Live data from Hacker News

Viewing profile — imrehg

imrehg

HN member
Joined
Wed, Dec 16, 2009, 8:53 AM UTC
HN karma
1,395
Public activity
597 items

About imrehg

Senior Machine Learning Engineer at Faculty https://faculty.ai

Contact: gergely@imreh.net

Based at http://gergely.imreh.net

Taipei Hackerspace co-founder https://taipeihack.org

[ my public key: https://keybase.io/imrehg; my proof: https://keybase.io/imrehg/sigs/Vqc30FUDkAzeR6b4jTTQown3Qjned_Xju7DqvvT6DfU ]

Recent public activity

  1. comment
    Comment #49268793

    I did report my test results on the first one. :) The second I didn't say it's any of llama.cpp's "fault", but it is _related_ to llama.cpp since it's being shipped in another syst…

  2. comment
    Comment #49268627

    I just switched to Vulkan, and be done with it. :) As much as I can tell, the ROCm version of llama.cpp would be a bit faster on prompt processing, but about the same on the token …

  3. comment
    Comment #49268518

    llama.cpp works pretty well for me on the Framework 13 laptop, but the current era of "move fast, break things, rarely fix" (sorry, that's how it feels), bites here quite a bit. Tw…

  4. comment
    Comment #49156851

    I have sent countless small bug fixes even before, to tools that I use that I could dig into when something was off... No big heavy lifting, but plenty of "tinker on your tools" st…

  5. comment
    Comment #49057808

    Well, a had some new account causing havoc (posting crap on packages, requesting delete of packages...) and they were registered today. I guess I'll have to start flagging. Also, h…

  6. comment
    Comment #48920608

    It's scary how your first paragraph describes by recent (but not long term) experiences. Looks like I have some checks to do as well.

  7. comment
    Comment #48877847

    That's about the speed I get on a AMD Ryzen AI 9 HX 370 (inside a Framework 13), with Qwen3.6-35B-A3B, so doing the same on that much larger model...

  8. comment
    Comment #48801043

    As long as places like Taiwan are effectively blocked from being in the UN[0], I don't believe we should be adding more to their power and responsibilities than what they have now,…

  9. comment
    Comment #48785704

    I've relatively recently migrated over to using btop[0], and it's the kind of modern interface, useful and informative, that I needed. As others mention it - it seems to shows the …

  10. comment
    Comment #48784642

    Should be, judging by their LinkedIn from the bottom of the page (Centaur Technology - Member of Technical Staff: 2005 - 2013) This is a blast from the past. Centaur was doing VIA …

  11. comment
    Comment #48743555

    From the Huggingface page, it is a fine-tuned version of the Qwen3.6-35B-A3B that is my alternative, and the benchmarks seems to be better. So I'm using as a "likely some quality g…

  12. comment
    Comment #48729454

    On the MoE versions of these models the MTP versions have only marginal benefit. In my trials the speed-up is 15 token/s... on my device. I still use the MTP version as it _feels_ …

  13. comment
    Comment #48729345

    I'm having a decently good time time with `qwen3.6-35b-a3b-mtp` (unsloth's multi-token prediction version) and and `qwen-agentworld-35b-a3b`. On a 2021 M1 Pro (32GB RAM) I can get …

  14. comment
    Comment #48670141

    I value Open Source very highly, and enjoy contributing. Fix a bug here, add a small feature there, most OSS projects have low hanging opportunities. And these days, to "switch off…

  15. comment
    Comment #48657000

    Above a certain size you cannot get the quota-based models. In my org, as our Team account (quota-based) was growing, to onboard more people we _had to_ switch to Enterprise (and t…

  16. comment
    Comment #48610524

    Oh, this was one that I played a lot as a kid! (Alongside F-19 Stealth Fighter, F-117A Nighthawk Stealth Fighter - the two that apparently came before and after this game - TIL, an…

  17. comment
    Comment #48475228

    I encountered this when I was checking why my gluten-free bread came out the bread machine the way it did. I guess it latched onto some yeast-related points and it fell back to Opu…

  18. comment
    Comment #48280195

    Yeah, I've ended up mapping "M5" to "End" (so that "Home" has a counterpart). This current one still has the Ins/Del/PgUp/PgDn/Home for direct access, but certainly some of the lac…

  19. comment
    Comment #48264323

    I've just recently got a mechanical keyboard, after doing a workspace ergonomics assessment, and also cultural influence from various other developer friends, etc... It's a Keychro…

  20. comment
    Comment #47945517

    Yup, that's exactly it: - get the gravitational constant with these two known masses - then can deduct the mass of the unknown Earth by its interaction with other masses (say the "…

  21. comment
    Comment #47295904

    I haven't really seen many alternatives besides what's listed, and looking for it (FOSSin my personal development, closed source at $WORK). Imho the best options there are - GitLab…

  22. comment
    Comment #45473011

    Reminds me of PoC||GTFO. Loads of fun in this and in that. And does indeed make me sometimes question if I know anything about programming (in a good way:)

  23. comment
    Comment #44665912

    Yes, the site seems to save the checked cocktails into Local Storage (you can click some, and in your browser's inspection tools you can check. Eg. in Firefox > Inspect > Storage >…

  24. comment
    Comment #44540039

    News TV channels in Taiwan that I usually watch, very often use videos from Threads for local news reporting (stuff sent in by the public). X-originated ones pretty much disappeare…

  25. comment
    Comment #43601691

    I still have my Thinkpad X201 from 2011 running, and I use it as my personal machine (have an M1 MacBook from/for work). Of course, had to replace the hard drive once or twice, rep…