FloorEgg

23 ms·

1. 331.▲
   by FloorEgg 1y ago

   Thank you for saying that :)
2. 332.▲
   by FloorEgg 1y ago

   Hey in the part 3 that introduces PAYG it jumps from $100 down $40-65 a month to $0.21 a day / $1.50 a week. It seems your mixing up examples since they're off by an order of magnitude. Once I read that my trust in everything else
3. 333.▲
   by FloorEgg 1y ago

   "Minimize surprise" and "maximize accurate predictions" are the same thing mathematically. Minimize free energy = minimize prediction error. LLMs do everything modelled in the free energy principle, they just don't
4. 334.▲
   by FloorEgg 1y ago

   > An llm generates the next word based on what best matches its training, with some level of randomisation. This is sort of accurate, but not precise. An LLM generates the next token by sampling from a probability distribution over possi
5. 335.▲
   by FloorEgg 1y ago

   I did mean "how so", haha. Thanks. I haven't used meetup since the pandemic, so all this is somewhat disheartening news.
6. 336.▲
   by FloorEgg 1y ago

   LLMs embody the free-energy principle computationally. They maintain an internal generative model of language and continually minimize “surprise”, the difference between predicted and actual tokens, during both training and infeence. In Fri
7. 337.▲
   by FloorEgg 1y ago

   How do?
8. 338.▲
   by FloorEgg 1y ago

   It seems maybe you mixed up non-fungible value-bottlenecking human services with "everything". Many things can get cheaper, some things get more expensive, and median person gets more wealthy and buys more of both. In reality, yes
9. 339.▲
   by FloorEgg 1y ago

   I already answered under the other comment asking me why and if your curious I suggest looking for it. Very short answer is Karl Friston's free energy pricniple
10. 340.▲
    by FloorEgg 1y ago

    In a sense, yes. If one is to quantify consciousness it would probably make sense to think of it as an area of awareness and cognizance across time. Awareness scales with sensory scale and resolution (sensory receptors vs input token limits
11. 341.▲
    by FloorEgg 1y ago

    That's a good point of positivity and I welcome it :)
12. 342.▲
    by FloorEgg 1y ago

    Karl Friston's free energy principle is probably roughly 80% of my reasons to think they're coupled. The rest comes from studying integrated information theories, architecture of brains and nervous systems and neutral nets, more b
13. 343.▲
    by FloorEgg 1y ago

    I think LLMs are conscious just in a very limited way. I think consciousness is tightly coupled to intelligence. If I had to guess, the current leading LLMs consciousness is most comparable to a small fish, with a conscious lifespan of a fe
14. 344.▲
    by FloorEgg 1y ago

    The good: The internet enabled access to a wider diversity of cultures. The bad (I think?): Culture decoupled from geography, which makes it harder to connect with local community. The ugly: The attention economy optimizes for what is novel
15. 345.▲
    by FloorEgg 1y ago

    What do you mean by OP? The comment I responded to didn't refer to pop stoicism or use the word vacuous. The word vacuous also didn't appear in the article. Stoicism is the study and reflection of what is and isn't in our con
16. 346.▲
    by FloorEgg 1y ago

    That doesn't sound like stoicism to me. At its core stoicism is about having the best possible judgement and taking the best possible actions. Sometimes acting makes a situation worse and so patience or restraint are what's best.
17. 347.▲
    by FloorEgg 1y ago

    Edit: the following is incorrect. I didn't know that the change to IRC § 174 was cancelled this summer. ------ What's crazy is that with the 2021 changes to IRC § 174 most software r&d spending is considered capital investment
18. 348.▲
    by FloorEgg 1y ago

    For me I can "see" things in my minds eye and it almost doesn't matter if my eyes are closed or not. The detail isn't perfect and it's unlike anything else I would ever "see", such as a blurry image or sim
19. 349.▲
    by FloorEgg 1y ago

    I can only speculate so take it with a spoon of salt... But people who are really good at sussing out other people's wants can find ways to make more people happy. It's like a maze of interests and I guess he's just really mo
20. 350.▲
    by FloorEgg 1y ago

    I guess so, but it's relative. An hour isn't hard if you're driven by curiosity, but it is if it's a chore you're doing for someone else.
21. 351.▲
    by FloorEgg 1y ago

    What version of chatGPT was this? That sounds like a gpt-3.5 kind of response. I wonder how the latest models would fare in these kinds of conversations.
22. 352.▲
    by FloorEgg 1y ago

    I bought a Samsung phone back in like 2014, and shortly after bought smartwatch to pair with it. A year later, Samsung released an update that removed the pairing functionality so my smartwatch could no longer pair. They did this in conjunc
23. 353.▲
    by FloorEgg 1y ago

    Actually it took me almost an hour to re-do a portion of the research and lay it out, which I did further down the thread, if you care to look. Also you didn't ask me to link you to my sources, you asked me what my sources were. I answ
24. 354.▲
    by FloorEgg 1y ago

    Actually, I work for many people: My customers, my colleagues, my family. I just don't work for strangers on HN. My mistake was answering judahmeek's question directly. They asked "What are your sources?" and I answered
25. 355.▲
    by FloorEgg 1y ago

    I agree completely, though I continually wonder why. Is it by specific design? By economic incentive? Is it because novel threats attract attention, and having a need to be validated continually satisfied maintains attention? I can't h
26. 356.▲
    by FloorEgg 1y ago

    I became aware of the legal contradictions last summer and spent a few hours doing searches and reading through the relevant regulatory language for a few industries. I don't have all the references handy. I don't work for you. It
27. 357.▲
    by FloorEgg 1y ago

    The federal regulations... It's not hard to find if you go looking...
28. 358.▲
    by FloorEgg 1y ago

    > it's not like there were DEI advocates out there giving guides on how to break the law I think you're very mistaken. Not only were their guides, but there were federal regulations mandating that the laws be broken. It is&#x2F
29. 359.▲
    by FloorEgg 1y ago

    I don't think that's quite fair, as in many cases there were federal regulations that pressured industries into behavior that was discriminatory to one group in order to favor others. In fact there was an accumulation of contradic
30. 360.▲
    by FloorEgg 1y ago

    I am wondering the same thing. It's interesting that they didn't report at all on the median and only the average. Also find the timing interesting, as I can't help but suspect that both the justification and incentive for se