Xenograph

Is this test available somewhere? Would like to test it out on my models.

Is this with browser tooling attached to the agent for review/iteration?

> I know many are scared off by the closed bootloader of the teensy (though I feel it's a fair thing to do).

Why is it a fair thing to do?

In "yiwei"/"creer" case, Juan believes that they are going to promote him (but his belief is not very well calibrated and is likely false). yiwei/creerse asserts something about the truth value of the belief, in addition to what the belief is.

In the "doubts" case, Juan believes that they are not going to promote him. There is no assertion regarding the truth value of that belief.

That makes sense, wasn't sure if it was as simple as tweaking those two numbers or not, thanks for sharing!

If there's any insight you can share about your AGENTS.md prompting, it may also be helpful for others!

Have you tried continue.dev's new open completion model [1]? How does it compare to llama.vscode FIM with qwen?

[1] https://blog.continue.dev/instinct/

Would love for you to share the Codex patches you needed to make and the AGENTS.md prompting, if you're open to it.

Which models do you enjoy most on your 4090? and why vLLM instead of ollama?

I don't think it's the heat, rather, the wasps seem attracted to the radio wave EM radiation. I'm not sure why, but I've seen it first hand when turning on high powered outdoor point-to-point WiFi antennas (eg. airMAX LiteBeam 5AC). The moment I connected the power, several wasps immediately began crowding around the antenna.

Why is GC problematic for web servers?

I put my tri-ply stainless pans in the dishwasher all the time.

Why shouldn't I?

We're talking about normal dish soap for hand washing (eg. Dawn).

Dishwasher soap is a whole different story. It contains abrasives, among other things.

KVM switches are often slow and/or buggy and/or expensive. Plus, looking glass is more flexible since you can display both OSs at the same time.

As a sibling comment stated, you can get a dummy monitor plug on Amazon for like $5 which fools your GPU into thinking there's a monitor attached.

It's in the title of the post. You can run your VM in a regular window but at high performance (think gaming and professional stuff that's GPU heavy). You otherwise can't do this without a hardware KVM switch (or tossing the VM and just dual booting). Hardware KVM switches are expensive and often don't work very well. And still, with a KVM switch it wouldn't be running inside a window. It would consume your whole screen.

What exactly is limiting about it? I use it on a daily basis in order to flip between high performance Linux and high performance Windows instantly with the flick of a hotkey.

I use it for gaming, software dev, and just in general it's nice to be able to switch OSs for any reason instantly. What better options are there?

Just want to say that gnif is amazing. He is always happy to help out people on the looking glass support discord. Plus he's absolutely dedicated to continuing to improve the software he writes and the ecosystem around it. Top notch guy.

I think you forgot to mention why you didn't like it.

Let me rephrase: this is a good thing as long as production is meeting or exceeding demand.

When it becomes clear that there aren't enough hands on deck (or at desks, or in factories) to "keep the lights on" then we can say there's a problem.

Until then, it's good for people to have more financial independence and be able to find a job that doesn't treat them like cattle.

As you mention, this breaks down when you take it to an extreme (eg. 200 mil). But that's not where we are and nor is it where we're headed so I don't see the equivalence.

The first step in conducting scientific inquiry is formulating hypotheses, which are typically based on intuition and anecdote.

This blog post is clearly telling an anecdote from the author's own personal life experience. From these experiences they have formed their hypotheses about behaviors that may affect intelligence. This blog post presents their experiences and conclusions.

There's nothing unscientific about this. Perhaps you think that it is important for the author to attach a disclaimer that this blog post is not asserting definite scientifically proven truths about reality. But it's fairly obvious to me that the author was speaking from their own experience and opinions rather than attempting to share scientifically proven facts (why even write a long blog post to do that? The research would speak for itself).

That explanation sounds circular to me - by definition any word you'd use in that way is an adverb.

A more intuitive explanation is perhaps that "overseas" came from the combination of two words, "over" (a preposition), and "seas" (a noun). And when you write it out as three words "going over seas" it makes perfect grammatical sense.

Disclaimer, I have no idea if my explanation is rooted in facts about the English language or how it actually developed. But that's how I'd answer the question if someone asked me for an intuitive explanation about why we don't use "to". Because there's already a preposition ("over")! You wouldn't write "going to over seas", would you?

> I used all the extra earnings to secure a moderate income outside my job.

How? You make it sound straightforward, but I wouldn't know where to begin.

> But I mean, normal people also don't really understand or believe in Bitcoin?

I don't think that's really the case anymore. I've heard a lot of positive sentiment from "normal" people about BTC lately, probably because its been all over the mainstream media due to it seeing such strong gains.

People may not understand Bitcoin, but most everyone understands "asset that is seeing massive gains, making people rich, being adopted by companies around the world, and seems like it may continue to grow".

Honest question: what would cause a downturn in the housing market? People will always need/want houses, right?

While I appreciate the intention, I think it's a bad analogy.

Can you think of a small piece of machine code that you could inject into arbitrary computers that would produce useful effects in some significant proportion of the computer population? (Don't think so...)

Computers and biological systems are so wildly different. Although perhaps you'd find better analogies comparing brains to machine learning systems. Eg. "Trying to find out how ADs work is like applying some simple numerical transformation on some particular part of a deep neural network and trying to find out what changes are produced".

Oddly enough.. I think ML researchers often do find little "tricks" that improve the performance of neural nets without being able to fully grasp how or why. It's somewhat similar to how we use and understand ADs.

It is market manipulation, of course, but it's not the kind that's illegal.

Now it's mostly folks worried about their 401k.

Not that I disagree necessarily, but to play devil's advocate: kids don't learn any formal grammar rules before they speak their first language fluently.

The point is that all the data is centralized in a single database in this case.

People being employed doesn't imply that they aren't working jobs which needn't exist.

Not really - I think you missed the point of the original claim, it wasn't meant to be interpreted literally. In this case, "not enough work" does not translate to "not enough jobs". It translates to tons of useless jobs that employ people to count beans and create powerpoint presentations.

Perhaps the original claim would've been better phrased as "not enough real/useful/impactful work".