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

OpenAI's new open-source model is basically Phi-5

seangoedecke.com

Post reply on HN
Print view View original

181–190 of 233 posts

Re: OpenAI's new open-source model is basically Phi-5

#181

Post by vidarh »

slt2021 wrote: post #58
kristopolous wrote: post #25

Earlier quoted context omitted.

Porn is always the frontier. It's a well-understood self-contained use-case without many externalities and simple business models. What more, with porn, the medium is the product probably more than the content. Having it on home-media in the 80s was the selling point. Getting it over the 1-900 phone lines or accessing it over the internet ... these were arguably the actual product. It might have been a driver of earl…

even if it is victim free, it can affect mental health in a way that a consumer will be more compelled to do a criminal act and create a real victim. let's say you publish a Steam game how to be a school shooter and shoot kids, wouldn't that lead to real school shootings ? who can definitely say that computer generated content about criminal behavior, won't lead to real crime with real victims? https://en.wikipedia.o…

I think you ask valid questions, but I don't think there is any credible evidence to answer those questions "yes".

Re: OpenAI's new open-source model is basically Phi-5

#182

Post by siva7 »

wmf wrote: post #20

I saw a bunch of people complaining on Twitter about how GPT-OSS can't be customized or has no soul and I noticed that none of them said what they were trying to accomplish. "The main use-case for fine-tuning small language models is for erotic role-play, and there’s a serious demand." Ah.

It's not about role-play, it just should be a bit more into my language, that's all i ask for.

Re: OpenAI's new open-source model is basically Phi-5

#183

Post by bko »

orbital-decay wrote: post #176

I gave it a random sci-fi novel and made it translate a chapter, which is something I do with all models. It refused to discuss minors in sexualized contexts. I was like W.T.F.?! and started bisecting the book, trying to find the piece that triggers this. Turns out there was some absolutely innocent, two sentence long romantic remark involving two secondary 17 years old characters in an unrelated place. Another issue…

That's so funny. I noticed this as well one time. I got some transcript from a podcast unedited (no punctuation, speaker id, etc) and it had this line I wanted to extract:

> If you’re a gay person, you might be told that if you ever move from Manhattan to Hoboken you’ll be beaten up by bat-wielding thugs right away. If you’re a woman living in a rat-infested apartment in San Francisco, where the rent is going up and up while you fantasize about a nice suburban house in Reno, Nevada, you might hear that, well, if you ever dare to move to Reno, you are going to be chained to your bed and forced to carry a baby to term. The only logical explanation is that a crazed, ideological intensification has distracted us from what’s really going on.

So naturally I threw it in an LLM to get that line and I got something that totally glossed over the "chained to a bed" with some euphemism. I wish I could find its translation again, but I tried just now translating it to Spanish and then back but it recreated that part pretty much exactly so it didn't happen again.

Re: OpenAI's new open-source model is basically Phi-5

#184

Post by wkat4242 »

CjHuber wrote: post #154
wkat4242 wrote: post #112

From the article: > For the same reason that Microsoft probably continued to train Phi-style models: safety. Releasing an open-source model is terrifying for a large organization. Once it’s out there, your name is associated with it forever, and thousands of researchers will be frantically trying to fine-tune it to remove the safety guardrails. I don't think this is really an issue in practice. Llama 2 and 3 were unc…

If I think about Llma, I think about uncensored. not that I ever used one, but there were not many use cases for censored llama when others were so much better at other things

Hmm? For me it's still the best model for general use. That I've tried at least. Gemma and phi I didn't like so much. Qwen is chinese and I've had bad experiences with those (reverting to Chinese when they get confused).

Not for knowledge, but I combine it with searches of course. None of the small models are good at knowledge.

Re: OpenAI's new open-source model is basically Phi-5

#185

Post by wkat4242 »

Clueed wrote: post #151
wkat4242 wrote: post #103

Earlier quoted context omitted.

It's not just erotic role play that the censorship affects. My life involves a lot of sexual discussions and that means that everyday talk, chat summaries, email rewrites or translations will cause the model to shut down. I do the latter a lot especially to find colloquialisms because Google translate is often too literal. It's so annoying. Right now I'm using abliterated llama 3.1. I have no need for vision but I wa…

I found DeepSeek R1 (better for questions) and V3 (better for prose) to be very willing to discuss sex with a simple system prompt, as well as being very pleasant in articulation. I guess, I prefer them because they are almost SOTA and very large. Not through the official interface though. Needs to be hosted by a third party. OpenRouter has a generous free tier for both. I just saw that there is an abliterated versio…

Oh yeah I was mainly thinking about smaller models because I want to run them locally. Not just because of the abliteration. I just want everything personal to run locally only. I don't trust any of the cloud companies.

Re: OpenAI's new open-source model is basically Phi-5

#186

Post by wkat4242 »

Alex-Programs wrote: post #175
wkat4242 wrote: post #103

Earlier quoted context omitted.

It's not just erotic role play that the censorship affects. My life involves a lot of sexual discussions and that means that everyday talk, chat summaries, email rewrites or translations will cause the model to shut down. I do the latter a lot especially to find colloquialisms because Google translate is often too literal. It's so annoying. Right now I'm using abliterated llama 3.1. I have no need for vision but I wa…

I'm curious whether you can get https://platform.nuenki.app (my deep translation service; I'm just polishing it before release) to run into "too sexual". It's designed to be resilient to that kind of thing, and it works well in my testing, but maybe you have some better "stress test" content than me!

I don't use hosted AI much. But one of the reasons I like AI is that I can ask it something like "Hmm I don't like how this sounds, can you formulate it another way, more colloquial?" and it'll take that into account. Or what I often get is that it translates "you" to singular when I need plural and vice versa. In most languages that differs but not in English.

Re: OpenAI's new open-source model is basically Phi-5

#187

Post by wkat4242 »

itsdesmond wrote: post #149
jofzar wrote: post #143

Earlier quoted context omitted.

Sorry just to ask, what kind of job do you have? Sex therapist sounds like the closest?

I don’t think they’re using it for work.

Correct, I'm just very kinky/bdsm/polyamorous and so are all my friends. I do have a sexuologist though, and I often write stuff up for her when she asks me to. Where the AI also comes in sometimes.

Re: OpenAI's new open-source model is basically Phi-5

#188

Post by wkat4242 »

Havoc wrote: post #171
wkat4242 wrote: post #103

Earlier quoted context omitted.

It's not just erotic role play that the censorship affects. My life involves a lot of sexual discussions and that means that everyday talk, chat summaries, email rewrites or translations will cause the model to shut down. I do the latter a lot especially to find colloquialisms because Google translate is often too literal. It's so annoying. Right now I'm using abliterated llama 3.1. I have no need for vision but I wa…

Pretty sure I saw an uncensored version yesterday on localllama

huh interesting thanks! I'll have a look.

Re: OpenAI's new open-source model is basically Phi-5

#189

Post by idiotsecant »

slt2021 wrote: post #100
NikolaNovak wrote: post #93

Earlier quoted context omitted.

My wife has recently joined board of directors for a local non-profit helping victims of childhood sexual abuse. They are not religious or pornography prescriptive - they are liberal atheist whose entire focus is essentially sessions to help people who are currently adults and carrying baggage and impact of dreadful acts committed to them decades ago. Anyhoo, The current "state of the art", as-scientific-as-we-curren…

I agree you captured my thoughts exactly. most of the perpetrators of child sexual abuse are victim's close people: teacher/cousin/brother/uncle/father/etc. the reinforcement of their lust will only remove whatever remaining barrier against such repulsive behavior, and once the novelty from synthetic CP wears out, it will create urge to commit real crime with real victims

A lot of opinions in this thread but very little data. What you want to believe and what is true are very often not the same. That's why we do the science.

Re: OpenAI's new open-source model is basically Phi-5

#190

Post by umeshunni »

teruakohatu wrote: post #118
wkat4242 wrote: post #112

From the article: > For the same reason that Microsoft probably continued to train Phi-style models: safety. Releasing an open-source model is terrifying for a large organization. Once it’s out there, your name is associated with it forever, and thousands of researchers will be frantically trying to fine-tune it to remove the safety guardrails. I don't think this is really an issue in practice. Llama 2 and 3 were unc…

> researchers will be frantically trying to fine-tune it to remove the safety guardrails. It is a really weak excuse. They are more likely to take a reputation hit for having silly guardrails than for having someone to remove them. Imagine if Bill Gates decided not to release MS Paint in 1985 because someone could have drawn something offensive with it.

> Imagine if Bill Gates decided not to release MS Paint in 1985 because someone could have drawn something offensive with it.

Quite believable that this could have been the case in 2021.

Post reply on HN