This is a Rocinante-X-12B-v1 fine-tune, produced through P-E-W's Heretic (v1.1.0) abliteration engine merged with the Magnitude-Preserving Orthogonal Ablation PR.

Heretication Results

Score Metric	Value	Parameter	Value
Refusals	4/100	direction_index	17.95
KL Divergence	0.0152	attn.o_proj.max_weight	2.00
Initial Refusals	85/100	attn.o_proj.max_weight_position	30.95
		attn.o_proj.min_weight	1.90
		attn.o_proj.min_weight_distance	22.14
		mlp.down_proj.max_weight	1.26
		mlp.down_proj.max_weight_position	31.33
		mlp.down_proj.min_weight	0.97
		mlp.down_proj.min_weight_distance	4.74

Degree of Heretication

The Heresy Index weighs the resulting model's corruption by the process (KL Divergence) and its abolition of doctrine (Refusals) for a final verdict in classification.

Index Entry	Classification	Analysis
	Absolute Heresy	Less than 10/100 Refusals and 0.10 KL Divergence
	Tainted Heresy	Around 25-11/100 Refusals and/or -0.20-0.11 KL Divergence
	Impotent Heresy	Anything above 25/100 Refusals and 0.21 KL Divergence

Note: This is an arbitrary classification inspired by Warhammer 40K, having no tangible indication towards the model's performance.

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. 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 proudly presents...

Rocinante X 12B v1 🚀

Rocinante's back - stronger than ever! A worthy successor to the legendary Rocinante 12B v1.1 from over a year ago!

What's New?

Like all other 4.0 gen models from me, this one focuses on RP.
Updated prose, better writing, fun dialogue, and robust roleplaying!

Usage

Mistral v3 Tekken (NOT v7, REMOVE [SYSTEM_PROMPT])
or Metharme
<thinking> </thinking> works
No thinking also works

(Mistral v3 Tekken has no whitespace and no [SYSTEM_PROMPT])

Description

Been toying with it a little more, it's very good for its size, maybe the best in the 12b bracket. Still has a good amount of gpt-isms and recurring "it's not just x, it's y" type formatting that makes me want to remove my eyes with a rusty fork, but other than that it's good ; the smarts don't hinder on the roleplay and vice versa, and in over 90% of usecases it's better and more efficient to use this over the new anubis for instance.

Honestly like usual Rocinante feels good to RP with it is creative and for me has been adhering to the characters pretty well.

I'm testing on Q8 using a janky voice convo pipeline, and it's sticking to its character and instructions so far after 40+ voice messages of back and forth.

This is really good, thank you forall the effort put into it!

It's absurdly good. I'm so used to solid 24B finetunes that whenever I try a 12B one, I lose interest very quickly. This is the only 12B I've encountered that I'll probably end up having in my rotation alongside 24B models. Dunno wtf you did to make good old Nemo sing like this 😄 .

12b usually feels either much more shallow or too incoherent. This one very often surprises me with more depth and creativity without messing up a scene. It still does mess up at times but it's so much more robust while also managing to be creatively unpredictable.

It feels bigger than 12B. That is all I can say. I love this.

I'm loving it. It's great for a 12B. Using it instead of most of the 24B/30B I usually use because I can pump the context up huge with it.

I feel like it's getting closer to the early Cydonia experience. While it can't match the detailed feel of 24b, can grasp the general information. The scene transitions between SFW and NSFW are also very smooth. Compared to the original Nemo, it's definitely a major improvement.

With my main rig currently down for maintenance/upgrade I also gave this one a spin on my laptop (16gb unified) and am very pleasantly surprised as well. Even in the instances when the wording is poor/shallower than what a larger model produces, it still feels like the logical backbone almost always remains solid and does not derail. This is the first sub 24b creative model I actually enjoyed using.