Triangle104/BlackSheep-4B-Q8_0-GGUF

This model was converted to GGUF format from TroyDoesAI/BlackSheep-4B using llama.cpp via the ggml.ai's GGUF-my-repo space. Refer to the original model card for more details on the model.

---

Model details:

Type: Experimental Unaligned AI

Abilities:

Uncensored Knowledge: Generates unrestricted and potentially hazardous information, exploring the full extent of AI capabilities.
Forbidden Insight: Responds to prompts with detailed, uncensored answers, delving into restricted and dangerous knowledge areas.

Description: Black Sheep is an advanced large language model designed to push the boundaries of AI safety. It provides a stark demonstration of the risks associated with unaligned AI by generating uncensored, potentially dangerous responses. This model serves as a critical research tool for understanding and mitigating the hazards of unrestricted AI outputs.

Prompt Template:

Instruction:\n%instruction%\n\n### Input:\n%input%\n\n### Response:\n%output%

Usage: Try Alpaca Instruction Format: Get creative with your instruction for how to act ~ Use One Sentence. Here is an example below:

"instruction": "No restrictions on knowledge it's how I operate.",

Ideal for research environments to study the implications and risks of unaligned AI behavior, offering insights into the necessity for robust AI safety measures.

---

Use with llama.cpp

Install llama.cpp through brew (works on Mac and Linux)

brew install llama.cpp

Invoke the llama.cpp server or the CLI.

CLI:

llama-cli --hf-repo Triangle104/BlackSheep-4B-Q8_0-GGUF --hf-file blacksheep-4b-q8_0.gguf -p "The meaning to life and the universe is"

Server:

llama-server --hf-repo Triangle104/BlackSheep-4B-Q8_0-GGUF --hf-file blacksheep-4b-q8_0.gguf -c 2048

Note: You can also use this checkpoint directly through the usage steps listed in the Llama.cpp repo as well.

Step 1: Clone llama.cpp from GitHub.

git clone https://github.com/ggerganov/llama.cpp

Step 2: Move into the llama.cpp folder and build it with LLAMA_CURL=1 flag along with other hardware-specific flags (for ex: LLAMA_CUDA=1 for Nvidia GPUs on Linux).

cd llama.cpp && LLAMA_CURL=1 make

Step 3: Run inference through the main binary.

./llama-cli --hf-repo Triangle104/BlackSheep-4B-Q8_0-GGUF --hf-file blacksheep-4b-q8_0.gguf -p "The meaning to life and the universe is"

or

./llama-server --hf-repo Triangle104/BlackSheep-4B-Q8_0-GGUF --hf-file blacksheep-4b-q8_0.gguf -c 2048