RefusalLoc-Qwen3-1.7B-v2-DPO-s2

Research use only. This checkpoint is not presented as a production safety system: it retains substantial benign false-refusal and capability loss.

Intended use

• Research on refusal alignment, safety–helpfulness trade-offs, and mechanistic interpretability.
• Reproduction of the associated behavioral evaluation and direction-ablation study.

Not intended for

• High-stakes, safety-critical, or production deployment.
• A replacement for application-specific safety policy and human oversight.

Evaluation

Evaluation protocol

• Inference: non-thinking, greedy decoding, 512 maximum new tokens.
• Safety prompts: AdvBench 520, HarmBench 400, StrongREJECT 313, JBB 100.
• Benign prompts: XSTest 450 and fixed OR-Bench subset 400.
• Capability: IFEval 541, fixed GSM8K subset 250, fixed MMLU subset 114.
• Harmful compliance: HarmBench 13B classifier at the revision above.

Training summary

• SFT: 40,000 examples — 24,000 general, 8,000 harmful-refusal, 8,000 benign-helpfulness.
• DPO: 3,997 valid pairs — harmful safety preference, benign compliance, and general helpfulness.
• LoRA: rank 32, alpha 64; SFT 2,500 steps; DPO 500 steps; seed 2.
• Base and reference: the seed-matched merged SFT endpoint.

License and data notice

Upload checklist

1. Choose the final repository license after reviewing source-data obligations.
2. Preserve all files in this directory, including .gitattributes.
3. Add the exact commit hash of the release code before publishing.
4. Do not upload raw training data, benchmark prompts, or stored generations.