Live data from Hacker News

Viewing profile — aradox66

aradox66

HN member
Joined
Sun, Oct 03, 2021, 12:18 AM UTC
HN karma
457
Public activity
220 items

About aradox66

No profile information was provided.

Recent public activity

  1. comment
    Comment #47200750

    [flagged]

  2. comment
    Comment #46501817

    Research suggests that aerobic fitness largely compensates for the negative health effects of being overweight. https://www.physiologicallyspeaking.com/p/physiology-friday-... e.g.…

  3. comment
    Comment #45724417

    How could that make sense? The emergent capabilities of the models are obviously critical to the evolution of that whole system.

  4. comment
    Comment #45706226

    I thought private equity was all about liquidating nearly bankrupt businesses

  5. comment
    Comment #45701378

    Don't "real" consequences apply for setting weights? There's an actual monetary cost to train these models, and they have to actually perform to keep getting trained. Sure it's VC …

  6. comment
    Comment #45596012

    It's not an assumption, it's an extremely developed international field of tactical and strategic study that leads to these conclusions

  7. comment
    Comment #45486385

    I love location sharing with friends. "Find my Jerry", "Find my Anna" we call it in our friend group. It simplifies logistics, gives us a narrow but fun little window into each oth…

  8. comment
    Comment #45160049

    You want to always use mineral zinc and to learn for your own body and the sun conditions of your environment whether and how often you need to reapply. You're not going to be doin…

  9. comment
    Comment #45158982

    It's a common failure mode and most other OTCs won't kill you if you occasionally accidentally take an extra dose or two. Agreed that it's safe at the correct dose but the hazards …

  10. comment
    Comment #45082898

    Arguably there are some evolutionary advantages to the presence of elders in social species with cultural transmission but your point stands

  11. comment
    Comment #45048695

    Seidenberg's work is really interesting but he's definitely not arguing that sweat lodges are a part of historical Jewish practice. He's doing a compare/contrast.

  12. comment
    Comment #44769032

    caught in the wild, the moving goalposts of the Turing test

  13. comment
    Comment #44724664

    Hahaha I love that idea. An LLM enters the Obsidian vault and responds to a prompt by following an arcane and elaborate sequence of calculations. Does it really understand?

  14. comment
    Comment #44718746

    I'm also not an expert here, but looking through the figures [1], that one HR result for Asperger's in a figure 4 is...surprising to see, given the headlines. This is far from stro…

  15. comment
    Comment #44534303

    Nope! Rabbi Akiva, who, as the story goes, was an illiterate shepherd until he started studying in his 40s, and went on to become one of the most renowned scholars of his era. This…

  16. comment
    Comment #44427987

    I think this mostly has been debunked in the last 15 years, except in the unscientific health influencer space https://pubmed.ncbi.nlm.nih.gov/39017376/

  17. comment
    Comment #44406238

    Also, at temperature 0 LLMs can behave deterministically! Indeterminism isn't necessarily quite the right word for the kind of abstraction LLMs provide

  18. comment
    Comment #44406193

    Determinism isn't always ideal. Determinism may trade off with things like accuracy, performance, etc. There are situations where the tradeoff is well worth it.

  19. comment
    Comment #44261262

    I imagine there are some boundary constraints - mobile video can only be so compelling. I would guess we're already pretty close to that threshold.

  20. comment
    Comment #44250117

    It's a different chemical cocktail, but more importantly consenting to it makes a bit of a difference in terms of how it registers morally.

  21. comment
    Comment #44119341

    There is no shortage of cranks generating novel cosmological theories, but this writer doesn't seem to be one of them. He's interested in predictive power! I think it's fascinating…

  22. comment
  23. comment
    Comment #44015959

    Can't recommend dbt highly enough to get over the SQL boilerplate problem. After using dbt for a few years it's unimaginable to work without it.

  24. comment
    Comment #43968436

    Ah, yes, the ol "it was just one tab, but it's very strong/pure."

  25. comment
    Comment #43823435

    Indeed this kind of utilitarian suffering quantization leads to some weird, some would say "repugnant" conclusions