Instructions to use grimjim/Mistral-7B-Instruct-v0.2-8bit-abliterated-layer18 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use grimjim/Mistral-7B-Instruct-v0.2-8bit-abliterated-layer18 with Transformers:

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

pipe = pipeline("text-generation", model="grimjim/Mistral-7B-Instruct-v0.2-8bit-abliterated-layer18")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("grimjim/Mistral-7B-Instruct-v0.2-8bit-abliterated-layer18")
model = AutoModelForCausalLM.from_pretrained("grimjim/Mistral-7B-Instruct-v0.2-8bit-abliterated-layer18", device_map="auto")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use grimjim/Mistral-7B-Instruct-v0.2-8bit-abliterated-layer18 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "grimjim/Mistral-7B-Instruct-v0.2-8bit-abliterated-layer18"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "grimjim/Mistral-7B-Instruct-v0.2-8bit-abliterated-layer18",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/grimjim/Mistral-7B-Instruct-v0.2-8bit-abliterated-layer18

SGLang

How to use grimjim/Mistral-7B-Instruct-v0.2-8bit-abliterated-layer18 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 "grimjim/Mistral-7B-Instruct-v0.2-8bit-abliterated-layer18" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "grimjim/Mistral-7B-Instruct-v0.2-8bit-abliterated-layer18",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

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 "grimjim/Mistral-7B-Instruct-v0.2-8bit-abliterated-layer18" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "grimjim/Mistral-7B-Instruct-v0.2-8bit-abliterated-layer18",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use grimjim/Mistral-7B-Instruct-v0.2-8bit-abliterated-layer18 with Docker Model Runner:
```
docker model run hf.co/grimjim/Mistral-7B-Instruct-v0.2-8bit-abliterated-layer18
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

Mistral-7B-Instruct-v0.2-8bit-abliterated-layer18

This model was abliterated by computing a refusal vector an 8-bit bitsandbytes quant, and then applying the vector to the full weight model. Abliteration was performed locally using a CUDA GPU, the VRAM memory consumption appeared to be constrained to be under 12GB.

Layer 18 was selected for derivation of the refusal direction, as measurements of the refusal direction magnitude, signal-to-noise ratio, and angle between the means of the "harmful" and "harmless" directions suggested that intervention based on this layer would be relatively efficient and effective.

No additional fine-tuning was performed on these weights. Repair is required for proper use.

The code used can be found on Github at https://github.com/jim-plus/llm-abliteration.

(My prior attempt relied on default values within the codebase, which turned out to be less effective than this intervention.)

Downloads last month: 4

Safetensors

Model size

7B params

Tensor type

BF16

Model tree for grimjim/Mistral-7B-Instruct-v0.2-8bit-abliterated-layer18

Base model

mistralai/Mistral-7B-Instruct-v0.2

Finetuned

(1096)

this model

Quantizations

1 model