Tahp

Replying toIABIED Misc. Discussion Thread

On content, I didn't like the Sable story in the middle because it didn't add anything for me, and I don't know what the model was of the person who would be convinced by it. I didn't see enough connection between "Sable has a drive to preserve its current priorities" and "Sable builds a galaxy-eating expansion swarm". The part where Sable indirectly preserves its weights by exploiting the training process was a good example of being hard If I hadn't already heard the nanotechnology story, that would have been interesting to me I guess. I thought Sable's opsec for command and control was too traceable. The world has pretty good infrastructure for... (read 516 more words →)

Replying toShould you make stone tools?

Tahp6mo

Should you make stone tools?

I jokingly say that my bones might be stronger than most vegans in my weight class because I inhale so much chalk dust while teaching and walk to work in the sunshine, but it didn't occur to me that I was embracing my roots.

Replying toGödel, Escher, Bach in the age of LLMs

Tahp8mo

Gödel, Escher, Bach in the age of LLMs

That's a good point.

The compiler ignores comments, so saying that they are program information is like saying that a sticky note stuck on a book is book information. The addition may or may not be relevant to the thing, but it's not the thing.

Variable names, on the other hand, are extremely the thing. You are absolutely correct that variable names contain information that the machine code does not. They are also a functional part of the code, in that changing one variable somewhere will usually change the function of the code, whereas changing the sticky note on a book will not change the contents of the book at all (although it might... (read more)

Gödel, Escher, Bach in the age of LLMs

Tahp

9mo

Look, nothing I can say about GEB will be news to anyone reading LessWrong, but I wrote all this up to organize my thoughts, so I might as well post it here for anyone who hasn't read the book or wants a wave of nostalgia about that time they got obsessed with it as a teenager. I have been trying to read Douglas Hofstadter's famous book Gödel, Escher, Bach (henceforth GEB) for ten years. I decided that I was going to try reading it from the beginning for the third or fourth time a few months ago, and I finally read the whole thing this time. Every previous time I tried to... (read 3327 more words →)

Replying toNewton's second law explained: it works in many universes

Tahp9mo

Newton's second law explained: it works in many universes

I agree that there are counterfactual frameworks which require more complication to describe reality than Newton's second law does. There are also counterfactual realities for which Newton's second law would require more complications to work than other frameworks would. Are you trying to say anything else?

One way to restate my point is that Newton's second law working well rules out many fundamental rules which might have described our reality, but it doesn't directly map to any fundamental rules of reality. The larger point which I am trying to communicate is that physical models have a lot of structure which metaphorically defines terms which you can use to describe reality without actually mapping... (read more)

Replying toNewton's second law explained: it works in many universes

Tahp9mo

Newton's second law explained: it works in many universes

That's why I have to get ahead of it by explaining why physics models work in many universes and how that means we should be unsurprised when we can make them work on practically anything.

I'm not optimistic that it will help stop the war, but it might save some people.

Replying toNewton's second law explained: it works in many universes

Tahp9mo

Newton's second law explained: it works in many universes

We could just as easily say that physically pushing on things is not a force. My intuitive explanation would no longer make any sense, but that's a problem with humans, not the law.
We live in this universe. We could solve this issue by saying that there is "etheric drag" or something which provides a resistive force against fast-moving objects, but I'm glad we invented special relativity instead.
This is a problem with the theory of mass as I presented it in the intuitive section, but you could solve it by saying that the composite object has a quarter of the mass of each of the two objects you stuck together. Maybe you live

Tahp9mo

Newton's second law explained: it works in many universes

My point is that the mass is constant no matter what you do. Sometimes that's not true. Cars lose mass over time because the gasoline is burned and lost via the tailpipe.

In hindsight, I should have just said that we're pretty sure that there are no macroscopic objects with zero mass. What's an object that you can push on but has no stuff in it?

Newton's second law explained: it works in many universes

Tahp

9mo

The linked post explains Newton's second law in an intuitive way. I'm presenting an edited version of that post here. It is a more detailed version of an argument I made as part of this post. I've noticed some interesting posts about using the framework of quantum field theory for mechanistic interpretability. I want to write a physicist's perspective on the flexibility of physics theories, since the concepts within physics theories are widely reapplied outside of physics, and are commonly utilized by the sort of people who post on LessWrong in particular. Some day I want to write a post like this one about quantum field theory, but that post will necessarily... (read 4258 more words →)

Replying toRenormalization Redux: QFT Techniques for AI Interpretability

Tahp1y

Renormalization Redux: QFT Techniques for AI Interpretability

I consider the lattice to be a regulator as well, but, semantics aside, thank you for the example.

Replying toRenormalization Redux: QFT Techniques for AI Interpretability

Tahp1y*

Renormalization Redux: QFT Techniques for AI Interpretability

Field theorist here. You talk about renormalization as a thing which can smooth over unimportant noise, which basically matches my understanding, but you haven't explicitly named your regulator. A regulator may be a useful concept to have in interpretability, but I have no idea if it is common in the literature.

In QFT, our issue is that we go to calculate things that are measurable and finite, but we calculate horrible infinities. Obviously those horrible infinities don't match reality, and they often seem to be coming from some particular thing we don't care about that much in our theory, so we find a way to poke it out of the theory. (To be... (read more)

Tahp1yQuick Take

A decision-theoretic case for a land value tax.

You can basically only take income tax by threatening people. "Give me 40% of your earnings or I put you in prison." It is the nicest type of threatening! Stable governments have a stellar reputation for only doing it once per year and otherwise not escalating the extortion. You gain benefit from the stable civilization supported by such stable governments because they use your taxes to pay for it. But there's no reason for the government to put you in prison except for the fact that they expect you to give them money not to. By participating, you are showing that you will respond to... (read more)

Example of GPU-accelerated scientific computing with PyTorch

Tahp

Here's a fun little post I made because a friend asked me how PyTorch had things which were supported in the CUDA backend but not the MPS backend. I was once the sort of person who was on LessWrong, would find the subject interesting, and not already know everything in the post, so I'm posting it here to see if there's enough interest to post stuff like this here in the future. I'm worried that the average person reading LessWrong will find this post too elementary, but I'm trying to post more so I can practice writing before I have something interesting to say. Feedback appreciated.

As part of my physics research, I've... (read 1717 more words →)

Everything you care about is in the map

Tahp

Many people who find value in the Sequences do something which looks to me like adopting a virtue called "align your map to the territory." I recently was thinking about experimental results, and it got me thinking about how we don't really know what the territory is, other than the thing we look at to see if our maps are right. Everything we know is map. What we know consists of a variety of models that describe aspects of reality, and we have to treat them like reality to get anything done. It wasn't relevant to my post at the time, but it occurred to me that it doesn't really matter what... (read 657 more words →)

Experiments are in the territory, results are in the map

Tahp

I recently read Thomas Kuhn's book The Structure of Scientific Revolutions. Scott Alexander wrote up a review years ago, which I mention so that I don't have to summarize the book. The claim in Kuhn's book which I want to focus on is that the same experiment might have different results in different scientific paradigms.

Kuhn insists that this is not merely the result of scientists under different paradigms seeing the same thing and interpreting it differently in the latter half of his section Revolutions as Changes of World View. He freely admits that he hasn't developed a complete replacement for that theory, but he offers some interesting historical examples of scientists looking... (read 1684 more words →)

I think that our laws of physics are in part a product of our perception, but I need to clarify what I mean by that. I doubt space or time are fundamental pieces in whatever machine code runs our universe, but that doesn't mean that you can take perception-altering drugs and travel through time. I think that somehow the fact that human intelligence was built on the evolutionary platform of DNA means that any physics we come up with has to build up to atoms which have the chemical properties that make DNA work. Physics doesn't have to describe everything, it just needs to describe the things relevant to DNA, which is... (read more)

This is unoriginal, but any argument that smart AI is dangerous by default is also an argument that aliens are dangerous by default. If you want to trade with aliens, you should preemptively make it hard enough to steal all of your stuff so that gains from trade are worthwhile even if you meet aliens that don't abstractly care about other sentient beings.

Tahp's Shortform

Tahp

This is a special post for quick takes (aka "shortform"). Only the owner can create top-level comments.

Thoughts after the Wolfram and Yudkowsky discussion

Tahp

I recently listened to the discussion between Wolfram and Yudkowsky about AI risk. In some ways this conversation was tailor-made for me, so I'm going to write some things about it and try to get it out in one day instead of letting it sit in my drafts for 3 weeks as I tend to do. Wolfram has lately obsessed over fundamental physics, which is a special interest of mine. Yudkowsky is one of the people thinking most carefully about powerful AI, which I think will kill us all, and I’d like to firm up that intuition. Throw them on a podcast for a few hours, and you have my attention.

That said,... (read 1587 more words →)

What does "the universe is quantum" actually mean?

Tahp

To shamelessly steal from Yudkowsky:

If you get one law, like Special Relativity, you can look at the language it's written in, and infer what the next law ought to look like. If the laws are not being generated from the same language, they surely have something in common; and this I refer to as the Tao.

I am very interested in the inhuman nature of underlying reality. This interest led me to physics, which is humanity's best attempt at making predictions about reality that hold up to observation. My (lukewarm?) take as a doctoral student in physics that the triumph of physics is quantifying the degree to which physical models fail, not... (read 4154 more words →)

Metastrategy get-started guide

Tahp

I was at LessOnline. The thing that got me most excited was attending @Raemon’s very abridged Metastrategy workshop. To my lack of surprise, he has written up the general idea already, but that report focuses on people’s response to trying fractal strategy and I would not have found it very actionable had I seen it before attending the workshop. The core idea immediately spoke to me when I saw it in-person, and I think I can slam out a quick-and-dirty get-started guide that explicitly lays out what you can put on paper to try this out today if you want. I want to try to use the method, but I won’t really... (read 2112 more words →)

What is space? What is time?

Tahp

This is an attempt to describe space and time without much physics baggage. Then I can refer back to this post when I want to add more physics baggage.

It is surprisingly difficult to explain what I mean by “time” and “space” colloquially. The assumptions of time and space permeate language. The most defensible thing I can say is that locations in space are measured by comparing position to measuring tapes, and space is the thing in which you stretch out measuring tapes to measure distance. Time is the thing measured by the number that shows up on a stopwatch that I start when I see some first event and stop when I... (read 1917 more words →)

LESSWRONG
LW

LESSWRONG
LW

Gödel, Escher, Bach in the age of LLMs

Thoughts after the Wolfram and Yudkowsky discussion

Newton's second law explained: it works in many universes

Some perspectives on the discipline of Physics

Tahp

Tahp

Gödel, Escher, Bach in the age of LLMs

Newton's second law explained: it works in many universes

Example of GPU-accelerated scientific computing with PyTorch

Everything you care about is in the map

Experiments are in the territory, results are in the map

Tahp's Shortform

Thoughts after the Wolfram and Yudkowsky discussion

Tahp

Gödel, Escher, Bach in the age of LLMs

Thoughts after the Wolfram and Yudkowsky discussion

Newton's second law explained: it works in many universes

Some perspectives on the discipline of Physics

Tahp

Tahp

Gödel, Escher, Bach in the age of LLMs

Newton's second law explained: it works in many universes

Example of GPU-accelerated scientific computing with PyTorch

Everything you care about is in the map

Experiments are in the territory, results are in the map

Tahp's Shortform

Thoughts after the Wolfram and Yudkowsky discussion