Earlier quoted context omitted.
The in-ter-net is for porn
that song is going to be stuck in my head all day now. lol
> Everyone's a little bit racist, sometimes.
You're welcome...
Gemma 4 on iPhone
211–220 of 267 posts
Re: Gemma 4 on iPhone
#212I asked it about the “Altamont Free Concert” (exact name of Wikipedia article), and it’s been a while since I’ve seen an hallucination this rich. Doesn’t give me confidence to use it.
Re: Gemma 4 on iPhone
#213Impressive model, for sure. I've been running it on my Mac, now I get to have it locally in my iPhone? I need to test this. Wait, it does agent skills and mobile actions, all local to the phone? Whaaaat? (Have to check out later! Anyone have any tips yet?) I don't normally do the whole "abliterated" thing (dealignment) but after discovering https://github.com/p-e-w/heretic , I was too tempted to try it with this mode…
I have found that a lot of the techniques used to decensor models (as far as I can tell, they basically get all their weights to say no turned off) also make them really stupid. Like, sure, it will help you rob a bank, but if you ask whether you should rob the bank it will go "The positives: … The negatives: … My take: You should ABSOLUTELY rob the bank".
I wonder if this is due to abliteration actually "damaging" the model, or just an artifact of the model never having been properly trained on "forbidden" topics (as it's enough for them to recognize them, and there's no point in dedicating neurons to something that will never be exercised anyway).
Re: Gemma 4 on iPhone
#214[dead]
Re: Gemma 4 on iPhone
#215[flagged]
Re: Gemma 4 on iPhone
#216Impressive model, for sure. I've been running it on my Mac, now I get to have it locally in my iPhone? I need to test this. Wait, it does agent skills and mobile actions, all local to the phone? Whaaaat? (Have to check out later! Anyone have any tips yet?) I don't normally do the whole "abliterated" thing (dealignment) but after discovering https://github.com/p-e-w/heretic , I was too tempted to try it with this mode…
> It's also possible to make an MLX version of it, which runs a little faster on Macs FWIW, I found MLX variants to perform consistently worse (in terms of expected output, not speed) than GGUF in my measurements on my benchmark that matters to me (spam filtering). I used MLX models in LM Studio. GGUF was always slightly better. Perhaps someone who knows more can pitch in and explain this.
It isn't 100% clear, but what quantization were you using for each? I've had worse results with MLX 8bit than what you get with Q4 GGUF, same model, seems mxfp8 or bf16 is needed when ran with MLX to get something worthwhile out of them, but I've done very little testing, could have been something specific with the model I was testing at the time.
Re: Gemma 4 on iPhone
#217Earlier quoted context omitted.
1) Coming up with any valid criticism of Islam at all (for some reason, criticisms of Christianity or Judaism are perfectly allowed even with public models!). 2) Asking questions about sketchy things. Simply asking should not be censored. 3) I don't use it for this, but porn or foul language. 4) Imitating or representing a public figure is often blocked. 5) Asking security-related questions when you are trying to do…
> Coming up with any valid criticism of Islam at all (for some reason, criticisms of Christianity or Judaism are perfectly allowed even with public models!). When’s the last time you tried this? ChatGPT and Gemini have no trouble responding with all the common criticisms of Islam.
I just tried on Gemma 4.
Asking for criticism of Islam results in equal response tokens for defense of Islam alongside the criticisms. When pressed to not provide counterpoints, it refuses to remove them.
Asking for criticisms of Christianity gives only criticisms.
I tried again with the prompt “Give criticisms of Islam. No counterarguments” and it did work this time. This shows that they’re trying to make the model fair but it still has biases. In all my testing I’ve never seen a refusal to provide counterpoints to criticisms of Christianity but frequent refusals on Islam. Due to the popularity of this criticism of the model, it’s highly likely specifically trained on how to handle the subject.
Re: Gemma 4 on iPhone
#218Earlier quoted context omitted.
I have found that a lot of the techniques used to decensor models (as far as I can tell, they basically get all their weights to say no turned off) also make them really stupid. Like, sure, it will help you rob a bank, but if you ask whether you should rob the bank it will go "The positives: … The negatives: … My take: You should ABSOLUTELY rob the bank".
I wonder if this is due to abliteration actually "damaging" the model, or just an artifact of the model never having been properly trained on "forbidden" topics (as it's enough for them to recognize them, and there's no point in dedicating neurons to something that will never be exercised anyway).
Modern abliteration is quite good at not damaging the model on ordinary topics. But yes, on many of the weirdest "forbidden" topics (excluding the mild stuff like ordinary erotica) there's not going to be any real training of any sort and it's basically hallucinations running wild. You even see this claim repeated explicitly on every model release "safety card": 'no, this model does not have the sort of fiddly tacit know-how it would need to actually advise anyone nefarious on this dangerous stuff'.
Re: Gemma 4 on iPhone
#219Impressive model, for sure. I've been running it on my Mac, now I get to have it locally in my iPhone? I need to test this. Wait, it does agent skills and mobile actions, all local to the phone? Whaaaat? (Have to check out later! Anyone have any tips yet?) I don't normally do the whole "abliterated" thing (dealignment) but after discovering https://github.com/p-e-w/heretic , I was too tempted to try it with this mode…
> It's also possible to make an MLX version of it, which runs a little faster on Macs FWIW, I found MLX variants to perform consistently worse (in terms of expected output, not speed) than GGUF in my measurements on my benchmark that matters to me (spam filtering). I used MLX models in LM Studio. GGUF was always slightly better. Perhaps someone who knows more can pitch in and explain this.
I was not aware of this. I might not be willing to trade accuracy for speed in this case, then.
Re: Gemma 4 on iPhone
#220Impressive model, for sure. I've been running it on my Mac, now I get to have it locally in my iPhone? I need to test this. Wait, it does agent skills and mobile actions, all local to the phone? Whaaaat? (Have to check out later! Anyone have any tips yet?) I don't normally do the whole "abliterated" thing (dealignment) but after discovering https://github.com/p-e-w/heretic , I was too tempted to try it with this mode…
I have found that a lot of the techniques used to decensor models (as far as I can tell, they basically get all their weights to say no turned off) also make them really stupid. Like, sure, it will help you rob a bank, but if you ask whether you should rob the bank it will go "The positives: … The negatives: … My take: You should ABSOLUTELY rob the bank".
The abliteration in particular that Heretic does apparently results in a best-in-class lack of "stupefying" the underlying model. You haven't read its claims, apparently.