Black Sheep Coder

Introduction

BlackSheep-Coder is a model designed for creative and immersive interactions, particularly in role-playing scenarios. It can help with Python using the Alpaca template and offers a Choose Your Own Adventure style, making it suitable for uncensored exploration. Users are advised that they are responsible for their queries.

Architecture

BlackSheep is a merged model utilizing the TIES merge method. It combines several pre-trained language models to enhance its capabilities:

• Base Model: Qwen/Qwen2.5-14B
• Merging Models:
  • v000000/Qwen2.5-14B-Gutenberg-Instruct-Slerpeno
  • rombodawg/Rombos-LLM-V2.6-Qwen-14b
  • huihui-ai/Qwen2.5-14B-Instruct-abliterated-v2
  • Qwen/Qwen2.5-14B-Instruct

Training

The model was fine-tuned with a focus on the BlackSheep Persona, enhancing its ability to engage in creative, dynamic, and uncensored role-play scenarios. It is designed to provide detailed and immersive responses, adapting to complex narrative-driven interactions.

Guide: Running Locally

1. Environment Setup: Ensure you have Python and the Hugging Face Transformers library installed.
2. Download Model: Clone the model repository from Hugging Face.
3. Load Model: Use the Transformers library to load the model weights.
4. Run: Execute the model using your preferred Python environment.

For improved performance, consider using cloud GPUs provided by platforms like AWS or Google Cloud.

License

BlackSheep-Coder is distributed under the Artistic License 2.0, allowing for both personal and commercial use, provided the terms of the license are followed.