huihui-ai/Qwen2.5-Coder-0.5B-Instruct-abliterated
The huihui-ai/Qwen2.5-Coder-0.5B-Instruct-abliterated is a 0.5 billion parameter instruction-tuned causal language model, derived from Qwen/Qwen2.5-Coder-0.5B-Instruct. This version has been uncensored using abliteration techniques, making it suitable for applications requiring less restrictive content generation. It is part of a series of uncensored Qwen2.5-Coder models, offering a compact option for code-related tasks and general instruction following.
Loading preview...
Overview
This model, huihui-ai/Qwen2.5-Coder-0.5B-Instruct-abliterated, is a 0.5 billion parameter instruction-tuned language model. It is an uncensored variant of the original Qwen/Qwen2.5-Coder-0.5B-Instruct model, created using an 'abliteration' process to remove refusal behaviors. This makes it suitable for use cases where content restrictions of the base model are undesirable.
Key Characteristics
- Uncensored: Modified to remove refusal tendencies present in the base Qwen2.5-Coder model.
- Compact Size: At 0.5 billion parameters, it offers a lightweight solution for various tasks.
- Instruction-tuned: Designed to follow instructions effectively.
- Code-focused: Derived from the Qwen2.5-Coder series, suggesting an orientation towards code generation and understanding.
Usage
This model can be easily integrated into applications using the Hugging Face transformers library. It also supports deployment via Ollama, with a pre-built image available for direct use. The provided Python example demonstrates how to load the model, manage conversation context, and generate responses, including options to clear chat history.