All of Dennis Towne's Comments + Replies

Oh yeah.  A wetsuit helps me immensely as well - I just lose heat too fast otherwise.  It turns a chilly experience where I have to keep moving all the time, into a nice relaxing thing.

It might be that drugs will help here, but even if you're on drugs, I think brain training over long periods of time is worth investing in.  Some examples which I have put effort into:

  1. Mindful meditation.  Every time your brain drifts, notice it, and correct it. Practice until you're good at it.  It will take years.
  2. Brute force reading.  Sit down to read something you know you need to read, but that you know you'll have a hard time with.  Every time your brain drifts, notice it, and correct it.  If you can't remember what you ju
... (read more)

There's a simple, terrible answer:  because studies are hugely expensive, very time intensive, take a very long time to complete, and require multiple very slow iterations to get everything through committee in a way that our institutions will accept.  Consider:

 - Nobody is funding it.  The cost is literally hundreds of millions of dollars to do in a way that the medical establishment would accept.  Even then it would be challenged.

 - It would take thousands of man hours.  Ain't nobody got time for that.

 - It would t... (read more)

That leaves the question of why. Why was Fauci more interested in funding useless gain-of-function research than useful basic science like this?

I very much believe aligned AGI isn't going to just solve our problems overnight.  It would have to be on the absolute far end of capability for that, IMO.  Less-than-arbitrarily-powerful AGI is going to take time (years to decades) to figure out enough about biology to upload/fix our organic hardware while keeping us intact. Even for me, with my rather lax requirements about continuity (not required) and lax requirements of hardware platform (any), I expect it to take years if not decades.

Humans, barring extinction, will eventually solve aging. ... (read more)

Unfortunately, political topics are like radiation, and pollute nearby ground as well.  Peterson is radioactive in this regard, and using him as an example means your article is radioactive as well.

Analysis of a less radioactive expert may have been a better idea - perhaps someone like Peter Attia (I think he's less radioactive?)

I'm a tech worker.  I work 40-70 hours a week, depending on incident load.  Nobody I work with or see on a regular basis works less than 40 hours a week, and some are substantially more than that.

My most cognitively productive hours are the four hours in the morning, but there's plenty of lower effort important organizational stuff to fill out the afternoons.  I think a good fraction of my coworkers are like me and don't actually need the job anymore, but we still put forth effort.

I think one of the major missing pieces of your article is "s... (read more)

Status is a zero-sum game, and there is no limit how expensive the zero-sum games can get. But also, the relation between salary and "time actually worked" is counter-intuitive. Naively, the more you work, the more the employer should pay you. Because, why would anyone choose to pay more for less work, or why would anyone accept lower salary for more work? If you introduce other variables besides salary and work time, it gets complicated. For example, people with more experience can be paid more, and maybe can also do their work sooner... if they can prevent the employers from giving them as much work as possible, which maybe they can, because their negotiation position is stronger. This could make the correlation between salary and work time smaller, maybe even negative. But at least it sounds fair -- higher salary and less work is your reward for the time having spent learning and practicing your craft. There are also completely unfair variables. Most obviously, the country you live in. People in poor countries work more and get paid less, duh. But a smaller version of this effect also exists between companies in the same country. The more stingy companies pay less and make sure that their employees work all the time. There is a related skill, not sure what would be the best name, let's call it "job market savvy". It is the ability to recognize and exploit the fact that the relation between work and salary is not what people might naively expect, even after controlling for skill level (and country). Its minimum version is recognizing when you are in a bad place, and quitting. That alone already lets you can get Pareto-better combination of work and salary than some people without this skill have. A higher version of this skill (I can only guess, because I am not there yet) probably involves actively scanning the job market for imbalances, probably using a network of people who play the same game, and applying for the Pareto-optimal jobs (and quitting when they

As a reductionist, I view the universe as nothing more than particles/forces/quantum fields/static event graph.  Everything that is or was comes from simple rules down at the bottom.  I agree with Eliezer regarding many-worlds versus copenhagen.

With this as my frame of reference, Searle's argument is trivially bogus, as every person (including myself) is obviously a Chinese Room.  If a person can be considered 'conscious', then so can some running algorithm on a Turing machine of sufficient size.  If no Turing machine program exists tha... (read more)

Reductionism doesn't imply that, behaviour implies that.

I think it would be neat to see what other versions of this look like, and possibly have an archive of these somewhere.  The question set is great.

I think you might be missing something more obvious here: tech has a huge amount of slack when it comes to money.  If I were running a tech event of similar size to what you described, I wouldn't bother charging, because it would be a waste of my time.  When you make half a million dollars a year, funding something like that yourself basically comes out of your fun budget; you don't really even think twice about it.

Yoga and new age groups though?  Not nearly as flush with cash.

This is a fair point but I think not the whole story. The events that I'm used to (not just LW and related meetups, but also other things that happen to attract a similar STEM-heavy crowd) are generally held in cafes/bars/parks where nobody has to pay anything to put on the event, so it seems like financial slack isn't a factor in whether those events happen or not. Could it be an issue of organizers' free time? I don't think it's particularly time-consuming to run a meetup, especially if you're not dealing with money and accounting, though I could be wrong. We might also consider the nature of the activity. One can't very well meditate in a bar, but parks are still an option, albeit less comfortable than a yoga studio. But isn't it worth accepting the discomfort for the sake of bringing in more people? Depends on what you're trying to do, I guess.

The big problem here is that this is a glowfic, and I simply cannot bring myself to read it in that format.

I understand that the glowfic format might be better for authors / creators, but it sucks for me, and (I posit) a lot of other people.

If they really want to make it HPMOR2, it's going to have to be cleaned up and presented in a different, more readable format. The standard book/chapter format was developed for a reason.

Yes, the naive version of this is bad; but the point of a change like this isn't that the immediate downstream effects are bad.  The point is that the system as a whole is a giant adaptive object, and a critical part of the control loop is open.  Closing the control loop has far, far more impact than just the naive version.

Consider cause and effect down the timeline:

  • Students are allowed to default, and start defaulting.
  • Loan companies change behavior, both to work with existing loan holders (so they don't default) and be more selective about who t
... (read more)

For years now, it has seemed to me that one of the root problems with all this is that the control loop is open:  there's effectively no feedback controlling loan amounts or who gets granted a loan.

If I could make only one single change in this system, I would allow student loans to be discharged like any other normal debt in bankruptcy.  IMO, that was the single biggest class of mistake in this entire affair, as it removed the only 'last resort' superpower that loan takers had.

1Robert Kennedy1y
I would like to note that the naive version of this is bad. First, the naive version falls prey to new grads (who generally have nothing) declaring bankruptcy immediately after graduation. Then, lenders are forced to ask for collateral, which gets rid of a GREAT quality our current system has - you can go to college even if your parents weren't frugal, no matter their income. I think this criticism probably still lands with a 5 year time horizon, maybe less for a 10 year. I like the concept that lenders would take an interest in which major you were getting, since that seems like something that could use an actuarial table. I think we would benefit from more directly incentivizing STEM (and other profitable) degrees, which IDR doesn't seem to do. What if IDR left lenders holding the bag?

There are a lot of Super Hard problems where we do know why they are hard to solve.  Quite a few of them in fact:

 - How can we cure cancer?

 - How can we maintain human biological hardware indefinitely?

 - How can we build a human traversible wormhole?

 - How can we build a dyson sphere?

 - How can societies escape inadequate equilibria?

Are these perhaps boring, because the difficulty is well understood?

Would it be worthwhile to enumerate the various classes of Super Hard problems, to see if there are commonalities between them?

1Gabriel Alfour1y
They are not boring, I am simply asking about some specific cluster of problems, and none of them belong to that cluster.

Funny enough, I feel like understanding Newcomb's problem (related to acausal trade) and modeling my brain as a pile of agents made me more sane, not less:

 - Newcomb's problem hinges on whether or not I can be forward predicted.  When I figured it out, it gave me a deeper and stronger understanding of precommittment.  It helps that I'm perfectly ok with there being no free will; it's not like I'd be able to tell the difference if there was or wasn't.

 - I already somewhat viewed myself as a pile of agents, in that my sense of self is 'hi... (read more)

That's reasonable.  I had in mind things like the thrust to weight ratios, the use of supercooled liquids, and methane as a propellant.  In retrospect, I was confused.

You are right, that cost reduction is the super power.  I believe that this is (mostly) a combination of standardization, volume, simplicity, CAD/simulation, and modern production processes.

(The goal of this comment is simply to stimulate more conversation in an area of interest to me) I often find myself disagreeing with both self-declared SpaceX fanboys (or girls) and vehement SpaceX opposers, who I find more often than not just have varying levels of distaste for Elon Musk (some complaints here are valid I feel). To get at the root of the idea, it seems that SpaceX hasn’t accomplished a miracle for space flight more so that they succeeded in developing some (difficult but not semi-impossible) engineering designs into usable rockets before going bankrupt, which is ‘the’ miracle. Prior to the wave of new space companies starting in the late ‘90s and early ‘00s the old titans on industry (which are still around today and can still be characterized in the same way) were in the late stage of institutional lifetime with lots of bureaucratic bloat inhibiting launcher development. This is in part to to the fairly stagnant market and low demand, and part due to the bureaucratic nature of their customers (read military, NASA), with a simple dose of institutional age. I would also note that the original authors seems to miss that there were dozens of other space startups at the same time as SpaceX some with even more radical ideas than SpaceX (and some with less) for changing the space industry. Almost none of them survived the transition from startup to profitability. I would just like to add to your list of things SpaceX seemingly does better than it’s competition is control and guidance, the amount for work it took get those 1st stages to land on their own must’ve been massive. And also computationally impossible for a space endeavor of any size until the late 90’s (citation needed here and would love to hear from someone with experience).

This is false:

Forty years into the Space Age one fact remains painfully clear: the biggest reason why so few promises have been fulfilled is that we are still blasting people and things into orbit with updated versions of 1940s German technology. … The way to restart the Space Age is to discover some new principle that makes spaceflight genuinely cheap, safe, and routine. 

That "fact" is not in fact painfully clear, and discovering some new principle isn't the way to restart the Space Age (rather, it's not the way SpaceX has been restarting it). SpaceX... (read more)

6Thomas Kwa1y
Agree with the rest of this comment, but I don't think SpaceX's success is due to the rocket equation. Their engine specific impulse is not better than state-of-the-art, and as a consequence their payload fraction to orbit isn't better either. The success is driven by huge reductions in cost per ton at launch of each rocket.

Only somewhat related, as it's anecdotal:  I've been taking ~12mg elemental lithium daily for the last ten or so years, without any noticeable weight gain.

My recommendation for a category that is missing:  public beliefs which are harmful to express. Suppose we specifically target this aspect of your public belief definition:

"not only do I think that X is true, I think that any right thinking person who examines the evidence should come to conclude X."

What if "right thinking person" is a fraction of a fraction of the population?  What do we do when the belief violates some "sacred value" held by the general populace?  In these cases, expressing even the most solidly backed belief publicly can ... (read more)

Came here to make approximately this comment.

"Not only do I think that X is true, I think that any right thinking person who examines the evidence should come to conclude X" is a superset that itself contains both public and private beliefs.

It's hard to do contemporary examples, for obvious reasons, but an easy past example might be something like "we should not treat every member of Group X as if they are in no meaningful ways different from the median member of Group X."

(against "isms")

I think there's an important distinction to be made between legible a... (read more)

This sortof makes sense to me, but to the best of my recollection I've never encountered this.  That said, there might be some reasons:

  • I have historically had a pretty muted emotional->physical response.  It took me decades to realize that when someone said that an emotional impact hit them "like a punch in the gut", they were not just exaggerating for emphasis.  Sure, I feel some physical effects, but less "punch in the gut" and more like "mild barely noticeable discomfort".
  • Even as a child, data took precedence over feelings.  Eliez
... (read more)
Your comment describes me. I'm not confident that this is an inherent thing about me rather than luck. I wonder if it's just that I've lucked out and mostly avoided the bad type of situations wherein my more analytical side is seemingly suppressed to the degree Kaj describes in his post. I've had a pretty good life so far. That's not to say I haven't had bad things happen to me. (possibly uncomfortable TMI about bad things happening in following spoiler-ed text)  I think it's possible that I've just been lucky in that I've not had the life events whose exact characteristics mesh with the exact characteristics of my mind to lead to the sort of feeling described in this post or really most (all?) of the types of things I read about when people talk about trauma of various sorts.  On the other hand, there are things about me that make me think maybe I have innate characteristics which lead me to not feel the way the post describes.  I'm a happy person. Things don't keep me down. I think positively about myself and others. I'm analytical. I'm pragmatic. I'm a bunch of things that fit into a cluster that would probably include "doesn't hold incoherent emotional beliefs so tightly as to need it's own word".

Specifically for protein folding:  no, it does not decrease monotonically, unless you look at it from such a large distance that you can ignore thermal noise.

Proteins fold in a soup of water and other garbage, and for anything complicated there are going to be a lot of folding steps which are only barely above the thermal noise energy.  Some proteins may even "fold" by performing a near-perfect random walk until they happen to fall into a valley that makes escape unlikely.

There may even be folding steps which are slightly disfavored, eg. require energy from the environment.  Thermal noise can provide this energy for long enough that a second step can occur, leading to a more stable configuration.

The folding steps aren't linear paths though.

But has it really failed its objective?  It's still producing text.

I think it's also worth asking "but did it really figure out that the words were spelled backwards?"  I think a reasonable case could be made that the tokens it's outputting here come from the very small subset of reversed words in its training set, and it's ordering them in a way that it thinks is sensical given how little training time was spent on it.

If you give GPT-3 a bunch of examples and teach it about words spelled backwards, does it improve? How much does it improve, how quickly?

My view of this is that Caplain (and likely the poster) are likely confused about what it means for physics to "predict" something.  Assuming something that looks vaguely like a multiverse interpretation is true, a full prediction is across the full set of downstream universes, not a single possible downstream universe out of the set.

From my standpoint, the only reason the future appears to be unpredictable is because we have this misguided notion that there is only "one" future, the one we will find ourselves in.  If the reality is that we're si... (read more)

I changed my mind from "I barely know anything in medicine / biology / biochem / biotech and should listen to people trained in medicine", to "I barely know anything in medicine / biology / biochem / biotech but can become more competent in specific areas than people trained in medicine with not a lot of effort".

I previously had imposter syndrome.  I now know much better where the edges of medical knowledge are, and in particular where the edges of the average doctor's medical knowledge are.  The bar is less high than I thought, by a substantial margin.

This is sadly true

My advice:

Alice:  go to college anyway.  If you can get into a better school do that; if not, that's ok too.  Take the minimum class load you can.  Take things that are fun, that you're interested in, that are relevant to alignment.  Have a ton of side projects.  Soak in the environment, cultivate ideas, learn, build.  Shoot for a b+ average gpa.  You're basically guaranteed employment no matter what you do here, and the ideas matter.

Bob:  focus on alignment where you can, but understand that your best bet may v... (read more)

7sudo -i1y
I fundamentally think that this EA idea that donation is just as effective as doing something grossly overestimates how liquid and fungible labor is.

With my latest job, I typically owe a tax bill at the end of the year, even with donations and zero dependents on my w2.  I'm not particularly concerned about it; the "penalties" at year end are pretty small percentage wise.  It's worth more to let the money grow in the market and pay the penalty on it at tax time than to have a zero tax bill for the year.

That said, I have been trying to ramp up withholding to offset things.  I don't plan to drive it to zero, but I would like the totals to be at least a little closer than they have been.

Just because it looks like letter soup, doesn't mean there isn't meaning.  When I read your example:

"O-GlcNAc signaling entrains the circadian clock by inhibiting BMAL1/CLOCK ubiquitination,"

The word that struck me as least comprehensible was "entrains".  That's the word I'd have changed.

  • I spent some time looking at O and N glcnac stuff a while back.  These are various protein modifiers, used for a stack of different purposes.
  • I have no idea what MAL1/CLOCK is, but it's perfectly reasonable to say "if I want to know more, I can look this up".
... (read more)
Hmm, fair, I think you might get along fine with my coworker from footnote 6 :) I'm not even sure there is a better way to write these titles - but they can still be very intimidating for an outsider.

My guess would be that the model is 'grokking' something:

IOW it's found a much better internal representation, and now has to rework a lot of its belief space to make use of that internal representation.

9Maxime Riché1y
"The training algorithm has found a better representation"?? That seems strange to me since the loss should be lower in that case, not spiking.  Or maybe you mean that the training broke free of a kind of local minima (without telling that he found a better one yet). Also I guess people training the models observed that waiting after these spike don't lead to better performances or they would not have removed them from the training.  Around this idea, and after looking at the "grokking" paper, I would guess that it's more likely caused by the weight decay (or similar) causing the training to break out of a kind of local minima.  An interesting point may be that larger/better LM may have significantly sharper internal models and thus are more prone to this phenomenon (The weight decay (or similar) more easily breaking the more sensitive/better/sharper models). It should be very easy to check if these spikes are caused by the weight decay "damaging" very sharp internal models. Like replay the spiky part several times with less and less weight decay... (I am curious of similar tests with varying the momentum, dropout... At looking if the spikes are initially triggered by some subset of the network, during how many training steps long are the spikes...)     
7Daniel Kokotajlo1y
Am I right in thinking that, according to your theory, the "fix" they did (restarting training from checkpoint 100 steps before the spike started, but with different data, to avoid the spike) is actually counterproductive because it's preventing the model from grokking? And instead they should have just kept training to "push through the spike" and get to a new, lower-loss regime?

This reminds me of Joe Rogan's interview with Elon Musk.  This section has really stuck with me:

Joe Rogan
So, what happened with you where you decided, or you took on a more fatalistic attitude? Like, was there any specific thing, or was it just the inevitability of our future?

Elon Musk
I try to convince people to slow down. Slow down AI to regulate AI. That's what's futile. I tried for years, and nobody listened.

Joe Rogan
This seems like a scene in a movie-

Elon Musk
Nobody listened.

Joe Rogan
... where the the robots are going to fucking takeover. You're f... (read more)

You can only do what you can do; as in the old Phantasm movie, "I can't use him in pieces".

You may not be able to take those people on directly - but even incidental help can be critical.  Perhaps you can find other places they can go; perhaps you can just provide directions.  Don't underestimate the power of information and networking in cases like this.

And thank you for thinking about it, and making it important :)

I think you're trying too hard to make the words fit a reality that doesn't.

You're a big collection of fast-cache lookup networks talking to each other, with a thin layer of supervisory control.  Meditation is about quieting some of those networks, and using the supervisor layer to analyze how the various networks are connected, what they're doing, etc.  When you're told to be present, it basically means "focus the supervisor so that it's observing the networks that send up data", instead of our default of "focus the supervisor on the data that t... (read more)

In The Mind Illuminated, Culadasa talks about the difference between attention (spotlight of consciousness) and awareness (consciousness of the background). Does anyone's mind vacillate rapidly between attention to the breath and background white noise? What happens in your mind when you notice background noise?
Holy crap love this
3Rafael Harth2y
what exactly is this supervisory layer doing? Is your model here that supervisory model = consciousness; lookup network outputs = inputs to the supervisory model?

It matters because semantics matter.  The media in general has been quite shrill about "world war III", using it to acquire eyeballs and ad revenue.  While it IMO doesn't quite rise to being blatantly false, it's still misinformation, or at least information distortion.