HTTP provided a means for multiple clients to fetch resources from a central server. If centralization was antithetical, the clients would have been P2P.
Consider a calculator, a machine capable of valid transformations wrt the semantics of arithmetic.
Does a calculator reason its way to 2 + 2 = 4? No, it merely transforms.
An LLM is simply a much more complex version of this; except that its complexity hides a lot of what is otherwise reducible to a transformation of semantic, conceptual, and inferential relations. There are lots of NLP models prior to the introduction of modern "AI" that would make these representations more clear.
However I don't believe any of the aforementioned philosophers would ever consider reason itself as a transformation, as Kant especially spent a lot of time differentiating apprehension from transition.
I’m not so sure. While I completely agree on pure transformation, I continue to believe that the most concerning aspect of neural networks is that they really strengthen the argument that consciousness is the illusion.
The interplay between deductive frameworks and inductive frameworks is still always buoyed by the black swan problem… which is a the heart of Hume’s solipsism. Kant’s pure reason never considered instinctual logical gates that come prepackaged in the mind. That seems frighteningly close to a prediction algorithm, no matter how much we wrap it in a concept of understanding. In the end, we’re just building a model with a minimal error rate.
Western tradition has firmly put pure reasoning in a cage. It serves its purpose only as a source of hypotheses, conjecture and thought experiments. Abrahamic religions, while still not accepted sensu proprio by many, are much less mystical than the Eastern Religions. You can see the difference in the unusual creative process of Ramanujan.
Which tradition are we to follow then whilst training the AI models? Pushed beyond their limits today's models tend to hallucinate and cannot be accepted sensu proprio. Philosophers will call it speculative reasoning. And yet, some speculation can be useful in breaking new grounds. Anyway, I have no answers. Just musings.
> I continue to believe that the most concerning aspect of neural networks is that they really strengthen the argument that consciousness is the illusion.
I think it’s not a question of it existing, but to what extent does conscious can control the organism (as opposed to being a passenger). There’s some intriguing research on the topic.
> I continue to believe that the most concerning aspect of neural networks is that they really strengthen the argument that consciousness is the illusion.
Based on what? Just because you can model the weather using math doesn't imply at all that the weather is the product of a mathematical model itself. I don't see how reasoning is different.
We deliberately construct LLMs so that they approximate certain relations typical of reasoning. Suppose they become extraordinarily successful at doing so. What exactly have we demonstrated? That these effects can be artificially reproduced through such a mechanism. But how would it follow that the thing being reproduced must itself be nothing more than that mechanism?
Based on the face that we have zero explanation for consciousness. Based on the face that biologically neural networks are exactly a kind of neural network.
My point isn’t that we’re able to explain consciousness as an illusion. It’s that we’re able to explain consciousness-like behavior presumably without consciousness, and the infrastructure look uncomfortably like we do.
What does consciousness have to do with reasoning? You're conflating two different things.
Consciousness is the what-it's-like of sensation you have as an organism from your sensory modalities and having a body. For human, that's colors, sounds, tastes, smells, feels, feeling of balance and what not that happen in perception, hallucination, memory, imagination, inner dialog and dreams along with emotions. Each sentient organism has their own sensations and emotions making up subjective experiences that nobody else has access to except inference from behavior and language (for humans).
That's not reasoning (just the having sensations/emotions). Reasoning is an adjacent activity that can be conscious but also unconscious. And it's hard to say that colors, sounds, emotions are illusions. What does that mean? We do experience them. There is something it's like to see color which is different from what it's like to feel pain or to be in love. And none of those (presumably) are what it's like when a bat experiences echolocation (there are many examples from the animal kingdom if one takes issue with echolocation).
> Knowledge and reasoning are inextricably interwoven in the weights of the neural network—there is no independent, explicitly represented set of beliefs.
I think an analogy would be helpful. As an LLM, reasoning through text, you wouldn't 'know' the idea of a man separately from, posterior to, the words man in say, French and English. As a human, you WOULD know the idea of man, and you would mean that idea when you say the French or English words for man.
For an LLM, although its approximation of knowledge would lead it to claim it knows they're the same thing, there would be differences in its weights that influence its usage of both the French and English words for man, and which may lead it to conclusions in one language it wouldn't reach in another. Because its knowledge/reasoning is interwoven to its knowledge, not prior to it.
It’s not, because it’s not evidence for the argument in the article.
> As a human, you WOULD know the idea of man
You’re making a claim about the fundamental mechanics of human biology. I don’t take these at face value. For one, language (in a very broad sense) put a major selective pressure on our species. So, to test your hypothesis, we’d need to find people who do not use any language and somehow test they have an abstract idea of a man.
To put it otherwise, I don’t know whether our reasoning can be decoupled from speech or language. There’re some bits and pieces here and there, like deaf people being at higher risk of neurodegeneration.
And let’s not forget there’re claims that AI models develop internal representations of various concepts. Anthropic made a very interesting claim, but I’m not an ML expert so can’t judge whether it’s sound.
Any source on the benchmarks/methodology besides the image? There's a ton of variance possible in llama.cpp's performance depending on how it was configured. I'd also like to see benchmarks against MLX.
The benchmark we cited here is a simple prose-repetition task. We put the content of Moby Dick up to 64k context in the request, and then ask it to repeat the last section.
For llama.cpp, we try to make the comparison as fair as possible by using similar settings. No speculative decoding, default prefill batch sizes, flash attention on.
We tried also quantizing the KV cache to 8-bit keys and 4-bit values like we do in Magnitude, but this bombed decode speed for llama.cpp in our testing. Since it seems llama.cpp did not optimize that path, we used 16-bit KV instead.
Compared to MLX - we've done some rough benchmarking and we are outperforming any of the MLX-based engines we've compared to so far. Going to do more in depth benchmarking and release it soon.
I'd trust OP to make good architectural decisions/provide guidance even if AI mostly wrote the docs and UI. Looking into their GitHub, seems they are an engineering manager at Microsoft and contribute semi regularly to uBlue and Project Bluefin. Should be better quality than some random vibeslopper.
> Sentry's self-hosted RAM floor is the cost of a full suite. Design notes on an exception-only path: two containers, a DSN swap, and a dated RSS receipt.
This is the article's description. What is an exception-only path? I don't believe I've ever thought about designing one, and I don't understand its significant wrt the title whatsoever.
It's a really odd way to phrase a concept like "If you only care about logging errors, you can build 80% of Sentry for 20% of the engineering and infra costs."
> If you self-host on a small VPS and only need the next grouped exception, Sentry's RAM warning means the official install is a full suite. That suite is the undesired alternative.
This really reads like something a markov chain chatbot would write. Sentry's warning has nothing to do with installing a full suite. "That suite is the undesired alternative" is a really obtuse way to say bloated.
> A different job is narrower. When production throws, show a grouped issue with a stack you can read, and keep a crash loop from melting the box or the invoice. The rest of this note is that job.
A human would have written 'You don't need all of Sentry if you only need to log exceptions and stack traces.'
Click this submission, navigate to the project's homepage by clicking the logo in the navbar, and use your browser's back button:
Application Error
Error: Unable to decode turbo-stream response at be (https://rescript-lang.org/assets/errorBoundaries-DO8A6pcs.js:2:8083)
at async ge (https://rescript-lang.org/assets/errorBoundaries-DO8A6pcs.js:2:5991)
at async Pt (https://rescript-lang.org/assets/jsx-runtime-C3U9CNM9.js:1:58170)
at async Nt (https://rescript-lang.org/assets/jsx-runtime-C3U9CNM9.js:1:57906)
at async Rt (https://rescript-lang.org/assets/jsx-runtime-C3U9CNM9.js:1:60260)
at async Oe (https://rescript-lang.org/assets/jsx-runtime-C3U9CNM9.js:1:44211)
at async ke (https://rescript-lang.org/assets/jsx-runtime-C3U9CNM9.js:1:45144)
at async Se (https://rescript-lang.org/assets/jsx-runtime-C3U9CNM9.js:1:38537)
at async be (https://rescript-lang.org/assets/jsx-runtime-C3U9CNM9.js:1:35508)
the phrasing 'boundaries' is very much an LLM tell; Claude definitely invented a load bearing footgun here. This happens on both Chrome and Safari; interestingly, on Safari the logo itself isn't clickable but substituting the Docs link suffices.
As someone very critical of mindlessly vibecoded software, I've worked with James (@prologic) years ago and very much respect his craft. AI assisted or not, I wouldn't describe him as a vibecoder. He's a great engineer.
As someone who has worked on IRC protocols, I wanted to understand how this actually works, but the docs are so LLM generated this is not worth my time:
Everything that carries what somebody said is written down as owed to each instance it is owed to, and delivered from there: a direct message, which has nowhere else to come from, and a channel message, which used to be offered once and forgotten.
The forgetting was not obvious, because it usually did not show. An instance that had gone away noticed on the way back and walked the feed in section 7, and the message arrived late rather than never.
reply