# This is a decensored version of [TheDrummer/Magidonia-24B-v4.3](https://huggingface.co/TheDrummer/Magidonia-24B-v4.3), made using [Heretic](https://github.com/p-e-w/heretic) v1.1.0 ## Abliteration parameters | Parameter | Value | | :-------- | :---: | | **direction_index** | per layer | | **attn.o_proj.max_weight** | 0.89 | | **attn.o_proj.max_weight_position** | 30.54 | | **attn.o_proj.min_weight** | 0.69 | | **attn.o_proj.min_weight_distance** | 4.37 | | **mlp.down_proj.max_weight** | 1.01 | | **mlp.down_proj.max_weight_position** | 23.69 | | **mlp.down_proj.min_weight** | 0.98 | | **mlp.down_proj.min_weight_distance** | 11.82 | ## Performance | Metric | This model | Original model ([TheDrummer/Magidonia-24B-v4.3](https://huggingface.co/TheDrummer/Magidonia-24B-v4.3)) | | :----- | :--------: | :---------------------------: | | **KL divergence** | 0.0485 | 0 *(by definition)* | | **Refusals** | 5/100 | 73/100 | ----- # Join our Discord! https://discord.gg/BeaverAI ## More than 9000 members strong 💪 A hub for users and makers alike! --- ## Drummer is open for new opportunities (I'm a Software Engineer). Contact me through any of these channels: https://linktr.ee/thelocaldrummer ### Thank you to everyone who subscribed through [Patreon](https://www.patreon.com/TheDrummer). Your support helps me chug along in this brave new world. ### FAQ for those out-of-the-loop

🐶 Who is Drummer?

Hi! I'm Drummer. I'm a Software Engineer with experience in JavaScript, Golang, Python, and generally engineering the crap out of things. Why I'm in the AI space: - **Exploration:** Everyone is trying to figure out how AI works and what it's capable of. I am too - just not in creating the smartest, safest model at all costs. - **Upskill:** The world is headed towards AI. It is here to stay. This has been my way of brushing up in this new form of computing challenge. - **Value:** I yearn to create value. I feel satisfaction and fulfillment in providing something meaningful for others. - **Fun:** It's just fun using and making models. It's also fun coming up with theories and realizing them in practice (training AI). I started my tuning venture back in mid-2024 when I wanted to improve its literary capabilities. I've come a long way since then and I have branched out and specialized. Foundational models today are optimized for non-creative uses, and I believe there is a place for AI in creativity and entertainment. I am here to take *the road less traveled by*.

❓ What are my models like?

**Bottomline:** My models are usually geared towards creativity, usability, and entertainment! While intelligence, correctness, and problem solving are not my priority, they are still one of many qualities I want in my models. The primary goal is to enhance the experience for users looking to use models for creative uses, and other use cases which require no alignment. In an effort to make it clear to myself and to others what I'm aiming for, I've identified certain qualities that my users often want: Creativity - **Writing:** Does it string together words and sentences in a pleasant & effective way? Does it feel like a writer? - **Dynamism:** How good is the AI at being compelling and intriguing in its storytelling? - **Imagination:** Can the AI navigate through a plethora of possibilities? Can it skirt incoherence and rise up to absolute coherence at the end of it? (Dis)alignment - **Attitude:** Does it refuse in both soft or hard ways? Does it lean towards certain corporate/religious/political ethics & beliefs? How does it see the user and itself? - **Morality:** Does it know ethics? Is its language infected with forced positivity? If not, can it still moralize over difficult & dubious themes? - **Formatting:** How stubborn is it with its established formatting? Can it create effective and novel formats to answer the prompt? Intelligence - **Adherence:** Can it follow instructions? Is it sticking to the prompt? Can it understsand you? - **Knowledge:** Does it know about the world in both fictional and non-fictional way? - **Perception:** Can it handle nuance, complexity, and logic? If it doesn't excel in one of these qualities, or if it's overall mediocre for its size, then I would most likely reiterate until I get something right.

💡 Philosophy

A person is defined by the language they use. Not whether they speak in English or German, but in how they perceive reality. Just like how we associate a serial killer as a mind that can't map 'murder' to 'evil', an innocent person is a mind that simply can't imagine 'murder'. They get confused when forced to deal with such subjects. AI's use of language speaks volumes about their 'perception' of reality. If a language model has been skewed and limited to a positive perception, then it's ability to imagine is also limited. Finetuning is an opportunity to adjust and broaden the language. Corporations use it to achieve safety and compliance. I'm here to

--- [Drummer](https://huggingface.co/TheDrummer) proudly presents... # Magidonia 24B v4.3 💿 ![image](https://cdn-uploads.huggingface.co/production/uploads/65f2fd1c25b848bd061b5c2e/nmZ0UYgdu9syT67CVitNR.png) ## What's New? - Much better roleplay - Might refuse/be more positive - But much better roleplay ## Usage - Mistral v7 Tekken - \[THINK\] \[/THINK\] might work - \ \ works - No thinking also works ## Description > The model often surprises me with how good some of the generated answers are. The overall level is also good. > i'm lurking here for weeks, testing out the builds in RP, this is by far the best one. It is pretty smart and aware of it's surroundings and holds the character card really well. > i tried many of your magistral finetunes and for sure this feels the best so nice job^^ > I've completed the analysis of v4zk's response, and it's excellent, with virtually no physics errors or contradictory intentions. This is significantly better than previous versions of Cydonia, which always had obvious trade-offs: either too erotic and lacking in physics, or a balance between physics and eroticism at the expense of character development, or simply inconsistency, performing terribly in certain test cases. v4zk's response is virtually error-free and novel, maintaining broad stability while allowing characters to express distinct inner intentions. I believe this is the best version of Cydonia to date. Simply put, it's good enough to approach the level of Skyfall—making up for most of the gains from parameter scaling. > It's a versatile tune, really adheres to system prompts and thinking prefill. OOC, Author's note and all that work flawlessly. Love the thinking on this one! It's good at reading between the lines when given good prompting. It will consider details from the cards in unique ways. Accepts a wide range of sampling parameters and is still stable (comparatively, it may be too creative at high temperatures). Swipes are minimal compared to other similarly-sized thinking models. Good for RP/CYOA/Dungeon/Chat/Companion/Author! You can also try its Mistral Small 3.2 counterpart: [Cydonia 24B v4.3](https://huggingface.co/TheDrummer/Cydonia-24B-v4.3) (The difference is mostly writing style, and most prefer the Magistral tune.) ## Links - Original: https://huggingface.co/TheDrummer/Magidonia-24B-v4.3 - GGUF: https://huggingface.co/TheDrummer/Magidonia-24B-v4.3-GGUF - iMatrix (recommended): https://huggingface.co/bartowski/TheDrummer_Magidonia-24B-v4.3-GGUF - EXL3: https://huggingface.co/ArtusDev/TheDrummer_Magidonia-24B-v4.3-EXL3 --- ![image](https://cdn-uploads.huggingface.co/production/uploads/65f2fd1c25b848bd061b5c2e/M_Vh3ifg4qZkRmsYQbMT8.png) Thank you to OpenRouter (and esp. @louisgv) for sponsoring my tune and providing me my first ever 'brandkit'! `config-v4zk`