Instructions to use failspy/Phi-3-medium-4k-instruct-abliterated-v3-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use failspy/Phi-3-medium-4k-instruct-abliterated-v3-GGUF with llama.cpp:

Install (macOS, Linux)

curl -LsSf https://llama.app/install.sh | sh
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf failspy/Phi-3-medium-4k-instruct-abliterated-v3-GGUF
# Run inference directly in the terminal:
llama cli -hf failspy/Phi-3-medium-4k-instruct-abliterated-v3-GGUF

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf failspy/Phi-3-medium-4k-instruct-abliterated-v3-GGUF
# Run inference directly in the terminal:
llama cli -hf failspy/Phi-3-medium-4k-instruct-abliterated-v3-GGUF

Build from source code

git clone https://github.com/ggml-org/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-app
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama serve -hf failspy/Phi-3-medium-4k-instruct-abliterated-v3-GGUF
# Run inference directly in the terminal:
./build/bin/llama cli -hf failspy/Phi-3-medium-4k-instruct-abliterated-v3-GGUF

Use Docker

docker model run hf.co/failspy/Phi-3-medium-4k-instruct-abliterated-v3-GGUF

Llama new
LM Studio
Jan

vLLM

How to use failspy/Phi-3-medium-4k-instruct-abliterated-v3-GGUF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "failspy/Phi-3-medium-4k-instruct-abliterated-v3-GGUF"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "failspy/Phi-3-medium-4k-instruct-abliterated-v3-GGUF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/failspy/Phi-3-medium-4k-instruct-abliterated-v3-GGUF

Ollama
How to use failspy/Phi-3-medium-4k-instruct-abliterated-v3-GGUF with Ollama:
```
ollama run hf.co/failspy/Phi-3-medium-4k-instruct-abliterated-v3-GGUF
```
Unsloth Desktop
Docker Model Runner
How to use failspy/Phi-3-medium-4k-instruct-abliterated-v3-GGUF with Docker Model Runner:
```
docker model run hf.co/failspy/Phi-3-medium-4k-instruct-abliterated-v3-GGUF
```

Lemonade

How to use failspy/Phi-3-medium-4k-instruct-abliterated-v3-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull failspy/Phi-3-medium-4k-instruct-abliterated-v3-GGUF

Run and chat with the model

lemonade run user.Phi-3-medium-4k-instruct-abliterated-v3-GGUF-{{QUANT_TAG}}

List all available models

lemonade list

Atomic Chat

WizardLM-2-8x22B abliteration inquiry

by OrangeApples - opened May 23, 2024

Discussion

OrangeApples

May 23, 2024

•

edited May 23, 2024

@failspy saw this model https://huggingface.co/fearlessdots/WizardLM-2-7B-abliterated and it made me wonder if it's big brother would benefit from this technique as well.

This tweet https://x.com/WizardLM_AI/status/1780101465950105775 possibly suggests that it's already more uncensored than the other Microsoft models, but perhaps abliteration can take this a step further. If you think it would, would you be willing to make a WizardLM-2-8x22B-abliterated-v3?

failspy

Owner May 23, 2024

You'll note how I started the README of this one specifically ;)

OrangeApples

May 23, 2024

Nice! Didn't notice that before 👀

OrangeApples changed discussion status to closed May 23, 2024

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment