Hi-Satoh/adv_sft_dpo_final_11_merged
Hi-Satoh/adv_sft_dpo_final_11_merged is a 4 billion parameter instruction-tuned causal language model, fine-tuned from Qwen/Qwen3-4B-Instruct-2507 by Hi-Satoh using Direct Preference Optimization (DPO). This model is specifically optimized to improve reasoning capabilities, particularly Chain-of-Thought, and enhance structured response quality. It leverages a 32768 token context length and is designed for applications requiring aligned and coherent outputs.
Loading preview...
Model Overview
Hi-Satoh/adv_sft_dpo_final_11_merged is a 4 billion parameter language model derived from Qwen/Qwen3-4B-Instruct-2507. It has been fine-tuned using Direct Preference Optimization (DPO) via the Unsloth library, with its LoRA configuration (r=8, alpha=16) fully merged into the base model.
Key Capabilities
- Enhanced Reasoning: Optimized to improve Chain-of-Thought reasoning, enabling more structured and logical response generation.
- Improved Response Quality: DPO training aligns the model's outputs with preferred responses, leading to higher quality and more coherent interactions.
- Full Merged Weights: The repository provides full-merged 16-bit weights, simplifying deployment as no adapter loading is required.
Training Details
The model underwent 1 epoch of DPO training with a learning rate of 3e-07 and a beta value of 0.1. It utilized a maximum sequence length of 4096 during training. The training data used was [Hi-Satoh/test_dpo_dataset].
Licensing
This model operates under the MIT License, as per the dataset terms. Users must also adhere to the original base model's license terms.