Model Overview

The tatsuji1962/dpo-qwen-cot-merged model is a specialized 4 billion parameter language model derived from Qwen/Qwen3-4B-Instruct-2507. It has undergone Direct Preference Optimization (DPO) using the Unsloth library, integrating the full 16-bit weights directly without requiring adapter loading.

Key Capabilities

• Enhanced Reasoning: Optimized to improve Chain-of-Thought (CoT) reasoning, making it suitable for tasks requiring logical progression and problem-solving.
• Improved Structured Responses: Fine-tuned to produce higher quality and more aligned structured outputs based on preference datasets.
• Direct Usage: As a fully merged model, it can be used directly with the transformers library, simplifying deployment.

Training Details

The model was trained for 1 epoch with a learning rate of 1e-07 and a beta value of 0.1, using a maximum sequence length of 1024. The training leveraged the u-10bei/dpo-dataset-qwen-cot dataset, focusing on aligning model responses with preferred outputs. The base model's license terms must be followed, and the merged model itself is released under the MIT License.