Sparse probing and activation patching of AI-text detection neurons in frozen BERT (EMNLP 2026)
-
Updated
Sep 14, 2026 - Python
#
sparse-probing
Here are 2 public repositories matching this topic...
Automated detection, visualization and suppression of hallucination-associated neurons in open-source LLMs — LLM mechanistic interpretability research tool
ai-safety pacmap model-editing mechanistic-interpretability transformerlens llm-hallucination llm-alignment h-neurons sparse-probing interpretability-research
-
Updated
Mar 19, 2026
Add this topic to your repo
To associate your repository with the sparse-probing topic, visit your repo's landing page and select "manage topics."