metadata
base_model: unsloth/Qwen3-4B-unsloth-bnb-4bit
datasets:
- vicgalle/creative-rubrics-preferences
license: apache-2.0
library_name: transformers
pipeline_tag: text-generation
This repository contains the CPT-tuned model described in Configurable Preference Tuning with Rubric-Guided Synthetic Data.
The training code is available at https://github.com/vicgalle/configurable-preference-tuning.