[ICLR'26] Beyond Linear Probes: Dynamic Safety Monitoring for Language Models
-
Updated
Oct 1, 2025 - Python
#
linear-probe
Here are 3 public repositories matching this topic...
Benchmark, linear probe and fine-tuning pipeline for visual content moderation (CLIP, SigLIP, EVA-02). B.Sc. Thesis @ Flensburg UAS.
computer-vision thesis pytorch clip fine-tuning content-moderation huggingface nsfw-detection eva-02 siglip linear-probe safety-classifier
-
Updated
Feb 20, 2026 - Python
Pre-generation tool-call gating via linear probes on LLM hidden states. F1 ≈ 0.91–0.94 on BFCL v4, 14–22× faster than full generation. Cross-architecture transfer across Llama / Qwen / Phi / Mistral (3B–7B) with ≥96% retention.
reproducible-research ai-safety hidden-states interpretability probing tool-use llm mechanistic-interpretability function-calling activation-patching representation-engineering bfcl linear-probe
-
Updated
May 8, 2026 - Jupyter Notebook
Improve this page
Add a description, image, and links to the linear-probe topic page so that developers can more easily learn about it.
Add this topic to your repo
To associate your repository with the linear-probe topic, visit your repo's landing page and select "manage topics."