Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Interview with gwern

dwarkeshpatel.com

Post reply on HN
Print view View original

201–210 of 262 posts

Re: Interview with gwern

#201

Post by tptacek »

A_D_E_P_T wrote: post #193
okasaki wrote: post #102

Interesting that there's no mention of human biodiversity (aka blacks are dumb), as if you spend five minutes on #lesswrong you'll notice that that's a big issue for gwern and the other goons.

"Goons." Nice. Thing is: Are you absolutely sure that notion of human biodiversity is wrong? IQ is heritable, as height is heritable. You'll grant that there are populations that differ in their genetic potential for height -- e.g. Dalmatians vs. Pygmies -- so how is it that you dismiss out of hand the notion that there might be population-wide differences in the genetic potential for intelligence? I can hear it now:…

What do you think heritability means?

Re: Interview with gwern

#202

Post by Vecr »

arthurofbabylon wrote: post #200

This was a tough listen, for two subtly similar reasons. The voice was uncanny. Simply hard to listen to, despite being realistic. I mean precisely that: it is cognitively difficult to string together meaning from that voice. (I am adjacent to the field of audio production and frequently deal with human- and machine-produced audio. The problem this podcast has with this voice is not unique.) The tonality and meaning…

It's a real voice. Probably with some splicing, but I don't know how much. Gwern admits he isn't that good at speaking, and I believe him. He also says he isn't all that smart. Presumably that's highly relative.

Re: Interview with gwern

#204

Post by A_D_E_P_T »

tptacek wrote: post #201
A_D_E_P_T wrote: post #193

Earlier quoted context omitted.

"Goons." Nice. Thing is: Are you absolutely sure that notion of human biodiversity is wrong? IQ is heritable, as height is heritable. You'll grant that there are populations that differ in their genetic potential for height -- e.g. Dalmatians vs. Pygmies -- so how is it that you dismiss out of hand the notion that there might be population-wide differences in the genetic potential for intelligence? I can hear it now:…

What do you think heritability means?

Farmers, who depend on this sort of thing, have it figured out:

> https://extension.missouri.edu/publications/g2910

Heritability is simply a measure of how much of the variation in a trait, like height or IQ, is due to genetic factors rather than environmental influences in a given population. Could be feedlot steers, could be broiler chickens, could be humans. In humans, traits like height are very highly heritable, at ~0.8. Certain others, like eye color, are heritable to ~0.98.

Re: Interview with gwern

#206

Post by tptacek »

A_D_E_P_T wrote: post #204
tptacek wrote: post #201

Earlier quoted context omitted.

What do you think heritability means?

Farmers, who depend on this sort of thing, have it figured out: > https://extension.missouri.edu/publications/g2910 Heritability is simply a measure of how much of the variation in a trait, like height or IQ, is due to genetic factors rather than environmental influences in a given population. Could be feedlot steers, could be broiler chickens, could be humans. In humans, traits like height are very highly heritable,…

Right. It's a simple ratio of genetic variation to phenotypical variation. How does evidence of heritability support HBD claims, which are based on genetic determinism, a notion orthogonal to heritability?

Re: Interview with gwern

#207

Post by A_D_E_P_T »

tptacek wrote: post #206
A_D_E_P_T wrote: post #204

Earlier quoted context omitted.

Farmers, who depend on this sort of thing, have it figured out: > https://extension.missouri.edu/publications/g2910 Heritability is simply a measure of how much of the variation in a trait, like height or IQ, is due to genetic factors rather than environmental influences in a given population. Could be feedlot steers, could be broiler chickens, could be humans. In humans, traits like height are very highly heritable,…

Right. It's a simple ratio of genetic variation to phenotypical variation. How does evidence of heritability support HBD claims, which are based on genetic determinism, a notion orthogonal to heritability?

I don't think that HBD claims -- at least, those made in reasonably good faith -- are based on genetic determinism. A Bosnian guy from the Dinaric alps is much more likely to be >1.8m in stature than a Pygmy. This is not predetermined as such, it's just that one population has something like +3SD in stature over the other. (An admittedly wildly extreme example!)

Differences in IQ between groups are apparently far more modest, but, however distasteful, it's still possible to speak of them, and it's possible to make statistical statements about them. My position is simply that, on the one side, it should be done in good faith -- and, on the other side, it shouldn't be seen as something heretical.

Re: Interview with gwern

#208

Post by dilap »

YeGoblynQueenne wrote: post #161
dilap wrote: post #130

Earlier quoted context omitted.

By intellectually understand, I just mean you can ask Claude or ChatGPT or whatever, "how can I recognize if a string is in a^n b^n? what is the language being described?" and it can easily tell you; if you were giving it an exam, it would pass. (Of course, maybe you could argue that's a famous example in its training set and it's just regurgitating, but then you could try making modifications, asking other questions…

>> (Of course, maybe you could argue that's a famous example in its training set and it's just regurgitating, but then you could try making modifications, asking other questions, etc, and the LLM would continue to respond sensibly. So to me it seems to understand...) Yes, well, that's the big confounder that has to be overcome by any claim of understanding (or reasoning etc) by LLMs, isn't it? They've seen so much st…

Personally I am convinced LLMs do have real understanding, because they seem to respond in interesting and thoughtfull ways to anything I care to talk to them about, well outside of any topic I would expect to be captured statistically! (Indeed, I often find it easier to get LLMs to understand me than many humans. :-)

There's also stuff like the Golden Gate Claude experiment and research @repligate shares on twitter, which again make me think understanding (as I conceive of it) is definitely there.

Now, are the conscious, feeling entities? That is a harder question to answer...

Re: Interview with gwern

#209

Post by tptacek »

A_D_E_P_T wrote: post #207
tptacek wrote: post #206

Earlier quoted context omitted.

Right. It's a simple ratio of genetic variation to phenotypical variation. How does evidence of heritability support HBD claims, which are based on genetic determinism, a notion orthogonal to heritability?

I don't think that HBD claims -- at least, those made in reasonably good faith -- are based on genetic determinism. A Bosnian guy from the Dinaric alps is much more likely to be >1.8m in stature than a Pygmy. This is not predetermined as such, it's just that one population has something like +3SD in stature over the other. (An admittedly wildly extreme example!) Differences in IQ between groups are apparently far mor…

I don't understand the alternative interpretation you're alluding to. Stipulate the validity of IQ or the common g. If group variations in these metrics aren't caused by genes, why are they distasteful? If they are, you're describing genetic determinism, which, again, is orthogonal to heritability.

Re: Interview with gwern

#210

Post by freejazz »

Vecr wrote: post #202
arthurofbabylon wrote: post #200

This was a tough listen, for two subtly similar reasons. The voice was uncanny. Simply hard to listen to, despite being realistic. I mean precisely that: it is cognitively difficult to string together meaning from that voice. (I am adjacent to the field of audio production and frequently deal with human- and machine-produced audio. The problem this podcast has with this voice is not unique.) The tonality and meaning…

It's a real voice. Probably with some splicing, but I don't know how much. Gwern admits he isn't that good at speaking, and I believe him. He also says he isn't all that smart. Presumably that's highly relative.

> He also says he isn't all that smart

Oh, how humbling and modest of him!

Post reply on HN