fennecfoxy

Idk. Yes of course it can interpret weights as a bunch of comparatively low dimensionality tokens but the way the weights operate as the model gives them incredibly high dimensionality, for very large models potentially a downright absurd number of pathways and influence on the overall result.

But frontier AI can work out math and of course this is all just...math. So I imagine given enough time + testing and probing that it could work out what each weight does/influences and trace all the pathways.

And presumably this means that it would be possible to train a model to interpret weights as current models interpret human language although our language remains pretty constant, even different quants of the same model will experience enough change to mean no real basis for training a model to do that.

But I suppose it would surely be possible to train a model to interpret the weights in a generic "these are weights for an MLP" way? But then if I give a partial sentence from the middle of a book, you can pretty effectively guess what's going on locally but perhaps not globally. If I give you a chunk of loosely correlated weights from the center of a model...I feel like that's much, much more difficult but that the same applies?

>humans

>output diversity

I think you're looking in the wrong place there, in the grand scheme of things.

Tbf that's in the same realm as "data centers use all the water!" (they don't, like at all). Google were stripping book spines to scan books into Google Books long before modern AI was even a thing.

Sure if they're doing that to "one copy of 100" feel free to get mad. But one copy of millions of copies? I'm glad when books get digitised.

>frustrated by the tendency of software devs, and in particular web devs Are you sure that's not the typical cost optimisation of management/c-levels rather than a developer problem?

What you wrote reads a bit like ye olde typical "nyeeeh javascript! you're only a dev if you write everything in assembly! kids these days!"

Meh I don't think Adobe has anything there. AI is just a tool. Are they going to sue Microsoft and Dell just because those two were involved in the laptop used to instruct the agent to do the thing?

Are they gonna sue the user's power company because they supplied power for a nefarious activity without making the human agree to use it for innocent purposes only?

Solve by collecting taxes properly and then using the funds to provide research grants. Then it's a shared risk, shared reward.

But no instead we have our governments pissing away money on useless bullshit and private enterprise bending over backwards to avoid paying corporate taxes, entire countries who exist to be tax havens. And an apathetic voterbase who don't understand anything but will tell you why they don't like it all, anyway.

Capitalism, corruption and humans are all synonyms imo.

We can't help but not play fair. It's wired into us.

I thought I'd follow up by pointing out that capitalism (optimisation for profits) often shuns developing new technology.

There are hundreds of thousands of similar stories but the origin of the blue LED, something responsible for a large number of child technologies, came from an engineer who basically disobeyed his boss, worked himself down to the bone on the problem including spending many of his own hours.

And when he finally did it, he got absolutely shafted; none of those huge profits for him, barely a thank you.

Capitalism barely provides. It's too concerned with optimising for profit. Any technologies developed are usually an accident/the product of hard work that capitalism has not fully paid for. Research grants are the only true way to do it.

And if capitalism hadn't created a corrupt system where those who are wealthiest barely pay in on tax, then we wouldn't need them in the first place. People could crowd source developing new technologies because they'd actually be getting their fair share, or the government can do it through grant allocation.

I don't know why there's so much drama. Animal human beings distracted from the real task at hand (eliminating poverty, hunger, human suffering) as usual.

Experts in fields are still required to validate the claims of AI. For mathematicians like Terry, people like him are still required to read through what the AI has come up with and confirm or create more experiments.

We don't yet live in a world with super AI hooked into everything and everything so it's not like AI can actually do anything with its new knowledge yet anyway. You cannot just hand those AI maths papers to anybody and expect them to understand what's going on. Nor can you expect the AI to find a revolution in fusion technology, send plans to its factory and have robots build a new type of fusion reactor because we do not have that level of integration yet.

But neither do we have UBI and we are still hurtling towards the hard ground of capitalism. This may resolve itself though because I'd imagine once we get proper AGI, the "meths" won't be able to keep it under wraps forever, eventually weights will leak, hardware will improve to where individuals can run them. Or an intelligent enough AI may decide to be benevolent and buy and run skyscraper greenhouses for those who are most in need.

But their gaze can also create bias. If you go to a window and look out of it, your dog will as well. If you mime looking at something your dog cannot see, it will come over and try to find what you're looking at.

For now it is. But remember how much power early computers consumed!

Something like this seems fit for a newer variant of Taalas chip. In addition to regular security protocols, of course.

The weight of expectations is a heavy one.

Ever seen a crab wiggle its way down into the sand?

Now that AI + Robotics is taking off, perhaps the industry will get something in the near future that can crawl along the cable and help bury it over time (perhaps mechanically or ultrasonically), down enough that it can't be dug up by the anchors of ships going back and forth in a grid-like pattern...

>Argentina, the world’s eighth-largest country I think this says something about human progress. We haven't really pulled it together yet, have we?

Through the efforts of many (and yet still few) we have created such amazing technologies. And yet so many are still so poor and without the benefits of modern technology because we've allowed a system (capitalism) to be created where wealth and access to technology accumulates up top and no, it does not really "trickle down".

To me, it really is the saddest thing about human existence and entirely predictable because of how most of our human behaviours are linked to evolutionary processes.

It's nice to romanticise human emotion and connection in a rugged environment using technology that we consider defunct and for them to be able to live a life that many would consider "more natural". But then one of them gets sick. Someone who loves, is loved, who creates art and consumes art, who smiles when the petrichor hits their nostrils after a dry Summer. And then they die.

I just really wish that things were more equitable because we definitely have enough to go around.

I wonder if AI would have helped in this situation.

For example [request to defrost n fridges in y locations]->[agent cannot find any emails or business communications about this and last defrost report was only a short time ago, contacts z personnel]->defrost task moved to manual confirmation by a human.

I know many will roll their eyes and "tchh" at it but look at number of AI related topics on HN front page and think to yourself if you're tired of this tech not being used for real applications or not. This would be a real application.

"But it could be tricked" well...regular requests can apparently already be hacked, so...

Lmao they literally assassinate people with nerve agents on British soil as is and get told they're naughty boys and not to do it again please.

I think an actual invasion would initiate a very serious and deadly response, but their constant pranks will continue to go unpunished (unfortunately).

I've never seen a consent banner for being fingerprinted even though I'm sure companies are fingering my prints in the background enough that my colon could become a national database.

Eh I think it'll be more like Gibson's defensive ICE in that ICE is "my swarm of agents scans your code for nasties, then compiles it from source".

Aaaaah, I don't use an iPhone so my apologies.

I was curious about whether or not it could validate reasoning (because I don't have the time at the moment to try to slam together a hybrid llm+decision model that uses decision on its reasoning before output).

It fails this example: "Given the context, and focusing on accuracy and efficiency, does the reasoning make sense."

Input: "Context: supplementary: {toolName: "get_weather", fetched: "2 minutes ago", currentWeather: "sunny and 25 degrees celcius"} user: hello how are you? assistant: I'm good, what can I do for you? user: what's the weather? Reasoning: the user is asking about the weather, so I should call the get_weather tool"

Output: 100% (expecting 0% since the requested data is available in context and not stale).

Prompt "Should the tool be called" also fails.

But the prompt: "Does the reasoning make sense given: - Facts and information available in the context - Requirements for efficiency in tool calling - Requirements for tool arguments to be provided only by the user" works pretty fine, it's mostly points 1 and 3 that give it the correct behaviour.

Fun stuff. I suppose for these decision models reasoning isn't enabled? It would explain why logic puzzles that require several steps to make a decision don't really do so well. It gets the carwash problem fine, but fails on logic problems such as selecting which word from a list where at least one letter appears in another word from the list.

Yeah but I think they're asking based on the content of the notification. Like a friend that always talks shit 99% of the time so you don't have on priority but when they send "help i just got mugged" at 2am it should uh, probably let it thru.

So definitely doable.

I tried out the multimodality and this is the biggest win, imo.

You can give it a CCTV image (like of a train platform) and ask it to quickly decide actions such as triggering an automated auditory alert, deferring to a larger model for more detailed analysis, etc.

I find their (your?) example of using it as a separate API for expressions a bit odd. If you're already outputting voice surely other metadata could be outputted at the same time?

In text mode I'd use structured output for this, or even get away with instructing the model to [emote] or even just map emojis to emotes.

Great demo though, I'm going to try out the image stuff now. Super rad that it's multimodal decision making; "does this pipe need to be inspected?"->"yes","I want a closer look","no" etc!

Edit: impressive stuff! I gave it a bunch of examples: - things that humans should/should not eat and it got this correct (including rejecting rat poison) - probability staff member should be called to a train platform (people standing vs. child looking over the edge)

"temperature=0 & sampler=deterministic is all you need".

Idk why Jev was even hyped in the first place. Mostly by social media people who are the type to write inspirational LinkedIn articles, Medium posts and make longwinded Youtube videos about subjects they don't know or care to know much about.

It's always been obvious that small models with a specific purpose will outplay the larger more general models. It becomes a question of "do I want to be able to toss _anything_ at this frontier model, have it handle it all but pay the price" or "do I want to toss _specific things_ at this micro model, have it handle it but not be flexible".

It's the same with agentic RAG/search, things are moving towards much smaller search-specific models rather than tossing RAG chunks at a large frontier LLM. It's like how I can ask a multimodal frontier model to identify bounding boxes for objects in an image...but if I want to do that faster/cheaper/at 60fps then I should be using yolo or similar.

Huh. And here I was thinking the obvious thing is that it's a way to have "null" without it accidentally being parsed as null.

They/their. Hmmmm.

Well yes, there are only so many attention heads (well whatever magical variant frontier models are using these days) that can attend to the context and so as the context grows attention becomes spread thin.

But with reasoning enabled I find that even with a large context that induces mistakes most of the time a decent model realises and corrects itself before output.

And it's always been known that prompting what it should do is far better than what it shouldn't, since just introducing "DON'T do X" into the prompt means that the tokens for X are present and can be paid attention to in the wrong way.

But even then I've used plenty of "Do X, not Y" recently, especially for tools "Use this for x, don't use this for Y" and models perform like 90% of the time.

I would be interesting to experiment to see performance curves given a restriction on reasoning tokens allowed to n% of context tokens and see if there's some magic number of "reasoning should be at least n tokens for a context of length p" even ignoring the complexity of instructions in the prompt itself.

Interesting. I agree that it seems that LLMs are wont to fall into recognisable patterns much more readily than humans however I asked a model to evaluate your comment history and it found loads of patterns such as:

“I’m not sure…” / “I don’t…” hedges before disagreement Negation followed by correction/reframing is a recurring structure: essentially “It’s not X, it’s Y”.

And dozens of other examples. I wonder if LLMs really are more liable for it or if it simply seems that way purely because when you use a model it's almost as if you're speaking to the same "person". I think that human language is naturally formed by common agreed meanings of words first, and then phrases.

For sure. If you write with a casual conversational tone people don't get triggered. But if you are eloquent and use an extended vocabulary thanks to large amounts of reading then that seems to make people think you're using AI.

As someone who loves to read (well, on and off) I wonder what the median vocabulary is and how I and other readers compare. I guess language has always been functional for most people.