison.net/2024/Feb/21/gemini-pro-video/

  15. comment
    Comment #39336340

    Isn't this literally the case in the US? You list dependents on your tax form.

  16. comment
    Comment #39009798

    What is your current level? Is it shown on your LinkedIn?

  17. comment
    Comment #38285371

    And yet, a Rust Option (or really any option) can just be viewed as a list of one or zero elements. https://rust-unofficial.github.io/patterns/idioms/option-ite... In fact, in Hask…

  18. story
  19. comment
    Comment #37755623

    I ended up needing this so often for graph processing, and for values which might be inexact if using floating point, that I saved the formula in a blog post. https://vladfeinberg.…

  20. comment
  21. comment
    Comment #37451073

    What are some non-sanctuary cities which also have year round temperate, but not too hot or humid, weather?

  22. comment
    Comment #36952319

    Which papers?

  23. comment
    Comment #36927814

    I don't know if your questions are rhetorical, but have you ever been in the car with an exec? They very much do work in the car and take meetings from the car.

  24. comment
    Comment #36719588

    Intelectual parity to an average human by 2030? Would you be willing to bet on this?

  25. story