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

Sparks of Artificial General Intelligence: Early Experiments with GPT-4

arxiv.org

Post reply on HN
Print view View original

161–170 of 244 posts

Re: Sparks of Artificial General Intelligence: Early Experiments with GPT-4

#161

Post by voisin »

HopenHeyHi wrote: post #80
cowl wrote: post #71

Earlier quoted context omitted.

So you predict the top comments For a claim would be: 1. Dismisal 2. Trivialism 3. Non Well Formed Claim 4. I accept the claim 5. Disprove by counter example Are you sure you have not forgotten any tactic of debate to include in you prediction? I predict that you Prediction will result probably in these actions: 1. upvoted 2. downvoted

6. Meta comment for karma whoring 7. Like 6, but funnier A. Joke thread pile on B. Reprimands from humorless C. Dejected mods having to clean it all up

> B. Reprimands from humorless

You nailed it with this. I’ve never given it enough attention but this is totally bang on, found is a huge portion of highly commented posts.

Re: Sparks of Artificial General Intelligence: Early Experiments with GPT-4

#162

Post by IanCal »

SanderNL wrote: post #152
SanderNL wrote: post #140

Earlier quoted context omitted.

I agree. There is something special about layering these guys. To me this is like we are looking at a static combustion engine without the vehicle. “How is this useful?” It’s that I’m not sure what the best approach is here. Waiting for other smarter folks to put the pieces together.

I'm taking the liberty to spread my most recent words of visionary wisdom here. (/s) One of my main issues with these guys is their context window. Their memory. It's hard to see a LLM working on a code-base a few thousand tokens at a time and still being precise about it. To do that you need summary techniques. Feeding prompt with incrementally compressed summaries and hoping it will maintain cohesion. That sounds a…

Yes, there are two key things here I think.

1 - we don't hold everything in working memory. We don't even hold everything in our heads, we store things elsewhere. We then learn/have ways of bringing relevant information to the fore.

2 - we have roles that we take on.

The hierarchy/collaboration of differently prompted roles gives rise to a lot more depth. I already had this with a two LLM conversation about planning (one planner and one plan critic), drove out much more detailed actionable plans.

With the information hierarchy, for code you'd probably want something like:

High level goal summary/product description. Lower level summary about the area you're looking at. API docs of linked components. Full code of the class you're altering.

That's roughly what I have in mind I guess when working on a problem.

Re: Sparks of Artificial General Intelligence: Early Experiments with GPT-4

#163

Post by sebzim4500 »

bumbledraven wrote: post #63

GPT-4's proof of the infinitude of primes, as quoted in the paper, is wrong. Yes, I think I can, Though it might take a clever plan. I’ll start by noting Euclid’s proof, Which shows that primes aren’t just aloof. Assume we have a finite list, Of primes, and that none have been missed. Multiply them all together, And add one, just to be clever. The result will be a number, That has no prime factors, oh wonder! Not tru…

You missed a line

> Assume we have a finite list, > Of primes, and that none have been missed.

It's assuming that the finite list contains all primes and then noting that you can construct a new number which has no prime factors, which is a contradiction.

Re: Sparks of Artificial General Intelligence: Early Experiments with GPT-4

#164

Post by dns_snek »

csdvrx wrote: post #84
reidjs wrote: post #65

Well of course Microsoft is going to say something sensational about it, aren’t they in charge of the project somewhat? This is just an advertisement for them, by them.

IDK, but Microsoft seems to be now what Google was a many many years ago: a company creating tech I like to use such as Bing, Edge, Windows Terminal, VSCode, etc. Their Surface hardware is nice too (even if I prefer thinkpads) Oh and they're also helping with the linux kernel. Why can't old people let go? Companies aren't people - they respond to market incentices. Yes, Microsoft did bad stuff in the 1990s, but now t…

Not to detract from your overall point, but has Microsoft really done anything innovative when it comes to Edge, aside from painting over the Chromium skin?

The only noticeable difference that I've observed is its integration with Bing.

Re: Sparks of Artificial General Intelligence: Early Experiments with GPT-4

#165

Post by kerpotgh »

js8 wrote: post #139
ofrzeta wrote: post #137

Earlier quoted context omitted.

It can solve the variation "You have a rabbit, a wolf, a haypile and you want to cross a chasm". What kinds of variations do you have in mind?

If I add a small condition that makes the solution impossible, will it recognize that? Will it recognize for your example that it's a variation? Will it still be able to solve it when it is just a subtask of a bigger input? If I ask it a leading question that intentionally relies on a wrong solution, will it recognize that?

[dead]

Re: Sparks of Artificial General Intelligence: Early Experiments with GPT-4

#166

Post by SanderNL »

IanCal wrote: post #162
SanderNL wrote: post #152

Earlier quoted context omitted.

I'm taking the liberty to spread my most recent words of visionary wisdom here. (/s) One of my main issues with these guys is their context window. Their memory. It's hard to see a LLM working on a code-base a few thousand tokens at a time and still being precise about it. To do that you need summary techniques. Feeding prompt with incrementally compressed summaries and hoping it will maintain cohesion. That sounds a…

Yes, there are two key things here I think. 1 - we don't hold everything in working memory. We don't even hold everything in our heads, we store things elsewhere. We then learn/have ways of bringing relevant information to the fore. 2 - we have roles that we take on. The hierarchy/collaboration of differently prompted roles gives rise to a lot more depth. I already had this with a two LLM conversation about planning…

I think what we call "role-play" might be more integral to intelligence than we tend to give it credit for. Now I think of it, a "job description" could be a good prompt.

If you start with a CEO-like job agent, that can think of what other jobs are necessary then you can bootstrap from there. "I want to produce and sell red bread" => "We are going to need a bakery, accountant, marketeer, etc." and then those are "companies" of sorts with their own CEO that can think of how to solve their particular sub-problems.

Re: Sparks of Artificial General Intelligence: Early Experiments with GPT-4

#167

Post by lacksconfidence »

tandr wrote: post #43
tedunangst wrote: post #33

Earlier quoted context omitted.

Any variant of a "surprising" logic puzzle forces it to latch onto the surprising answer. Like whether two pounds of iron weighs more than one pound of feathers. Or any objects. It "expects" the twist, and always answers accordingly. It does so even if you change up the objects to be less tricky. > Which is heavier, a pound of marbles or two pounds of corn? Both weigh the same amount, which is a total of two pounds.…

I am terrible sorry, but I fail to see the logic in your 2+1=1+2 explanation of that answer. Would you be kind, and ELI5 it, if possible?

afaict the 2+1=1+2 was the response from the LLM, hence why they are dissapointed with the answer

Re: Sparks of Artificial General Intelligence: Early Experiments with GPT-4

#168

Post by ramraj07 »

parineum wrote: post #66
maxdoop wrote: post #22

Earlier quoted context omitted.

Our of curiosity, what is GPT-4 getting wrong so often? It’s prettily wild to my own , admittedly easily impressed, mind.

I've been trying to make it be a DM for a while now. It works seemingly well for a while but it's making everything up on the fly, which is fine as long as there's not state that needs to be maintained/remembered. Once you start trying to get it to do combat, it seems like it's working pretty well but eventually you realize that it's just making shit up behind the scenes. It'll tell you there are 4 goblins ambushing…

So you’re complaining it’s not smart in a way it was explicitly designed not to be (keep too long a context) so it doesn’t take over the world..

Re: Sparks of Artificial General Intelligence: Early Experiments with GPT-4

#169

Post by neilellis »

[This is in reply to the comments not the article!]

It's just a statistical model is the logical equivalent of human beings are just a bunch of atoms.

The amount of reductionist thinking that goes on in tech is hilarious. First define AGI then challenge an AI to meet those requirements. If it meets them it is AGI. Put aside your preconceptions of what technology you think is required to achieve the goals and stay empirical.

Note previous definitions of AI have been thrown away as AI passes through them one by one :-)

What goes on inside its 'head' is irrelevant. We still don't know what actually goes on inside our heads and we were damn sure we were intelligent long before we had a clue how our heads worked at all.

Also sentience != AGI. We can't even agree what sentience is in humans and other living beings so I'd stay clear of that one for now :-)

Re: Sparks of Artificial General Intelligence: Early Experiments with GPT-4

#170

Post by EntropyDenied »

dns_snek wrote: post #164
csdvrx wrote: post #84

Earlier quoted context omitted.

IDK, but Microsoft seems to be now what Google was a many many years ago: a company creating tech I like to use such as Bing, Edge, Windows Terminal, VSCode, etc. Their Surface hardware is nice too (even if I prefer thinkpads) Oh and they're also helping with the linux kernel. Why can't old people let go? Companies aren't people - they respond to market incentices. Yes, Microsoft did bad stuff in the 1990s, but now t…

Not to detract from your overall point, but has Microsoft really done anything innovative when it comes to Edge, aside from painting over the Chromium skin? The only noticeable difference that I've observed is its integration with Bing.

They made it vastly better in terms of resource utilization, specifically RAM usage. Anytime I restart Chrome to update the browser it's astonishing how much RAM is freed up after all my tabs open up again. Edge seems to have plugged a lot of memory leaks in comparison.
Post reply on HN