GoogleCloudPlatform/activation-model-scanner

Verify language model safety before deployment by analyzing activation patterns

1 best points · 0 comments · 1 Hacker News post · 34 GitHub stars · View on GitHub

Hacker News discussions

• AMS – Detect unsafe LLMs in 30 seconds via activation analysis (1 points, 0 comments, by gmessenger, 2026-04-30)