Sure. I buy tickets to their concerts, have bought CDs, movies, buy their game in the next Steam sale or on Humblebundle, rarely Patreon or support indie things on Ko-fi or whatever. I buy a novel if I enjoyed the first chapter(s) and want it on paper. Or go to the library. I just can't afford all the music and Spotify isn't paying the artists properly either. And I don't want a DVD collection, so for TV series they don't get money from me. Except for what the one streaming service I pay for forwards to them.
I've used laptops for more than a decade. And sure, in the early times thermal management wasn't that elborate. But I really haven't seen any laptop in many, many years that doesn't do it with perfect accuracy. And usually it's done in hardware so there isn't really any way for it to fail. And I played games and compiled software for hours with all CPU cores at 100% and fans blasting. At least with my current laptop and the two Thinkpads before. The first one had really good fans and never went to the limit. The others hit it with an accuracy of like 2 or 3 degrees. No software necessary. I'm pretty sure with the technology of the last 10 years, throttling doesn't ever fail unless you deliberately mess with it.
But now that I'm thinking of the fans... Maybe if the fan is clogged or has mechanically failed, there is a way... A decent Intel or AMD CPU will still throttle. But without a fan and airflow inside the laptop, other components might get too hot. But I'm thinking more of some capacitors or the harddisk which can't defend itself. The iGPU should be part of the thermal budget of the rest of the processor. Maybe it's handled differently because it doesn't draw that much power and doesn't really contribute to overheating it. I'm not sure.
Maybe it's more a hardware failure, a defective sensor, dust, a loose heat conductor, thermal paste or the fan? I still can't believe a laptop would enter that mode unless something was wrong with the hardware. But I might be wrong.
But reading that text like they tell you to do, is kind of an exercise in futility if you choose topic two. (the benefits of artificial satellites in telecommunications) I'd be angry at that point.
Why does it force the processor over the limit in the first place?
I think in every other laptop the CPU just throttles when it gets too hot. Meaning it can never exceed the maximum temperature. I wonder if this is a misunderstanding or if HP actually did away with all of that and designed a laptop that will cook itself.
And it's not even a good design decision to shutdown the PC if someone runs a game... Aren't computers meant to run them? Why not automatically lower the framerate by throttling? Why shut down instead?
Btw: Might be that you're behind a NAT (router) and that's why bittorrent doesn't connect. You'd need to figure out which port your torrent client is configured to listen on and then do "port forwarding" of that port to your machine in the router you got from your ISP. Or use something like UPnP that does this automatically.
Not sure if that applies in your case and it's unsolicited advice... But a fairly common issue with bittorrent.
I think most people use something like exllamav2 or vllm or use GGUF to do inference and it seems neither of those projects have properly implemented multimodality or this specific model architecture, yet.
You might just be at the forefront of things and there isn't yet any beaten path you could follow.
The easiest thing you could do is just use something that already exists, be it 4bit models, wait a few weeks and then upgrade. And I mean you can also always quantize models yourself and set the parameters however you like, if you have some inference framework that supports your model including the adapters for vision and has the quantization levels you're interested in...
And they're not even that woke. Afaik they still ocasionally eat animals in the 24th century. (Unless they're Vulcan.) Watch The Orville if you want some proper progressive shit 😆
Ja. Ich finde das auch schlimm. Vor allem, dass jeder Mensch unbedingt klar Stellung beziehen muss und das im Internet groß breitreten be jeder Gelegenheit. Mir tun einfach die Menschen dort leid die darunter leiden oder ihr Leben verlieren. Da fällt es mir auch echt schwer Sympathie für eine der Seiten aufzubringen und da zu verkünden wer Richtig liegt und wer keine Daseinsberechtigung erhält...
Ich skippe seit dem alle Artikel und Diskussionen zu dem Thema im Internet. Die sind meines Erachtens fast alle toxisch.
Also vor Oktober 2023 und der jüngsten Eskalation des Konflikts habe ich hier selten was zu Israel gelesen. Also würde ich da nicht zustimmen. Seit dem ist das Internet voll mit Hass und starken Meinungen dazu. Auch hier auf Lemmy.
Well, I'd say there is information in language. That's kinda the point of it and why we use it. And language is powerful. We can describe and talk about a lot of things. (And it's an interesting question what can not be described with language.)
I don't think the stochastical parrot thing is a proper debate. It's just that lots of people don't know what AI is and what it can and cannot do. And it's neither easy to understand nor are the consequences always that obvious.
Training LLMs involves some clever trickery, limit their size etc so they can't just memorize everything, but instead are forced to learn concepts behind those texts.
I think they form models of the world inside of them. At least of things they've learned from the dataset. That's why they can for example translate text. They have some concept of a cat stored inside of them and can apply that to a different language that uses entirely different characters to name that animal.
I wouldn't say they are "tools to learn more aspects about nature". They aren't a sensor or something. And they can infer things, but not 'measure' things like an X-ray.
I'm currently reading the paper. I occasionally debate here on Lemmy whether LLMs are just stochastic parrots, or if they actually grasp the concepts they're talking about. There's also evicence for that.
Ultimately I wonder if and when we'll get LLMs that address 'hallucinations' and expose a setting to adjust the factuality of the answer. I suppose that's somewhere in the model or at least possible to learn for the model. But certainly not controlled or factored in in the current generation of LLMs.
Look at the USA, UK or countries like China. I think they're all ahead if us. Leading in different fields. A skewed balance between capitalism and citizens rights, surveillance in general, and a dystopian surveillance state.
Thanks for taking the time to explain it to me. The Github issue also is very helpful. Seems that's exactly my answer to "Why do I need a fourth store in addition to F-Droid, AuroraStore and Obtanium" 😉
Have a nice day, thanks for the STT keyboard! I didn't really engage in the discussion because I'm exactly in the same situation as other people here. I already have the FUTO one and Sayboard... But eventually I'd like to replace FUTO software with free software alternatives. I don't like their licensing. So this is very welcome.
Sure. There isn't any paragraph on how it compares to other appstores or why the author started the project in the first place despite several other stores being available.
So I'm looking for the selling point. (Aside from your App being available there.)
Can someone enlighten me oabout the specifics of the accrescent.app appstore?
I guess it's somewhat like Obtanium in that it fetches releases packed by the original developers, just plus an index, metadata and signing, thus more convenient and secure? I guess it's open-source and everything? What are the unique benefits?
Well that paper only says it's theoretically not possible to completely eliminate hallucination. That doesn't mean it can be migitated and reduced to the point of insignificance. I think fabricating things is part of creativity. I mean LLMs are supposed to come up with new text. But maybe they're not really incentivised to differentiate between fact and fiction. I mean they have been trained on fictional content, too. I think the main problem is to control when to stick close to facts and when to be creative. Sure, I'd agree that we can't make them infallible. But there's probably quite some room for improvement. (And I don't really agree with the premise of the paper that it's caused solely from shortcomings in the training data. It's an inherent problem in being creative and that the world also consists of fiction and opinions and so much more than factual statements... But the training data quality and bias also has a severe effect.)
That paper is interesting. Thanks!
But I really fail to grasp the diagonal argument. Can we really choose the ground truth function f arbitrarily? Doesn't that just mean given arbitrary realities, there aren't hallucination-free LLMs in all of them? But I don't really care if there's a world where 1+1=2 and simultaneously 1+1=3 and there can't be an LLM telling the "truth" in that world... I think they need to narrow down "f". To me a reality needs to fulfill certain requirements. Like being contradiction free etc. And they'd need to prove that Cantor applies to that subset of "f".
And secondly: Why does the LLM need to decide between true and false? Can't it not just say "I don't know?" I think that'd immediately ruin their premise, too. Because they only look at LLMs who don't ever refuse and have to decide on a truth.
I think this is more related to Gödel's incompleteness theorem, which somehow isn't mentioned in the paper. I'm not a proper scientist and didn't really understand it, so I might be wrong with all of that. But it doesn't feel correct to me. And I mean the paper hasn't been cited or peer-reviewed (as of now). So it's more like just their opinion, anyways. I say (if their maths is correct) they just proved that there can't be an LLM that knows everything in any possible and impossible world. That doesn't quite apply because LLMs that don't know everything are useful, too. And we're concerned with one specific reality here that has some limitations. Like physics, objectivity or consistency.
FYI: There's also AnLinux, Linux Deploy, Termux, tainer, UserLAnd, ...
Some of them aren't maintained anymore. And they don't necessarily have hardware-acceleration. But don't all require root and system patches.