Phantomcloak19/gemma-dpo-full

Gemma-2 DPO Fine-tuned Model

• Developed by: Phantomcloak19
• License: Apache-2.0
• Base model: unsloth/gemma-2-2b-bnb-4bit
• Training framework: Unsloth + TRL (DPO)

This Gemma-2 (2B) model has been fine-tuned using Direct Preference Optimization (DPO) to reduce hallucinations and improve factual consistency, trained on the Unified Hallucination Benchmark.

The model was trained ~2× faster using Unsloth, enabling efficient low-VRAM fine-tuning.