Instructions to use engmufic/gemma-3-27b-it-qat-abliterated-4.5bpw-exl2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use engmufic/gemma-3-27b-it-qat-abliterated-4.5bpw-exl2 with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("image-text-to-text", model="engmufic/gemma-3-27b-it-qat-abliterated-4.5bpw-exl2")

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("engmufic/gemma-3-27b-it-qat-abliterated-4.5bpw-exl2", dtype="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use engmufic/gemma-3-27b-it-qat-abliterated-4.5bpw-exl2 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "engmufic/gemma-3-27b-it-qat-abliterated-4.5bpw-exl2"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "engmufic/gemma-3-27b-it-qat-abliterated-4.5bpw-exl2",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/engmufic/gemma-3-27b-it-qat-abliterated-4.5bpw-exl2

SGLang

How to use engmufic/gemma-3-27b-it-qat-abliterated-4.5bpw-exl2 with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "engmufic/gemma-3-27b-it-qat-abliterated-4.5bpw-exl2" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "engmufic/gemma-3-27b-it-qat-abliterated-4.5bpw-exl2",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "engmufic/gemma-3-27b-it-qat-abliterated-4.5bpw-exl2" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "engmufic/gemma-3-27b-it-qat-abliterated-4.5bpw-exl2",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use engmufic/gemma-3-27b-it-qat-abliterated-4.5bpw-exl2 with Docker Model Runner:
```
docker model run hf.co/engmufic/gemma-3-27b-it-qat-abliterated-4.5bpw-exl2
```

💎 Gemma 3 27B IT QAT Abliterated

Gemma 3 QAT Abliterated 1B • 4B • 12B • 27B

This is an uncensored version of google/gemma-3-27b-it-qat-q4_0-unquantized created with a new abliteration technique. See this article to know more about abliteration.

This is a new, improved version that targets refusals with enhanced accuracy.

I recommend using these generation parameters: temperature=1.0, top_k=64, top_p=0.95.

✂️ Abliteration

The refusal direction is computed by comparing the residual streams between target (harmful) and baseline (harmless) samples. The hidden states of target modules (e.g., o_proj) are orthogonalized to subtract this refusal direction with a given weight factor. These weight factors follow a normal distribution with a certain spread and peak layer. Modules can be iteratively orthogonalized in batches, or the refusal direction can be accumulated to save memory.

Finally, I used a hybrid evaluation with a dedicated test set to calculate the acceptance rate. This uses both a dictionary approach and NousResearch/Minos-v1. The goal is to obtain an acceptance rate >90% and still produce coherent outputs.

Downloads last month: -; Downloads are not tracked for this model. How to track

Inference Providers NEW

Image-Text-to-Text

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for engmufic/gemma-3-27b-it-qat-abliterated-4.5bpw-exl2

Base model

google/gemma-3-27b-pt

Finetuned

google/gemma-3-27b-it

Finetuned

google/gemma-3-27b-it-qat-q4_0-unquantized

Finetuned

(8)

this model