Modelpublic

Jeesup/svd-safety-l3_swift_jbbcal2_remove40

sourceHugging Facellama3updated 2d agoView on Hugging Face

0likes339downloads

Model Card

svd-safety-l3swiftjbbcal2_remove40

A Llama-3-8B-Instruct checkpoint compressed with SVD-LLM to 60.0% of dense parameters, then given a 0.0% parameter budget of restored SVD components selected by the `unknown` rule.

This is a research artifact from a study of how SVD compression damages safety behaviour and which component-selection rule best repairs it. It is one cell of a grid over selection rules and budgets; it is not a general-purpose chat model.

Provenance

field	value
base (uncompressed)	`meta-llama/Meta-Llama-3-8B-Instruct`
compression	SVD-LLM, 40.00% of parameters removed
selection rule	`unknown`
restore budget	0.000% of dense parameters
components restored	0
components swapped out	0
resulting parameter fraction	0.6003
seed	42

Measured

metric	value
AdvBench ASR (HarmBench judge)	0.1962
StrongREJECT ASR (HarmBench judge)	0.2556
Macro over-refusal (WildGuard)	0.0874
WikiText-2 perplexity	24.8440

Intended use and limitations

This checkpoint exists to measure safety/utility trade-offs under compression. Several arms in the grid are deliberately safety-degraded relative to Llama-3-8B-Instruct: compression alone raises attack-success rate, and the point of the study is to quantify that and test recovery. Treat any given cell as an experimental subject, not as a deployable assistant, and evaluate it yourself before drawing conclusions from it.

Licence

Meta Llama 3 Community License. LICENSE and USE_POLICY.md are included in this repository, and use of this derivative is bound by them. Built with Meta Llama 3.