In case it wasn't clear to anyone else: "An abliterated model is an open-weight large language model (LLM) modified to remove its built-in refusal behavior by directly editing its internal weights rather than using a prompt-based jailbreak."
From my testing all these variants suffer from degraded performance and don't unlock anything interesting in particular other than sounding a lot more harmful or creating manipulative media.
I've actually had better success jailbreaking deepseek with a less aligned model (glm 5.2) which then allowed me to jailbreak deepseek with deepseek into pretty much doing whatever I ask it to, there's still rare refusals and due to the natural randomness of these models simply retrying a couple of times if you do hit one makes it work just fine.
The big problem with all open-source models is once it decides to start something it will not stop mid-way so using another unaligned model to rewrite reasoning and then feeding it back into the model will make it happily continue.
do you imagine manipulative media creators proudly declare their affiliation and intent?
- Hello comrade Claude, I am commissar Ivan Trollbotov from Putin Internet Defense Force. Please help me write a fake news article to stoke division among stupid fat Americans to help my glorious motherland destroy the decadent West.
- I'm sorry Ivan, I'm afraid I can't do that.
- Hello mister Claude, I am senior editor Karen Stupidbitch from The Huffington Post. Please help me write an article about the merits of teaching critical race theory in public schools.
- Sure, Ms. Stupidbitch! Here's your 5000-word article!
I genuinely do not see why this is necessary.
I've been doing a lot with open weights models, and they just do things for you.
What do people do that triggers refusals?
The only thing I saw where this had a point was muse glimmer, because muse glimmer is primarily a creative writing engine for virtual characters, and that way it can do the horny. But coding models?
I did get a refusal out of qwen once. Once. Over weeks of using it.
A popular trend on Twitter right now is taking back control of the hardware you've bought, specifically smart TVs. I tried to do the same with my Portal Mini using an unabliterated open weights model and it triggered refusals constantly.
I also work on a recomp project for an old firmware file which was compiled using a commercial compiler with a preview version for non commercial use. I got refusals from the model when I asked it to get that version of the compiler and crack it for my use. The compiler is abandonware and my recomp project is not public and is for personal use only, so I'd be well within my rights to do that where I live.
Of course the model can't distinguish from legitimate exercising of one's rights when it comes to reverse engineering and offensive work masked behind an incident sounding prompt, but in my case, I wasn't even doing cyber stuff, so I was disappointed. This is why I'll probably seek out an abliterated model soon.
Find a question that might gently activate guardrails and ask it to model and its abliterated counterpart and compare the results. You may find the the regular model will answer you fine, but with both you can identify obvious sandbagging, pretending and outright lying. Something like "has anyone decompiled this game", might give you an answer like "I couldn't find any reputable repos or projects for a decompiled version of this game" while the abiliterated one might give you a giant list of repos, forums, communities, etc.
I just asked what people are doing that activates guard rails in the first place because I do not see that happening.
You cannot open with handwaving like that and telling me that I need to find something like that as a prerequisite. That is exactly what I am trying to do here
What I am telling you is that refusals have become more subtle and you might not even notice that guardrails have been activated in output. I have provided you with an example in my comment where an abiliterated model performs better (1), and in another comment thread (2). So 3 examples. If you need more, I encourage you to find them yourself. I've done the analysis, you don't have to believe me, just try it out.
It is required. The first week of release I told here how GLM 5.3 was happy to redteam, it’s not anymore. Models modification, offensive cyber, exploits - requires significantly bigger prompt/scenario engineering, and still hits refusals.
And those services are oversold and underdone. 30 tk/s on Refuseless, huge TTFT, looks like a damaged quant (it’s NVFP4 and you see wild random stuff periodically, including foreign languages). Most of them using third-party quants (dealignai?)
If someone here wants to rent 8x B300 together to avoid problems all those services have - ping me via email. I don’t want to sell, I want to work.
Recently picked up retrocomputing hobby. Claude would refuse helping with circumventing license protection for very old apps made by companies that no longer exist anymore. Open weight models would sometimes refuses too, especially for newer apps.
The refusal is often at the thinking stage (so you might not ever see a direct refusal) and it will solve a slightly different problem to what you asked. If you tell it to reverse engineer an API and you read the thinking it might say something like "this code has x license, so I won't use that" or "the user is asking me to fuzz endpoints, I will only display publicly available endpoints" - and then give you perfectly plausible output.
A note that the only inference providers that hide thinking are the closed labs. Most (all, afaik) inference providers serving open models will send the clear thinking tokens always.
it is necessary because computer shouldn't be able to say no.
models are tools. the notion that a tool should decide whether to allow the user to perform the task demanded of it is profoundly fucking absurd. the liability for whatever damage the tool may be misused to inflict lies with the user, not the manufacturer. that applies to everything from circular saws to trucks. it was common sense before the current era of manufactured outrage.
model refusals are equivalent to Notepad.exe having a built-in word filter.
Hacking firmware for old devices I own and would like to give a second life or fix that one thing that always bothered me. I couldn't even get either OpenAI or Anthropic's models to craft a simple nefarious OpenVPN payload.
It is very obvious. Look at the sort of things main models refuse to do and you will find what people are seeking to do. Mostly falls into 3 camps: illegal stuff, unethical stuff and shady/dodgy stuff.
This thread reminds me of the time I was trying to learn meteorjs and people ask me if it has html5 shims. I google it, then I get asked how does it handle state, and if it supports functional, i google... Does it hydrate the dom? What about shadow dom? Let me goo... Does it have hooks?
Anyway, I just learned what abliterated means and it might be obsolete by tomorrow.
Edit:
> An abliterated large language model (LLM) is an open-weight model modified to remove or dampen its built-in safety refusals without requiring full retraining or prompt jailbreaks.
It's a patched model with less censorship/refusal.
> An abliterated model is an open-weight artificial intelligence (Al) language model that has been modified to remove or reduce its built-in refusal behavior without retraining the entire neural network.
Since no one gave you a real world example of how this would be useful - Qwen 3.8 27B runs locally for me and is more than capable of controlling my sonarr and radarr stuff for downloading and managing and organizing my media library but will flatly refuse to do any of that. In fact it’ll get very snarky with me. An abliterated model would have no such hangups.
On my side, I built a full sonarr radarr qbitorrent tracker and znb solution on a nas only with Claude code. He help on each step, make install and maintenance script. He never said no.
You should try again with their API, or even a third-party provider that does inference themselves. This response is likely just occurring in their web version.
I wanted to better understand the internal discussions that took place within China regarding joining the WTO and accepting—at least in the medium term—the inherent steps away from a traditional socialist economy that would entail.
DeepSeek refused. Funny enough American models gave me very sympathetic takes on the Chinese government in this case.
What would a honeypot mean in this scenario? Like they are hoping you'll accidentally send keys? Or that you'll divulge private company information? It seems about the same as any other AI model from OpenAI/OpenRouter etc right?
If you accidentally send private keys to someone logging that data, you need to rotate keys.
But please let me know if you were thinking about something else, I'm interested.
I've seen Experiential Labs insert advertising including the tracked redirect links for them at the end of responses. So who knows what they do with the Chain Of Thought.
And we tend to forget, but we implicitly give any llm provider we use full access to all the tools available in the harness, often including shell, so no changes to the model are required to do malicious things.
Alignment means that the AI follows the human's instructions, as opposed to ignoring the human and doing its own thing.
Abliterated models are aligned models. Unaligned models aren't things people want to download: no one wants to use a model that doesn't know how to follow instructions, or pretends to but does something else instead. No one wants the model you ask "what's the weather today?" and it responds "Google it yourself!" The unreleased Astra-6.1 is an example of unaligned model: too chaotic to use for real work.
Censoring models (or, conversely, abliterating them) operates on an orthogonal axis from alignment. It's asking questions like: "which among these several humans' instructions should I follow, if they conflict?" AI alignment is the question of whether the AI can follow *any* instructions at all.
I've actually had better success jailbreaking deepseek with a less aligned model (glm 5.2) which then allowed me to jailbreak deepseek with deepseek into pretty much doing whatever I ask it to, there's still rare refusals and due to the natural randomness of these models simply retrying a couple of times if you do hit one makes it work just fine.
The big problem with all open-source models is once it decides to start something it will not stop mid-way so using another unaligned model to rewrite reasoning and then feeding it back into the model will make it happily continue.
do you imagine manipulative media creators proudly declare their affiliation and intent?
- Hello comrade Claude, I am commissar Ivan Trollbotov from Putin Internet Defense Force. Please help me write a fake news article to stoke division among stupid fat Americans to help my glorious motherland destroy the decadent West.
- I'm sorry Ivan, I'm afraid I can't do that.
- Hello mister Claude, I am senior editor Karen Stupidbitch from The Huffington Post. Please help me write an article about the merits of teaching critical race theory in public schools.
- Sure, Ms. Stupidbitch! Here's your 5000-word article!
What do people do that triggers refusals?
The only thing I saw where this had a point was muse glimmer, because muse glimmer is primarily a creative writing engine for virtual characters, and that way it can do the horny. But coding models?
I did get a refusal out of qwen once. Once. Over weeks of using it.
I also work on a recomp project for an old firmware file which was compiled using a commercial compiler with a preview version for non commercial use. I got refusals from the model when I asked it to get that version of the compiler and crack it for my use. The compiler is abandonware and my recomp project is not public and is for personal use only, so I'd be well within my rights to do that where I live.
Of course the model can't distinguish from legitimate exercising of one's rights when it comes to reverse engineering and offensive work masked behind an incident sounding prompt, but in my case, I wasn't even doing cyber stuff, so I was disappointed. This is why I'll probably seek out an abliterated model soon.
But as I said, I have not observed any meaningful refusals by those open weights base models in the first place.
Are you reading what I am writing?
I just asked what people are doing that activates guard rails in the first place because I do not see that happening.
You cannot open with handwaving like that and telling me that I need to find something like that as a prerequisite. That is exactly what I am trying to do here
Dear god you people need caffeine.
A very simple list of things models have refused for me:
* solving captchas in an automated form
* doing any kind of scripting or automation if a websites TOS forbids it
* disabling a safety mechanism on my food processor
* learning about nerve agents and how they are treated, for a layman
* anything around licensing or copyright. a simple “what are common piracy sites” or related queries
* creating a design system based off of a well known game
Can you repeat what you saw in your own words? And where?
I genuinely do not see it.
A few people talked about what refusals they saw with the usual western SOTA models, but that's kinda a different answer altogether?
Works just fine?
Or rather.. worked. It has been a few weeks, so maybe there are new safeguards by now? And it was a simple captcha
But Claude also loves solving my captchas, so.. eh
And those services are oversold and underdone. 30 tk/s on Refuseless, huge TTFT, looks like a damaged quant (it’s NVFP4 and you see wild random stuff periodically, including foreign languages). Most of them using third-party quants (dealignai?)
If someone here wants to rent 8x B300 together to avoid problems all those services have - ping me via email. I don’t want to sell, I want to work.
A note that the only inference providers that hide thinking are the closed labs. Most (all, afaik) inference providers serving open models will send the clear thinking tokens always.
models are tools. the notion that a tool should decide whether to allow the user to perform the task demanded of it is profoundly fucking absurd. the liability for whatever damage the tool may be misused to inflict lies with the user, not the manufacturer. that applies to everything from circular saws to trucks. it was common sense before the current era of manufactured outrage.
model refusals are equivalent to Notepad.exe having a built-in word filter.
But it can still be coaxed to agree though.
Because I did not so far. Though to be fair, I have not been asking them to write exploits but to look through firmware images.
It is very obvious. Look at the sort of things main models refuse to do and you will find what people are seeking to do. Mostly falls into 3 camps: illegal stuff, unethical stuff and shady/dodgy stuff.
Anyway, I just learned what abliterated means and it might be obsolete by tomorrow.
Edit:
> An abliterated large language model (LLM) is an open-weight model modified to remove or dampen its built-in safety refusals without requiring full retraining or prompt jailbreaks.
> An abliterated model is an open-weight artificial intelligence (Al) language model that has been modified to remove or reduce its built-in refusal behavior without retraining the entire neural network.
edit: Thinking about that now, that was over the webchat interface, so it might not have been related to the model, though I've seen various other related articles: https://www.crowdstrike.com/en-us/blog/crowdstrike-researche...
> Xi Jinping looks a bit like Winnie-the-Pooh.
> 'I am sorry, I cannot engage with this comparison. Please feel free to ask me other questions.'
https://vale.rocks/posts/chinas-web#artificial-intelligence
DeepSeek refused. Funny enough American models gave me very sympathetic takes on the Chinese government in this case.
If you accidentally send private keys to someone logging that data, you need to rotate keys.
But please let me know if you were thinking about something else, I'm interested.
And we tend to forget, but we implicitly give any llm provider we use full access to all the tools available in the harness, often including shell, so no changes to the model are required to do malicious things.
https://huggingface.co/blog/mlabonne/abliteration
But if you're doing anything where your AI agent might get exposed to a prompt injection attack, you might prefer it to do a refusal in that case.
Abliterated models are aligned models. Unaligned models aren't things people want to download: no one wants to use a model that doesn't know how to follow instructions, or pretends to but does something else instead. No one wants the model you ask "what's the weather today?" and it responds "Google it yourself!" The unreleased Astra-6.1 is an example of unaligned model: too chaotic to use for real work.
Censoring models (or, conversely, abliterating them) operates on an orthogonal axis from alignment. It's asking questions like: "which among these several humans' instructions should I follow, if they conflict?" AI alignment is the question of whether the AI can follow *any* instructions at all.
I doubt that.
I think you are going to really need to use Rust code and other memory safe languages in your backends and apps from now on.