Model Overview

This model, hiro7ka/dpo-qwen-cot-merged-ver3d, is a 4 billion parameter language model derived from Qwen/Qwen3-4B-Instruct-2507. It has undergone Direct Preference Optimization (DPO) using the Unsloth library, resulting in a merged 16-bit weight model that requires no adapter loading.

Key Capabilities

• Enhanced Reasoning: Optimized to improve Chain-of-Thought (CoT) reasoning, allowing for more structured and logical response generation.
• Improved Response Quality: Fine-tuned to align outputs with preferred responses, leading to higher quality and more relevant text.
• Direct Use: As a fully merged model, it can be used directly with the transformers library for inference.

Training Details

The model was trained for 0.5 epochs with a learning rate of 1e-08 and a beta value of 0.1, using a maximum sequence length of 1024. The DPO training leveraged the u-10bei/dpo-dataset-qwen-cot dataset. Users should adhere to the MIT License of the dataset and the original base model's license terms.