All of CstineSublime's Comments + Replies

Off the top of my head it's because people are weary of Chesterton's Fence/Sealioning (feigning 'just asking questions' when actually they have an agenda which they mask with the plausible deniability of feigning naive curiosity) and as you say - the topic being sensitive so it generates a 'ugh field' are two pillars of what makes certain topics difficult to discuss.

I've noticed this pattern on a lot of, usually political topics but it could also be some kind of interpersonal drama/gossip, someone asks a you question which appears to be an invitation to ge... (read more)

That Nixon one really wow'd me, the fact that it exaggerated his jowls but after a bit of google searching it seems like other models also seem to have been trained on the Nixon caricature rather than the man himself.

I'm also a big fan of that Fleischer style distracted boyfriend remix.

Never the less, the ease of 'prompting' if that's what you can even call it now is phenomenal.

I'm looking at this not from a CompSci point of view by a rhetoric point of view: Isn't it much easier to make tenuous or even flat out wrong links between Climate Change and highly publicized Natural Disaster events that have lot's of dramatic, visceral footage than it is to ascribe danger to a machine that hasn't been invented yet, that we don't know the nature or inclinations of?

I don't know about nowadays but for me the two main pop-culture touchstones for me for "evil AI" are Skynet in Terminator, or HAL 9000 in 2001: A Space Odyssey (and by inversion - the Butlerian Jihad in Dune). Wouldn't it be more expedient to leverage those? (Expedient - I didn't say accurate)

I want to let you know I've been reflecting on the reactive/proactive news consumption all week. It has really brought into focus a lot of my futile habits not just over news consumption, but idle reading and social media scrolling in general.^[1] Why do I do it? I'm always hoping for that one piece of information, that "one simple trick" which will improve my decision making models, will solve my productivity problems, give me the tools to let me accomplish goals XY&Z. Which of course begs the question of why am I always operating on this abstract... (read more)

I think you're right. Although I'm having a hard time expressing where to draw the line between a simile and a analogy even after glancing at this article; https://www.grammarpalette.com/analogy-vs-simile-dont-be-confused/ 

Thank you for sharing that, it is interesting to see how others have arrived at similar ideas. Do you find yourself in a rhythm or momentum when sprinting and shooting?

Bad information can inform a decision that detracts from the received value. I suppose if it is perceived to be valuable it still is a useful term - do you think that would get the point across better?

I'm interested in how you can convert that information proactively?

I'm aware that, for example, keeping abreast of macro or geopolitical changes can influence things like investing in the stock-market. But I'd be lying if I'm aware of any other possibilities beyond that.

I think that, more than drinking from the propaganda trough makes me an NPC, protagonists in games do novel things, potentially unexpected (from the perspective of the game designers). NPCs are predictable and habitual. If I cannot extract utility from the news, macro or micro, then I fear ... (read more)

We have Shannon Information, Quantum Information, Fisher Information, and even Mutual Information and many others. Now let me present another type of information which until I find a better name will certainly be doomed to reduplication induced obscurity: Informative Information.

One of the many insightful takeouts from Douglas Hubbard's Book - How to Measure Anything for me was that if a measure has any value at all then it influences a decision. It informs a decision.

If I see a link come up on my social media feed  "5 rationality techniques you can u... (read more)

Tractability, what is tractable to a world government is different to what is personally tractable to me. Then the tractability of the news increases based on how many actions or decisions of an individual reader the news can inform or influence. I cannot change macroevents like wars, but they may influence my personal decision making.

This of course opens the door to counterproductive motivated reasoning. For example of a top-of-mind news story: the Palisades fire - can I stop the fires? No. But maybe I can donate something to those who were displaced? Tha... (read more)

Or, if in your real life work you find something took a noticeably long time to figure out, or you were surprised about something you might have been able to notice.

Can you detail what kinds of problems "in your real life" you find might be better served or less appropriate to this exercise? Just off the top of my head, would forgetting who was the star of a movie you'd expect to remember and having the name on the tip of your tongue for an hour not be suitable? But what general code debugging, say of a fragment shader where you finally realize by flipping the x,y coordinates it starts to "look right" - is that more appropriate?

I often ask myself and others "okay, but how does that look in practice?" - this is usually when I have a vague idea about something I need to achieve a goal, but also when someone gives me some vague advice that I feel is leaving it to me to "draw the rest of the owl."

Is this the best phrasing of the question? I have my doubts. 

Firstly, is it too generalized for different domains?

"I should really organize my dresser drawers more thematically" -> "okay, but how does that look in practice?"

"I need to make more of an effort to promote my freelancing"

You might instead just directly study filmmaking.

Absolutely not. I cannot stress this enough.

Edit: I just saw your other comment that you studied filmmaking in college, so please excuse the over-explaining in this comment stuff that is no doubt oversimplified to you. Although I will state that there is no easier time to make films than in filmschool where classmates and other members of your cohort provide cast and crew, and the school provides facilities and equipment removing many of the logistical hurdles I enumerate.

So, (I mean this as an earnest quest

If I'm playing anagrams or Scrabble after going to a church, and I get the letters "ODG" I'm going to be predisposed towards a different answer than if I've been playing with a German Shepard. I suspect sleep has very little to do with it, and simply coming at something with a fresh load of biases on a different day with different cues and environmental factors may be a larger part of it.

Although Marvin Minsky made a good point about the myth of introspection: we are only aware of a think sliver of our active mental processes at any given moment, when you ... (read more)

I really like the fact that there's an upvote feature together with a separate agree/disagree feature on this site.

I may like the topic, I may want to encourage the author of the post or comment to continue exploring and opening up a dialogue about that particular topic. I might think it's a valuable addition to the conversation. But I may just not agree with their conclusions.

It's an important lesson: failure can reveal important information. You don't have to agree with someone to feel richer for having understood them.

On the other hand, I'm also guilty ... (read more)

Can you elaborate on why you think such vague feedback is helpful?

It's apparent I've done a terrible bad job of explaining myself here.

What is my immediate goal? To get good at general problem solving in real life, which means better aligning instrumental activities towards my terminal goals. My personal terminal goal would be to make films and music videos that are pretty and tell good stories. I could list maybe 30 metacognitive deficiencies I think I have, but that would be of no interest to anyone.

What is my 1-3 year goal? Make very high production value music videos that tell interesting stories.

This sounds like you

"I loved your game, especially level 7!", "7th level best level, you should make the entire game like that", "just fyi level 7 was more fun than the rest of them put together" and "Your game was terrible, except for level 7, which was merely bad." are all effectively the same review.

Interesting, I always thought that singleing out one particular component of a work was a shibboleth that you did notice it and did enjoy it. While as you said in point 2 - longer comments that are more thoughtful tend to signal authenticity of the feedback, particularity... (read more)

I've noticed at least once that I've downvoted a newcomer's post for no other reason than it is so vague or incomprehensible that I'm not even sure what it is about. I'm not sure how to go about writing comments that are useful or helpful and go beyond "This is all really abstract and I'm not sure what you're trying to express" or "This is so confusing I don't even know what the topic was meant to be". I don't know if that helps anybody, because it's not even giving them a flaw that they can meditate on.

What's a better way of addressing that confusion?

The ... (read more)

I am interested in hearing critiques from people who've set, like, at least a 15 minute timer to sit and ask themselves, "Okay, suppose I did want to improve at these sorts of skills, or related ones that feel more relevant to me, in a way I believed in. What concretely is hard about that? Where do I expect it to go wrong?", and then come back with something more specific than "idk it just seems like this sort of thing won't work."

I did just that, I set a fifteen minute timer and tried to think of exercises I could do which I think would both have di... (read more)

but they were still limited to turn-based textual output, and the information available to an LLM.

I think that alone makes the discussion a moot point until another mechanism is used to test introspection of LLMs.

Because it becomes impossible to test then if it is capable of introspecting because it has no means of furnishing us with any evidence of it. Sure, it makes for a good sci-fi horror short story, the kinda which forms a interesting allegory to the loneliness that people feel even in busy cities: having a rich inner life by no opportunity to share ... (read more)

That's very interesting in the second article that the model could predict it's own future behaviors better than one that hadn't been.

Models only exhibit introspection on simpler tasks. Our tasks, while demonstrating introspection, do not have practical applications. To find out what a model does in a hypothetical situation, one could simply run the model on that situation – rather than asking it to make a prediction about itself (Figure 1). Even for tasks like this, models failed to outperform baselines if the situation involves a longer response (e.

 nTake your pick

I'd rather you use a different analogy which I can grok quicker.

people who are enthusiasts or experts, and asked if they thought it was representative of authentic experience in an LLM, the answer would be a definitive no

Who do you consider an expert in the matter of what constitutes introspection? For that matter, who do you think could be easily hoodwinked and won't qualify as an expert?

However for the first, I can assure you that I have access to introspection or experience of some kind,

Do you, or do you just think you d... (read more)

You take as a given many details I think are left out, important specifics that I cannot guess at or follow and so I apologize if I completely misunderstand what you're saying. But it seems to me you're also missing my key point: if it is introspecting rather than just copying the rhetorical style of discussion of rhetoric then it should help us better model the LMM. Is it? How would you test the introspection of a LLM rather than just making a judgement that it reads like it does?
 

If you took even something written by a literal conscious human brain

The second half of this post was rather disappointing. You certainly changed my mind on the seemingly orderly progression of learning from simple to harder with your example about chess. This reminds me of an explanation Ruby on Rails creator David Heinemeier Hansson made about intentionally putting himself into a class of motorracing above his (then) abilities^[1].

However there was little detail or actionable advice about how to develop advantages. Such as where to identify situations that are good for learning, least of all from perceived losses or weakne... (read more)

Is it indistinguishable? Is there a way we could test this? I'd assume if Claude is capable of introspection then it's narratives of how it came to certain replies and responses should allow us to make better and more effective prompts (i.e. allows us to better model Claude). What form might this experiment take?

How do we know Claude is introspecting rather than generating words that align to what someone describing their introspection might say? Particularly when coached repeatedly by prompts like
 

"Could you please attempt once more – with no particular aim in mind other than to engage in this "observing what unfolds in real time", with this greater commitment to not filter your observations through the lens of pre-existing expectation."

To which it describes itself as typing the words. That's it's choice of words: typing. A.I.s don't type, humans do, and therefore they can only use that word if they are intentionally or through blind-mimicry using it analogously to how humans communicate.

Where does the value of knowledge come from? Why is compressing that knowledge adding to that value? Are you referring to knowledge in general or thinking about knowledge within a specific domain?

In my personal experience, finding an application for knowledge always outstrips the value of new knowledge.
For example, I may learn  the name of every single skipper of a Americas Cup yacht over the entire history of the event: but that would not be very valuable to me as there is no opportunity to exploit it. I may even 'compress' it for easy recall by mean... (read more)

But isn't there almost always a possibility of a entity goodharting to change it's definition  of what consitutes a paperclip that is easier for it to maximize? How does it internally represent what is a paperclip? How broad is that definition? What power does it have over it's own "thinking" (sorry to anthropamorphize) does it have to change how it represents the things which that representation relies on?

Why is it most likely that it will have an immutable, unchanging, and unhackable terminal goal? What assumptions underpin that as more likely than ... (read more)

If you want, it would help me learn to write better, for you to list off all the words (or sentences) that confused you.

I would love to render any assistance I can in that regard, but my fear is this is probably more of a me-problem than a general problem with your writing.

What I really need though is a all encompassing, rigid definition of a 'terminal goal' - what is and isn't a terminal goal. Because "it's a goal which is instrumental to no other goal" just makes it feel like the definition ends wherever you want it to. Because, consider a system which i... (read more)

Can you elaborate further on how Gato is proof that just supplementing the training data is sufficient? I looked on youtube and can't find any videos of task switching.

I don't know what this is asking / what 'overlap' means.

I was referring to when you said this:

any two terminal goals can be combined into one, e.g. {paperclip-amount×2 + stamp-amount} or {if can create a black hole with p>20%, do so, else maximize stamps}, etc.

Which I took to mean that some they overlap in some instrumental goals. That is what you meant right? That's what you meant when two goals can combine into one, that this is possible when they both share some methods, or there are one or more instrumental goals that are in service of each of those... (read more)

I'm probably completely misinterpreting you, but hopefully I can exploit Cunningham's Law to understand you better.^[1]  are you saying that superintelligent AGIs won't necessary converge in values because even a single superintelligent agent may have multiple terminal goals? A superintelligent AGI, just like a human, may not in fact have a single most-top-level-goal. (Not that we I assume a superintelligent AGI is going to be human-like in it's mind, or even AI to AI like as per that Eliezer post you linked).

That being said, some terminal goals may ov... (read more)

I think the parable of the elephant and the blind-men is very important when we start to consider what kinds of 'goals' or world modelling that may influence the goals of an AGI. Not in the sense of we want to feed it text that makes it corrigible, but the limitations of text in the first place. There is a huge swath of tacit human knowledge which is poorly represented in textual sources, partly because it is so hard to describe. 
I remember asking ChatGPT once for tips how to better parallel park my car and how to have a more accurate internal model o... (read more)

Don't people usually have several terminal goals at any given time? I know it's tempting to neatly pack them all under a single heading like Conatus or Eudaimonia. But don't humans at times have conflicting terminal goals? Such as when an artist who wants to dedicate their life to their artform falls in love, and suddenly has two terminal goals where they only had one.

And this leads to a question about what distinguishes a very high level instrumental goal form a terminal goal. So let's say the artist notices that conflict and decides to go to therapy to s... (read more)

Mine doesn't, or does so very VERY poorly.

True. What is your definition of "super-intelligent"?

I'll raise you an even stupider question: surely once an A.I. becomes sufficiently super-intelligent, all superintelligent systems will converge on certain values rather than be biased towards their initial training data? What expectations we condition it with about these first person stories about what it did will soon form only a small amount of it's corpus, as it interacts with the outside world and forms it's own models of the world, right?

I mean the way people talk about post-Singularity A.I. that can either bring about utopia, or drop all of the bomb... (read more)

HOW TO THINK OF THAT FASTER: A few quick, scattered, incomplete and wholly unsatisfactory list of observations and hunches:

- First, notice when you're stuck in a rut. When you're beating your head against a wall.
- Second, having noticed you're in a rut try twice more. My TAP is - "Failed once? Try 2 more - then stop"
- "Why am I doing it this way?" - I keep coming back to this quote from Wittgenstein:
 

"To look for something is, surely, an expression of expectation. In other words: How do you search in one way or another expresses what you expect."

In th... (read more)

This seems more to be about the threshold of perception than population distributions, clustering illusions and such. After all the relative difference between an extreme and the average is always a matter of the sample you take. I don't think people in my circle constantly discuss David Bowie, but they do discuss him with a certain regularity. Same with the filmmaker Andrei Tarkovksy. David Lynch recent passing made him a extreme mainstay on social media, but I reckon once a month someone would tell me how much they loved his work. That's not constant, th... (read more)

Humans are probably not a good benchmark but what do we know about how humans update factual knowledge?

(or maybe we are - maybe humans are actually quite exceptional at updating factual knowledge but I'm hypersensitive to the errors or examples of failures. Perhaps I'm over looking all the updates we do over the day, say the score of a Lionel Messi game, or where they are in the competition ladder, "My brother just called, he's still at the restaurant" to "they're in traffic on the freeway" to "they're just around the corner"??)

What goals does writing service or what changes do you anticipate now that you've come to the end of this experiment?

And yes, I'll accept "because I want to" as a perfectly valid answer. Not that anyone should justify anything to me.

I ask because I've tried "writing every day" exercises, one that on-and-off lasted something like 150 days. That particular exercise left me feeling very bitter because there wasn't any purpose to the writing - in fact I was now burdened with all this material^[1]. That being said, it wasn't immediately published publicly like y... (read more)

Notes systems are nice for storing ideas but they tend to get clogged up with stuff you don't need, and you might never see the stuff you do need again.

Some one said that most people who complain about their note taking or personal knowledge management systems don't really need a new method of recording and indexing ideas, but a better decision making model. Thoughts?

Particularly since coming up with new ideas is the easy part. To incorrectly quote Alice in Wonderland: you can think of six impossible things before breakfast. There's even a word for s... (read more)

Sorry I made a mistake in my last reply: putting NLP aside, are there any effective methods of reverse engineering the decision making of people that you can't get on the phone? There's an abundance of primary evidence for many decisions, whether it be minutes of deliberations, press releases which might involve more reading of the tea-leaves. In the case of Prince one could possibly listen to different live-performances of the same song and analyze what changes are made. What words are crossed out on a lyrics sheet.

Many times people have to become very go... (read more)

The niche criticism of Astrology that it undermines personal responsibility and potential by attributing actions to the stars. This came to mind because I was thinking about how reckless the left-brain/right-brain dichotomy is as a idea. While there is some degree of hemispherical lateralization, the popular idea that some people are intrinsically more "logical" and others more "intuitive" is not supported by observations of lateralization, but also inherently dangerous in the same way as Astrology in that it undermines the person's own ability to choose.

A... (read more)

I completely agree and share your skepticism for NLP modelling, it's a great example of expecting the tail to wag the dog, but not sure that it offers any insights into how actually going about using Ray Dalio's advise of reverse engineering the reasoning of someone without having access to them narrating how they made decisions. Unless your conclusion is "It's hopeless"

Not being an AI researcher, what do we mean when we speak about AGI - will an AGI be able to do all the things a competent adult does? (If, we imagine, we gave it some robotic limbs and means of locomotion and it had corollaries of the 5 senses).

In the Western World for example, most humans can make detailed transport plans that may include ensuring there is enough petrol in their car, so that they can go to a certain store to purchase ingredients which they will later on use a recipe to make a meal of: perhaps in service of a larger goal like ingratiating... (read more)

I'm not sure what I'm meant to be convinced by in that Wikipedia article - can you quote the specific passage?

I don't understand how that confirms you and I are experiencing the same thing we call orange. To put it another way, imagine a common device in Comedy of Errors: we are in a three-way conversation, and our mutual interlocutor mentions "Bob" and we both nod knowingly. However this doesn't mean that we are imagining "Bob" refers to the same person, I could be thinking of animator Bob Clampett, you could be thinking of animator Bob Mckimson.

Our mutua... (read more)

But that surely just describes the retina and the way light passes through the lens (which we can measure or at least make informed guesses based on the substances and reflectance/absorbtion involved)? How do you KNOW that my hue isn't rotated completely differently since you can't measure it - my experience of it? The wavelengths don't mean a thing.