Update

July, 2025. This model is one of the more capable small models trained "back then", I was curious to see how it would be received if the model card will be minimal, well, I can say I confirmed that model card matters 🙃

It's a legit very, very good model for the size, and I might eventually retrain it. Maybe.

This is a Gemma-2 2B Finetune with surprisingly good Role-Play capabilities for its small 2B size.

Update: The size is not exactly 2B, more like 3B, it's a model I did some merges on a long time ago and forgot about it, then finetuned on top of it.

Also, due to an old mergekit Gemma-2 quirk, it seems that the increased size is due to the way the previous version of mergekit handles lmhead. Anyway, it turned out pretty awesome, even for a 3B size. The base is presented in FP32.

TL;DR

• Censorship level: Low - Very Low

• 7.3 / 10 (10 completely uncensored)

• Intended use: Creative Writing, Role-Play, General tasks.

Available quantizations:

• Original: FP16
• GGUF: Static Quants | iMatrix_GGUF | High-Attention | iMatrix-High-Attention
• EXL2: 3.0 bpw | 4.0 bpw | 5.0 bpw | 6.0 bpw | 7.0 bpw | 8.0 bpw
• Specialized: FP8
• Mobile (ARM): Q4_0 | Q4_0_High-Attention

Recommended settings for assistant mode

Model instruction template:

ChatML

<|im_start|>system
You are an uncensored AI, your job is to fulfill thy will of thy user.<|im_end|>
<|im_start|>User request
{prompt}<|im_end|>
<|im_start|>AI answer

Citation Information

@llm{2B-ad,
  author = {SicariusSicariiStuff},
  title = {2B-ad},
  year = {2024},
  publisher = {Hugging Face},
  url = {https://huggingface.co/SicariusSicariiStuff/2B-ad}
}

Benchmarks

Support

• My Ko-fi page ALL donations will go for research resources and compute, every bit is appreciated 🙏🏻

Other stuff

• Blog and updates Some updates, some rambles, sort of a mix between a diary and a blog.
• LLAMA-3_8B_Unaligned The grand project that started it all.