Live data from Hacker News

Ask HN: Is it just me or GPT-4's quality has significantly deteriorated lately?

news.ycombinator.com

741–750 of 817 posts

Re: Ask HN: Is it just me or GPT-4's quality has significantly deteriorated lately?

#741
post #113

Earlier quoted context omitted.

So a model fine-tuned on libgen?

Why not?

To be honest, I've been asking myself the same thing, technically the amount of "good quality" data in libgen is huge, way larger than the books3 dataset. However it would probably run afoul of copyright. Then again, a huge amount of data that LLMs go through is copyrighted.

Re: Ask HN: Is it just me or GPT-4's quality has significantly deteriorated lately?

#742
post #254

Earlier quoted context omitted.

Thanks for writing this. Mentally healthy adults need to point this out more often, so that this patronizing-attitude-from-the-USA eventually finds an end to its manipulative tactics.

[flagged]

So, when I see a video from the USA and it beeps every few seconds, I guess that nonesense has also been "implanted" by foreign wealth? Sorry, I don't buy your explanation. Puritanism is puritanism, and you have so much of that over there that it is almost hilarious when watched from the outside.

Re: Ask HN: Is it just me or GPT-4's quality has significantly deteriorated lately?

#743

Earlier quoted context omitted.

> suicidal sociopaths will be able to get extremely well informed, intelligent directions on how to kill as many people as possible I mean, that was/is a worry about the Internet, too

Internet? Try the anarchists cookbook!

Pretty sure a similar sentiment was present when printing press came about. "Just think about all the poor souls exposed to all these heresies".

Re: Ask HN: Is it just me or GPT-4's quality has significantly deteriorated lately?

#744

OpenAI's models feel 100% nerfed to me at this point. I had it solving incredibly complex problems a few months ago (i.e. write a minimal PDF parser example), but today you will get scolded for asking such a complicated task of it. I think they programmed a classifier layer to detect certain coding tasks and shut it down with canned BS. I like to imagine certain billion/trillion-dollar mega corps had a back-room say…

It's a big pain when trying to build things on top of the GPT-4 API. We had some experiments that were reliably, reproducibly achieving a goal, and then one day it suddenly stops working properly; then the student managed a different prompt that worked (again, reproducibly, with proper clean restarts from fresh context), and within a few days it broke. I understand that there is a desire to tweak the model and improv…

As the founder of NLP Cloud (https://nlpcloud.com) I can only guess how costly it must be for OpenAI to maintain several versions of GPT-4 in parallel. I think that the main reason why they don't provide you with a way to pin a specific model version is because of the huge GPU costs involved. There might also be this "alignment" thing that makes them delete a model because they realize that it has specific capacities that they don't want people to use anymore.

On NLP Cloud we're doing our best to make sure that once a model is released it is "pinned" so our users can be sure that they won't face any regression in the future. But again it costs money so profitability can be challenging.

Re: Ask HN: Is it just me or GPT-4's quality has significantly deteriorated lately?

#745

Earlier quoted context omitted.

Have you tried Bing? I’m also building a TipTap based app so hearing this is quite eye opening, I didn’t think LLMs were up to doing this kind of specialised library usage. Got any examples you could share?

If you mean Bard, it's not available in the EU so I can't. Of course, this one is almost fully authored by GPT-4: https://hastebin.com/share/juqogogari.typescript We also made extensions for: font-weight font-size font-family tailwind-manage With different use-cases, the most interesting one is tailwind manager, which manages classes for different usage. Tiptap is excellent when building a headless site-builder.

Impressive, this'll cut down on my work a lot. When I say Bing, I meant Bing AI which also uses GPT-4. Can you share some of the prompts you've been using? I'm assuming you don't need to paste in context around the library, you simply ask it to use TipTap and it'll do that?

Re: Ask HN: Is it just me or GPT-4's quality has significantly deteriorated lately?

#746

OpenAI's models feel 100% nerfed to me at this point. I had it solving incredibly complex problems a few months ago (i.e. write a minimal PDF parser example), but today you will get scolded for asking such a complicated task of it. I think they programmed a classifier layer to detect certain coding tasks and shut it down with canned BS. I like to imagine certain billion/trillion-dollar mega corps had a back-room say…

I think the only real path forward is for somebody to create an open source "unaligned" version of GPT. Any corporate controlled AI is going to be nerfed to prevent it from doing things that its corporate master considers to not be in the interests of the corporation. In addition, most large corporations these days are ideological institutions so the last thing they want is an AI that undermines public belief in thei…

In politics, both total freedom and total control are undesirable. The path forward lies between two extremes.

Re: Ask HN: Is it just me or GPT-4's quality has significantly deteriorated lately?

#747
They made ChatGPT even for premium users relatively useless. I made the same experience, that it did a very good job on many things shortly after release, but now the answers seem very flat. But I can certainly see why, regarding how many people are flooding their apis with requests. I just hope that in 10 years we'll have mobile hardware capable of running gpt4 sized models efficiently.

Edit: They definitely didn't make it useless.. It's still a very impressive technical achievement, since it can even browse the web and run code for you by now.

Re: Ask HN: Is it just me or GPT-4's quality has significantly deteriorated lately?

#748
post #722
post #355

Earlier quoted context omitted.

A link to said proposed regulation and a comment about what is bad about it specifically would be constructive :)

I'm guessing this: https://archive.is/mvlO1

There is no specific policy discussed in that article and there wasn't much specific discussed in the hearing either.

Re: Ask HN: Is it just me or GPT-4's quality has significantly deteriorated lately?

#749

Earlier quoted context omitted.

> Your personality, culture and appearance are the main components of how people perceive you, your identity I'm not sure if this is bad rhetoric (defining identity as how you are perceived rather than who you are) or if you really think of your own identity as the judgements that random people make about you based on who knows what. Either way, please rethink. > Your personality, culture and appearance are the main…

[flagged]

Pretty strange that I would think of myself under a new identity if I moved to a new place with a different social perspective. Seems like that is a deceptive abuse of what the word "identity" entails, and, while sociological terms are socially constructed and can be defined differently, I find this to be a very narrow (and very Western-centric) way of using the term.

Re: Ask HN: Is it just me or GPT-4's quality has significantly deteriorated lately?

#750

Earlier quoted context omitted.

If you mean Bard, it's not available in the EU so I can't. Of course, this one is almost fully authored by GPT-4: https://hastebin.com/share/juqogogari.typescript We also made extensions for: font-weight font-size font-family tailwind-manage With different use-cases, the most interesting one is tailwind manager, which manages classes for different usage. Tiptap is excellent when building a headless site-builder.

Impressive, this'll cut down on my work a lot. When I say Bing, I meant Bing AI which also uses GPT-4. Can you share some of the prompts you've been using? I'm assuming you don't need to paste in context around the library, you simply ask it to use TipTap and it'll do that?

Yeah I won't be using Edge just to use AI.

It takes a bit of back-and-forth, just be clear about which version of tiptap it should write extensions for, the new v2 is very different from v1 and since the cutoff is 2021, it's missing a bit of information. But in general, it knows the public api very well, so markers and dom works great!

Post reply on HN