KempnerInstitute / overcompleteLinks
π Overcomplete is a Vision-based SAE Toolbox
β71Updated this week
Alternatives and similar repositories for overcomplete
Users that are interested in overcomplete are comparing it to the libraries listed below
Sorting:
- π Code for : "CRAFT: Concept Recursive Activation FacTorization for Explainability" (CVPR 2023)β66Updated 2 years ago
- LENS Projectβ48Updated last year
- A toolkit for quantitative evaluation of data attribution methods.β53Updated 3 weeks ago
- β14Updated 3 months ago
- Official code for the ICML 2024 paper "The Entropy Enigma: Success and Failure of Entropy Minimization"β53Updated last year
- Natural Language Descriptions of Deep Visual Features, ICLR 2022β65Updated 2 years ago
- Spurious Features Everywhere - Large-Scale Detection of Harmful Spurious Features in ImageNetβ32Updated last year
- Personal implementation of ASIF by Antonio Norelliβ25Updated last year
- Uncertainty-aware representation learning (URL) benchmarkβ105Updated 4 months ago
- NeuroSurgeon is a package that enables researchers to uncover and manipulate subnetworks within models in Huggingface Transformersβ41Updated 5 months ago
- Repository for PURE: Turning Polysemantic Neurons Into Pure Features by Identifying Relevant Circuits, accepted at CVPR 2024 XAI4CV Worksβ¦β18Updated last year
- ViT Prisma is a mechanistic interpretability library for Vision and Video Transformers (ViTs).β292Updated 2 weeks ago
- Build and train Lipschitz-constrained networks: PyTorch implementation of 1-Lipschitz layers. For TensorFlow/Keras implementation, see htβ¦β32Updated this week
- [NeurIPS 2024] Official implementation of the paper "MambaLRP: Explaining Selective State Space Sequence Models".β41Updated 9 months ago
- Universal Neurons in GPT2 Language Modelsβ30Updated last year
- MetaQuantus is an XAI performance tool to identify reliable evaluation metricsβ37Updated last year
- Codebase for Mechanistic Mode Connectivityβ15Updated 2 years ago
- Git Re-Basin: Merging Models modulo Permutation Symmetries in PyTorchβ77Updated 2 years ago
- Code release for REPAIR: REnormalizing Permuted Activations for Interpolation Repairβ48Updated last year
- Sparse and discrete interpretability tool for neural networksβ63Updated last year
- Recycling diverse modelsβ45Updated 2 years ago
- Sparse Autoencoder Training Libraryβ54Updated 3 months ago
- [NeurIPS 2024] Code for the paper: B-cosification: Transforming Deep Neural Networks to be Inherently Interpretable.β33Updated 2 months ago
- Data for "Datamodels: Predicting Predictions with Training Data"β97Updated 2 years ago
- This repository includes code to reproduce the tables in "Loss Landscapes are All You Need: Neural Network Generalization Can Be Explaineβ¦β37Updated 2 years ago
- Replicating and dissecting the git-re-basin project in one-click-replication Colabsβ36Updated 2 years ago
- ModelDiff: A Framework for Comparing Learning Algorithmsβ59Updated last year
- An eXplainable AI toolkit with Concept Relevance Propagation and Relevance Maximizationβ131Updated last year
- Official PyTorch implementation of improved B-cos modelsβ51Updated last year
- TopoTrans: Optimal Transport meets Topological Data Analysisβ14Updated 2 years ago