Skip to content
#

interpretability-ai

Star

Here are 3 public repositories matching this topic...

Language: All
Filter by language
All 3 Jupyter Notebook 1 Python 1

AliceArchiveX / concept-surgery

Star 3

Located the number "five" inside Qwen2.5-32B and DeepSeek-V4-Flash (304B) and replaced it with "four" at 128 neurons/layer. The model still reads 5 but computes with it as 4, and in every test we ran it never noticed. Full records, reproducible on low-end hardware.

ai-safety interpretability llm model-editing knowledge-editing mechanistic-interpretability llm-security causal-intervention open-source-llm qwen deepseek activation-patching representation-engineering activation-steering activation-engineering interpretability-ai concept-editing
  • Updated Oct 12, 2026
  • Python

HoussemDs / brain-xai-faithfulness

Star 0

Validating attribution methods for 3D brain MRI: faithfulness, randomized-weight sanity checks, and corrected region tests on a frozen SFCN using public pediatric data.

deep-learning medical-imaging neuroimaging brain-mri interpretability neuroimaging-analysis explainable-ai neuroimaging-data-science saliency-maps medical-imaging-with-deep-learning brain-mri-images explainable-ai-in-medical-imaging interpretability-methods medical-imaging-classification interpretability-and-explainability explainable-ai-xai interpretability-ai interpretability-research medical-imaging-analysis interpretability-experiments
  • Updated Nov 9, 2026
  • Jupyter Notebook

Add this topic to your repo

To associate your repository with the interpretability-ai topic, visit your repo's landing page and select "manage topics."

Learn more