Safety Beyond the Interface: Detecting Harm via Latent States in Large Language Models
Lightweight 12.6M-parameter probes on LLaMA-3.1-8B activations detect harmful prompts with F1 up to 99%, matching guard models 1000x larger at lower cost.
The study extracts activations from LLaMA-3.1-8B and trains lightweight MLP classifier probes with 12.6M parameters to detect harmful prompts. Evaluated on WildJailbreak, BeaverTails, and AEGIS 2.0, the probes achieve F1 scores of 99%, 83%, and 84%, respectively, competitive with guard models roughly 1000x larger. This internal-state approach addresses the latency, compute overhead, and assurance gap of external guardrail models in resource-constrained, time-critical deployments.
- Probes read LLaMA-3.1-8B internal activations instead of relying on external guardrail models.
- 12.6M-parameter MLP classifiers achieve F1 of 99%, 83%, and 84% on three benchmarks.
- Performance matches guard models roughly 1000x larger while cutting latency and compute.
- Targets resource-constrained, time-critical deployments where external guardrails add overhead.
- Vendors
- Meta
- AI models
- LLaMA-3.1-8B
From the original article
1 min read at arxiv.orgAutonomous systems increasingly rely on Large Language Models (LLMs) yet the safety infrastructure surrounding these models introduces latency and compute overhead. This limits utility in resource-constrained, time-critical deployments. Existing external guardrail models remain blind to the model's internal workings, creating a fundamental assurance gap.
By Alizishaan Khatri · arxiv.org
A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.