daveguy

40 ms·

721.
by daveguy 1y ago
The transformers are accurately described in the article. The confusion comes in the Reinforcement Learning Human Feedback (RLHF) process after a transformer based system is trained. These are algorithms on top of the basic model that make
722.
by daveguy 1y ago
Completely agree with you about social media. I'm not a fan of social media algorithms and believe they also produce much more harm than benefit.
723.
by daveguy 1y ago
Yeah, but should we have psychosis force multipliers?
724.
by daveguy 1y ago
And instead of coasting for a year or two he apparently crashed and burned on a ketamine fueled binge.
725.
by daveguy 1y ago
Tesla has garbage fit, finish and build. They are run by a cheap-out scammer who puts more energy into social media than his business endeavors. 250kw in 2020 vs 1000kw (4x) in 2025. When is Tesla coming out with a 1MW charger? Sell your st
726.
by daveguy 1y ago
You construct a theory of mind of the author of a work whether you recognize you are doing it or not. There are certain things everyone assumes about code based on the fact that we expect someone who writes code to have simple common sense.
727.
by daveguy 1y ago
Leaving out literally all of the context is pretty damn partisan, no?
728.
by daveguy 1y ago
Oh yeah, boo federal government bots because boo federal government. Hilarious.
729.
by daveguy 1y ago
> creating a dedicated, pre-trained foundation model is expensive. Not to mention a full LLM. Creating a dedicated pretrained model is a prerequisite of any LLM. What do you mean by "full LLM"?
730.
by daveguy 1y ago
They can't get impatient either. Only some vague semblance of what impatience means to a programmer or two.
731.
by daveguy 1y ago
> A sign of model collapse would be new LLMs having degraded performance. That's a good point. Stagnant performance would be a different thing -- more indicative of the technology itself.
732.
by daveguy 1y ago
None of the LLM models are self-building, self-healing or even self-thinking or self-teaching. They are static models (+rag, but that's a bolt-on). Did you have a specific tech in mind?
733.
by daveguy 1y ago
But, the internet is already polluted and being constantly polluted by LLMs. It feels like LLMs are the PFAS of the internet. So, I'm not sure what your argument is, but I would welcome clarification. Maybe I am misunderstanding you
734.
by daveguy 1y ago
All. Of. Them. It's quite literally what they do because they are optimistic text generators. Not correct or accurate text generators.
735.
by daveguy 1y ago
> the raw code quality that LLMs generate is usually top notch, even if the functionality is completely off. I'm not even sure what this is supposed to mean. It doesn't make syntax errors? Code that doesn't have the correc
736.
by daveguy 1y ago
I agree. Hedge funds coming out with "hey! look at us!" this late in the game is a huge red flag.
737.
by daveguy 1y ago
Looks like 65-90% of people can still easily tell an AI generated video from a real one. And sorry, bub, I'm not telling you why the AI videos are being identified. Human impressions are for humans.
738.
by daveguy 1y ago
> I just rewrote 300ish advanced PostgreSQL queries to mysql queries. Translation from one set of tokens to another is exactly the primary use case of LLMs. This is exactly what it should be good at. Developing new queries, much less so.
739.
by daveguy 1y ago
Surely most people can recognize the difference between NIMBY of a noisy wasteful cover for Bitcoin mining operation compared to the NIMBY of not wanting "the poors" nearby or the ability to retain high rent charges on hoarded hou
740.
by daveguy 1y ago
Or just don't mess with bitcoin and other crypto. It's pretty easy to avoid.
741.
by daveguy 1y ago
Oh cool, boxing is good exercise. Hadn't really thought about how fun that could be VR. Thank you.
742.
by daveguy 1y ago
If only Alexa could be trusted to buy something as seemingly simple as eggs.
743.
by daveguy 1y ago
Bezos thanks you for your acknowledgement and service.
744.
by daveguy 1y ago
Two questions, 1) What sensitive information was published in this article besides some superficial listening preferences and some Alexa interactions we have all had? I'm not sure identifying the extent of the use of the prefix "o
745.
by daveguy 1y ago
How does that work? I generally consider working out to be the complete opposite of VR. Was it a VR tennis-cardio type thing?
746.
by daveguy 1y ago
The point is, AI is not able to act capably or autonomously. If someone tells it to "act autonomously" it'll either sit like a lump or flail until the power runs out.
747.
by daveguy 1y ago
You say that like AI has some sort of autonomy. It doesn't.
748.
by daveguy 1y ago
This is excellent communication and respect for their users. Applause for Glitch. Especially allowing users to retrieve resources for a full 6 months after the closure. If you appreciate this level of communication and respect, avoid Digita
749.
by daveguy 1y ago
> Between Harvard, Yale, and possibly a few other schools, I thought they had influence throughout government. Turns out the "deep state" is just some made up bullshit to make people distrustful, angry, and easier to manipulate
750.
by daveguy 1y ago
Turns out teachers are people and general understanding evolves over time and not all at once. Who would have guessed. Well, Laplace maybe.

More ›