ml-engnr/dpo-qwen-cot-merged
The ml-engnr/dpo-qwen-cot-merged model is a 4 billion parameter language model based on the Qwen3-4B-Instruct-2507 architecture. It has been fine-tuned using Direct Preference Optimization (DPO) to enhance its reasoning capabilities, specifically Chain-of-Thought (CoT), and improve the quality of structured responses. This model is optimized for tasks requiring logical deduction and well-structured output, leveraging its 40960 token context length.
Loading preview...
Model Overview
The ml-engnr/dpo-qwen-cot-merged model is a 4 billion parameter language model, fine-tuned from the Qwen/Qwen3-4B-Instruct-2507 base model. It utilizes Direct Preference Optimization (DPO) via the Unsloth library to align its responses with preferred outputs. This model is provided as full-merged 16-bit weights, eliminating the need for adapter loading.
Key Capabilities
- Enhanced Reasoning: Optimized to improve Chain-of-Thought (CoT) reasoning, making it suitable for complex problem-solving tasks.
- Structured Response Quality: DPO training specifically targeted improving the quality and structure of generated responses based on a preference dataset.
- Direct Use: As a merged model, it can be directly loaded and used with the
transformerslibrary without additional configuration.
Training Details
The model underwent 1 epoch of DPO training with a learning rate of 5e-07 and a beta value of 0.4. The maximum sequence length used during training was 1024 tokens. LoRA configuration (r=8, alpha=16) was applied and subsequently merged into the base model.
Usage Considerations
This model is ideal for applications where robust reasoning and high-quality, structured outputs are critical. Users should adhere to the MIT License for the training data and the original base model's license terms for compliance.