Thank you for your perspective and I agree as repetitive as it is, I hope hn can continue to surface disagreement/substantial views on a topic.
These comments don't stay up much anymore and I can't tell if it's structural to the forum (flag weight + statistical mechanics of votes + guidelines) or if it's the userbase sentiment.
Except for now HN doesn't do that. Every single comment that points the issue with Grok being built by a pro-fascist team is being "locked" so that they can't be upvoted
For what it's worth, I've watched the vote fight on my comment swing back and fourth over the last hour. It got flagged, unflagged, and flagged again. It appears to now be locked and dead.
But I think it represents real malaise in the community. It's not a moderator plot, people here really just don't care and might even support this.
We really are in the minority of opinion for giving a damn about liberal democracy.
I think I was trying to bring up the mechanics of the forum as another reason.
Between the guidelines + user thoughts (e.g. repetition, low novelty/new info), there's other reasons these types of replies might end up dead.
I am worried that it leads to people self selecting to other forums biasing the remaining userbase vote/vouch/flag distributions. In an exit vs voice situation, the voice kinda dies out. Then we end up other-izing people and homogenizing our communities.
But I concede it's also possible that the minority opinion issue could be the core driving force.
Do you have examples of sites that you prefer that achieve this, or have a relatively high signal to noise ratio that also have a high volume of discussion? (preferrably not an individual's blog + comment section)
It seems like the docs[0] are a better explanation? The comparison to llm tokens is kinda confusing.
It looks like the model takes as input a state (structured text? not sure if multi-modal) and a question (as a "Choice", "Score", or "Noul") with some additional augmentations possible. Then outputs the question's answers as appropriate (e.g. a choice, accompanying probabilities, confidence).
Edit: On the AI primer page, it looks like they do the RLCD on a pre-trained base model?
I do agree that the comparison to LLM tokens is hard to understand (also because output tokens are not comparable).
But yes, text or structured state (like a JSON with multiple pieces of text in) -> decisions out (e.g. choice maps to "match" statement, "score" maps to sorting, "noul" short for bernoulli maps to if-statements)
Small request, can we get an explanation of the naming of "noul" in the docs[0]. I tried googling, and searching the docs and didn't understand why it was called that.
(I'd also argue something like p_yes or just probability might be a simpler name, but I'm sure there's a better reason behind Bernoulli maps).
here is how I attempted to explain it to my company's AI group chat, is this roughly accurate?
"instead of autoregressive string output it instead outputs structured type-safe 'decisions' with probabilities/confidence scores, each generated in parallel
so sort of more like a Large Classification Model than a Large Language Model? or, maybe better to think of it as a sort of "shift left" in the LLM's transformer architecture, allowing you to replace the predefined token vocabulary of an LLM with a prescribed set of 'decisions' that need to be made based off the input context; and exposing those probabilities directly so they can be integrated into the system logic, instead of just sampling from top-K.
all of this while still being instruction-tuned (!!!)"
It's always been possible to build classification pipelines using LLM embeddings as the input. seems like this is a much more sophisticated / useful application of that concept
the one nuance I'd get into is I'd call it "zero-shot" over "instruction-tuned" (the latter often implies a particular distribution), but very safe for sharing
For many day-to-day computing use cases, Jev seems far better suited than an autoregressive language model, if for no other reason than it is not wasting compute thinking about anything other than how to spit out a decision.
Do you have an architectural explainer yet for Jev or are you holding that close to your chest and letting the magic rip for now?
Hi - first congratulations, System One looks really promising.
The Doom demo really help me, at least, to understand how System One differs from LLMs. However the first demo (Side-by-side demonstration) - I'm struggling to understand what is going on here!
The demo is showing System One producing its output in parallel very quickly and for little cost compared to an LLM generating its answers token-by-token. The "noul" type is used to evaluate a yes/no question and return the probability that the answer is yes.
So this demo is showing System One offering much more nuanced responses and specific probabilities compared to an LLM's more crude responses (e.g. LLM shows "true" or "false" compared to "0.9" or "0.07" probabilities that the answer to some question is true).
I see this super interestingly as the "subconscious" to the llms "conscious" for lack of better terms. I'm super interested in this for broad and rapid decision making in the context of consumer agents so will be signing up for sure.
1. I am extremely on the same page
2. I do think that subconscious is not only much smarter than we give it credit for, but also much more robust than the "jagged frontier" of current LLMs
You really, really, really, really, REALLY REALLY REALLY should explain neologisms in your documentation! Really really! Having "here's a weird unexplained difficult-to-guess-impossible-to-Google-for term" in your documentation is aggressively user-hostile. Please don't.
The actual term is fine. Lack of an inline explanation is a blow-it move. Don't do that.
> the model takes as input a state (structured text? not sure if multi-modal)
Input, and criteria/instructions can both be defined as structured input (JSON). This ends up being pretty powerful because the model is trained to understand structure.
I assume this isn't really for consumers/individuals currently? Kinda feels like an improved magic 8 ball.
I can't really intuit how I should think about when the model will be accurate. Is there somewhere to read more about that? I assume customers would just have some tests or talk to you.
> I assume this isn't really for consumers/individuals currently?
Unless they're hackers, no. It's not really a chat interface, it's meant for consumption by machines and composing into higher level systems (pairs great with LLMs).
> Is there somewhere to read more about that? I assume customers would just have some tests or talk to you.
We're going to release some more info on evaluations over time, and yeah, join the waitlist! We offer faster access in exchange for good memes
I'm confused by this because I was involved in various distributed computing projects from about 1997 to about 2001 (as a person running compute notes for them) and from about 2001 to 2019 (as a person helping to administer a distributed computing related prize), and in the early part of that era we routinely talked about idle computer power as "wasted" because of the idea that the computer might as well be used to compute something rather than sitting idle. This may have been very credible in 1990s devices that consumed a roughly comparable amount of power regardless of what specific computation they were performing, but all modern devices have extremely variable power consumption depending on the load. You can easily feel this as devices have fans turn on or get hot when the CPU is loaded, and in many cases you can easily query the CPU with software to find out how its power consumption or clock rate or other factors get adjusted based on computational load.
This means that the idea that idle compute would have gone to waste is just no longer true on modern devices.
Now there is certainly compute that couldn't be sold to a paying cloud customer because it's too fragmented in some sense, but it still has some amount of energy cost, and, in a data center, corresponding cooling cost attributable to the marginal heat production. How can one actually say that there is literally no marginal cost at all? I just can't imagine a device that literally has the same power draw regardless of load factor!
If it's cloud compute that would be otherwise unused, the business is not paying for the electricity or wear. And they specifically put it in terms of marginal cost - Yeah there'd be an improvement from reselling this idle compute instead, but just using the cycles that would otherwise have been wasted doesn't change the status quo.
In the sense that the company paying to reserve a certain quantity of computer time isn't the owner or operator of the machines providing it, and has paid for that time by the hour, or something?
If I could manually set/select locations this might be perfect for me too.
I wonder why the scroll performance of every vibe coded app is the same kind of laggy. I assume they're defaulting to the same frameworks or implementations? Or the verification loops in training environments haven't prioritized frame rates/dropped frames yet.
“For example, if a philosophical zombie were poked with a sharp object, it would not feel any pain, but it would react exactly the way any conscious human would”
Doing non-verbal thinking has nothing to do with not feeling pain, I don’t think it sounds like philosophical zombies at all.
Maybe, I wouldn’t know. I just find it fascinating that other people have such a radically different way of experiencing thought. No idea what it’s like.
I wanted to say something dismissive. It feels like a lot of doublespeak, which I think undermines trust in them.
But I do understand that it's a little hard to characterize the amount of compute they're selling while obfuscating the business model enough that it still works. Demand, costs, supply, capabilities, competition change wildly week to week.
Selling a standardized "token" via the API with gas-like pricing works a lot better for transparency. But I think there's a world where the ambiguous subscription pricing is net beneficial for users and providers. Allowing some degree of price discrimination should benefit everyone in aggregate, assuming there's real competition and not yet another duopoly..
M5Stack's twitter says 2 to 3 months [0] to restock because they're out of eink driver ic.
When I submitted the story maybe 8 hours[1] after the announcement, it was already out of stock. This person says it sold out in 1 second - https://x.com/henry19840301/status/2090652363434647738 - if that's possible.
Maybe "one second" was a poorly phrased colloquialism meaning "right away".
One can sign up for new product releases and if I paid attention to my marketing emails I could have clicked through. I'm guessing there are enough people who do combined with limited initial stock. We don't know.
Thanks for passing through the info for those of us not on Twitter!
I think the signal on X has been deteriorating for years. Many valuable communities that existed purely on Twitter have scattered across the internet after Elon's transformation, often to many disconnected small communities across X, Mastodon instances, BlueSky, and reddit.
The labs might not be open. But they would want access to data. More open platforms (the web at large, but bluesky in particular has made open access an explicit goal) tend to provide that, relative to x/threads/facebook/reddit/youtube.
Some labs will be able to buy access. Not every platform will offer it. There are many more labs than the big ones.
These comments don't stay up much anymore and I can't tell if it's structural to the forum (flag weight + statistical mechanics of votes + guidelines) or if it's the userbase sentiment.
reply