| license: apache-2.0 | |
| datasets: | |
| - Lambent/schwartz-value-dpo | |
| base_model: | |
| - Lambent/Qwen3.5-9B-Base-Thoughtful-Interiority | |
| pipeline_tag: image-text-to-text | |
| A version of the base model lightly steered towards humane values. | |
| Methodology: | |
| Generated steering vectors for Lambent/Qwen3.5-9B-Base-Thoughtful-Interiority based on system prompts adapted from Schwartz portrait values. | |
| Relevant vectors for this model had the positive direction pointed at Benevolence and Universalism; and negative direction pointed at Achievement and Power. | |
| Asked GLM-5 to create scenarios that would test values against each other on these axes. | |
| Created a DPO dataset of 100 chosen/rejected based on the model's answers to those scenarios under the vector. | |
| Trained on DPO for the following iterations at batch size 1 and LoRA rank 256: | |
| 2e-7 for 4 epochs; 5e-6 for 1 epoch; 2e-7 for 4 epochs |