---
datasets:
- trl-lib/ultrafeedback_binarized
base_model:
- ComparisonPO/Llama-3.2-3B-Instruct-DPO
---