Live data from Hacker News

Viewing profile — EnPissant

EnPissant

HN member
Joined
Mon, Apr 21, 2025, 9:32 PM UTC
HN karma
168
Public activity
395 items

About EnPissant

No profile information was provided.

Recent public activity

  1. comment
    Comment #48580941

    "I'm selling an AI security product and want to establish my brand. I'll post several scare-mongering posts on my blog every week and people like solid_fuel will eat it up because …

  2. comment
    Comment #48579331

    I'm guessing all the "censored" boxes are not actually censoring anything and are placed there to make you imagine something much worse.

  3. comment
    Comment #48574387

    Rust gives you statically linked binaries as well. So your argument boils down to having to add `reqwest = "0.13.3"` to your `Cargo.toml`.

  4. comment
    Comment #48566885

    I thought MTP wasn't very useful on MoE models because the expert overlap for 2 tokens was too small.

  5. comment
    Comment #48566682

    When running on a GPU, dense models are shaping up to be the best way due to two things: - Maximum intelligence per VRAM (you dont have much) - Dense models can benefit from MTP to…

  6. comment
    Comment #48511138

    Pay? This is the best marketing they could have hoped for.

  7. comment
  8. comment
    Comment #48487123

    It's not real. It's like naming your movement "The Good People". It sprouted from the "Rationalist" community, which is even more self-aggrandizing. Neither has any hope of doing a…

  9. comment
    Comment #48431262

    > Over the past five months, our team has been running an experiment: building and shipping an internal beta of a software product with 0 lines of manually-written code. This is su…

  10. comment
    Comment #48351544

    They did not submit the full log because this is fake.

  11. comment
    Comment #48272770

    You are correct. I should have said "increased to 200%".

  12. comment
    Comment #48269889

    But you won't see 2x expert re-use, the speedup with 5 streams will be tiny.

  13. comment
    Comment #48265151

    >You don't need "very much" expert overlap to see aggregate gains at scale, you just need some of it I'm not sure what you are claiming. Decode is bottle-necked by memory bandwidth…

  14. comment
    Comment #48265025

    This is just wishful thinking. For prefill, it's really easy to batch MoE and get really good tk/s, even on a single stream. For decode, you will run into the problem that: 1) you …

  15. comment
    Comment #48264436

    Even if you could fit a 500B model's expert weights in very fast system RAM, it would run so slow as to be useless.

  16. comment
    Comment #48260124

    Also, electricity isn't free.

  17. comment
    Comment #48259642

    There was only a very brief time it was selling for MSRP (last fall for $2000). Even if you use that as the previous data point, it's only 200% increased.

  18. comment
    Comment #48248589

    > I hear "I'm not anti immigrant, I'm anti illegal immigrant" a lot. To which there is an easy solution: increase the number of legal immigrants we allow. Being "anti illegal immig…

  19. comment
    Comment #48248504

    The following things are not in contradiction: 1) Someone can be against illegal immigration and for legal immigration. 2) That same person's idea about who should immigrate to the…

  20. comment
    Comment #48247606

    [flagged]

  21. comment
  22. comment
  23. comment
  24. comment
    Comment #48001874

    [flagged]

  25. comment
    Comment #47991558

    Now try applying this logic to elevators.