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

Using ChatGPT to generate a GPT project end-to-end

github.com

Post reply on HN
Print view View original

81–90 of 225 posts

Re: Using ChatGPT to generate a GPT project end-to-end

#81

Post by ZephyrBlu »

c7b wrote: post #78
gwoolhurme wrote: post #61

Earlier quoted context omitted.

Could you explain in detail how that causes the singularity? I’m lost. I saw a cool tool used to make another cool tool. You saw the singularity. Where is this logic coming from? Why would terminal access which I’m fairly certain I’ve seen in autogpt, change much.

The cycle could look something like this: LLMv1 > prompt "get me the code for a better LLM" > paste output to terminal, execute > LLMv2 > repeat. There are lots of different singularity definitions on wikipedia, most focus on 'intelligence' (eg "an upgradable intelligent agent will eventually enter a 'runaway reaction' of self-improvement cycles, each new and more intelligent generation appearing more and more rapidl…

You can already do this in theory with tools like AutoGPT and BabyAGI. Why hasn't the singularity happened already?

Re: Using ChatGPT to generate a GPT project end-to-end

#82

Post by moffkalast »

bonzini wrote: post #69
mlboss wrote: post #67

Earlier quoted context omitted.

What does it has to do that will qualify it as artificial intelligence ? In my opinion, all the ingredients are there for artificial intelligence. Somebody just need to stitch everything up. It can understand text, reason about it, identity next steps, can write code to execute the steps, understand error messages and fix the code. That feels like AI.

Be able to say "I don't know".

GPT 4 tells me that on a daily basis because of the 2021 data cutoff.

Re: Using ChatGPT to generate a GPT project end-to-end

#83

Post by enraged_camel »

gwoolhurme wrote: post #68
choeger wrote: post #65

Earlier quoted context omitted.

Calm down. It is a language model . People have figured out how to predict the next word for a given prefix. That's very cool and it will definitely have a significant impact on software. But it is not artificial intelligence.

God damn the down votes. I agree with your overall thesis I don’t personally know what intelligence means. However it’s just a tool even if it is AI or whatever you want to label it. It’s extremely cool and powerful too. It scares the shit out of me as well. However I think we are also taking the hype to 11/10 when it should be much lower out of 10 than that…

It reminds of me the rhetorical question: if we found intelligent alien life out there in the universe, would we even be able to recognize it as such?

The same question may apply here: if artificial intelligence emerges, would we be able to realize that it is intelligent?

Re: Using ChatGPT to generate a GPT project end-to-end

#84

Post by gwoolhurme »

c7b wrote: post #78
gwoolhurme wrote: post #61

Earlier quoted context omitted.

Could you explain in detail how that causes the singularity? I’m lost. I saw a cool tool used to make another cool tool. You saw the singularity. Where is this logic coming from? Why would terminal access which I’m fairly certain I’ve seen in autogpt, change much.

The cycle could look something like this: LLMv1 > prompt "get me the code for a better LLM" > paste output to terminal, execute > LLMv2 > repeat. There are lots of different singularity definitions on wikipedia, most focus on 'intelligence' (eg "an upgradable intelligent agent will eventually enter a 'runaway reaction' of self-improvement cycles, each new and more intelligent generation appearing more and more rapidl…

What does a better LLM mean? We again… already have terminal usage in AutoGPT. It describes a little what you are talking about. What measurement do you use to test if a language model 1 is better than 2? Accuracy? Self reliance? Is that what a Transformer should be doing? Sure an LLM can compete with humans on all of those tasks. Fantastic. So are we at the singularity now as it’s defined? Do I quit my job now?

Re: Using ChatGPT to generate a GPT project end-to-end

#85

Post by gwoolhurme »

ZephyrBlu wrote: post #81
c7b wrote: post #78

Earlier quoted context omitted.

The cycle could look something like this: LLMv1 > prompt "get me the code for a better LLM" > paste output to terminal, execute > LLMv2 > repeat. There are lots of different singularity definitions on wikipedia, most focus on 'intelligence' (eg "an upgradable intelligent agent will eventually enter a 'runaway reaction' of self-improvement cycles, each new and more intelligent generation appearing more and more rapidl…

You can already do this in theory with tools like AutoGPT and BabyAGI. Why hasn't the singularity happened already?

Exactly AutoGPT is exactly what he is describing but I’m still lost as to what the verbiage is then. Is that the singularity?

Re: Using ChatGPT to generate a GPT project end-to-end

#86

Post by RedTwoOne »

I really appreciate the well documented, well commented code.

It's usually hard to find energy to do this on personal projects, if nothing else that aspect makes GBT coding worth it to me

Especially in this domain, most of the pytorch projects are impossible to read with the lack of comments and obscure variable names

Re: Using ChatGPT to generate a GPT project end-to-end

#87

Post by gwoolhurme »

enraged_camel wrote: post #83
gwoolhurme wrote: post #68

Earlier quoted context omitted.

God damn the down votes. I agree with your overall thesis I don’t personally know what intelligence means. However it’s just a tool even if it is AI or whatever you want to label it. It’s extremely cool and powerful too. It scares the shit out of me as well. However I think we are also taking the hype to 11/10 when it should be much lower out of 10 than that…

It reminds of me the rhetorical question: if we found intelligent alien life out there in the universe, would we even be able to recognize it as such? The same question may apply here: if artificial intelligence emerges, would we be able to realize that it is intelligent?

I have no idea. Probably not. The philosophy is a deep rabbit hole. A fun one to ponder, and I like having that discussion. Maybe the more cynical pragmatic old man swe that I am sees a super powerful calculator kind of. It’s very very cool, and obviously I can’t hold a conversation with a calculator but my analogy is more to say my calculator can do really complex integrals and even show its steps kind of! Especially for my handy TI-89. It was my best friend for all of engineering undergrad. I see chatgpt as a steroid version of that for all of Language. Code is another language, writing is a language, painting in some ways is a language.

Re: Using ChatGPT to generate a GPT project end-to-end

#88

Post by c7b »

gwoolhurme wrote: post #84
c7b wrote: post #78

Earlier quoted context omitted.

The cycle could look something like this: LLMv1 > prompt "get me the code for a better LLM" > paste output to terminal, execute > LLMv2 > repeat. There are lots of different singularity definitions on wikipedia, most focus on 'intelligence' (eg "an upgradable intelligent agent will eventually enter a 'runaway reaction' of self-improvement cycles, each new and more intelligent generation appearing more and more rapidl…

What does a better LLM mean? We again… already have terminal usage in AutoGPT. It describes a little what you are talking about. What measurement do you use to test if a language model 1 is better than 2? Accuracy? Self reliance? Is that what a Transformer should be doing? Sure an LLM can compete with humans on all of those tasks. Fantastic. So are we at the singularity now as it’s defined? Do I quit my job now?

Like I said in the original post, clearly telling which of two LLMs is better is difficult, and probably one of the things holding us back from having a singularity right now. But that doesn't seem to be an insurmountable problem. Most people who have used both seem to agree that GPT4 is 'better' than GPT3, so maybe we can formalize that intuition somehow.

Re: Using ChatGPT to generate a GPT project end-to-end

#89

Post by c7b »

avidphantasm wrote: post #75
c7b wrote: post #56

The technological singularity is approaching fast. It's pretty clear that, given terminal access and an appropriate outer loop, GPT models can iteratively create new GPT models (either by writing and executing Pyhton code, or later versions trained on LLM weights may even be able to output new weights directly). If the inner workings of the loop are sufficiently obfuscated (in the code-based version), it wouldn't nec…

Unless this singularity is able to construct for itself sensors and actuators that can refine existing knowledge and acquire new knowledge, it will forever be hamstrung by the incomplete and static knowledge of the universe that was used to train the initial model. This is where regulation is desperately needed. We must be very careful with the kinds of sensors and actuators that are integrated with these systems. Pe…

Well, if the terminal has internet access, it can download whatever new data has been created and train a new model on that. In theory, it could also hack into a 3D printing workshop, or get some funds (eg steal credit card data, or perhaps even legally) and just issue regular orders. This is clearly very speculative thinking, but I think we are making a similar point actually.
Post reply on HN