Code for our paper "Decomposing The Dark Matter of Sparse Autoencoders"
☆23Feb 6, 2025Updated last year
Alternatives and similar repositories for SAE-Dark-Matter
Users that are interested in SAE-Dark-Matter are comparing it to the libraries listed below. We may earn a commission when you buy through links labeled 'Ad' on this page.
Sorting:
- A tiny easily hackable implementation of a feature dashboard.☆17Oct 21, 2025Updated 10 months ago
- Multi-dimensional analysis of orthogonal safety directions in LLM alignment☆23Jun 12, 2026Updated 2 months ago
- Mamba support for transformer lens☆20Sep 17, 2024Updated last year
- Code for the paper "Distinguishing the Knowable from the Unknowable with Language Models"☆12Jul 18, 2026Updated last month
- Code for reproducing our paper "Are Sparse Autoencoders Useful? A Case Study in Sparse Probing"☆34Aug 15, 2026Updated 2 weeks ago
- Deploy to Railway using AI coding agents - Free Credits Offer • AdUse Claude Code, Codex, OpenCode, and more. Autonomous software development now has the infrastructure to match with Railway.
- Code for reproducing our paper "Not All Language Model Features Are Linear"☆92Nov 27, 2024Updated last year
- Code for reproducing our paper "Low Rank Adapting Models for Sparse Autoencoder Features"☆17Mar 31, 2025Updated last year
- DESSERT Effeciently Searches Sets of Embeddings via Retrieval Tables☆18Feb 21, 2024Updated 2 years ago
- Codebase for Obfuscated Activations Bypass LLM Latent-Space Defenses☆32Feb 11, 2025Updated last year
- A fast high dimensional near neighbor search algorithm based on group testing and locality sensitive hashing☆23Dec 9, 2023Updated 2 years ago
- ☆38Feb 20, 2025Updated last year
- Delphi was the home of a temple to Phoebus Apollo, which famously had the inscription, 'Know Thyself.' This library lets language models …☆275Updated this week
- The official repository of 'Unnatural Language Are Not Bugs but Features for LLMs'☆25May 20, 2025Updated last year
- Unsupervised Grammar Induction with Combinatory Categorial Grammars☆10Jan 28, 2021Updated 5 years ago
- 1-Click AI Models by DigitalOcean Gradient • AdDeploy popular AI models on DigitalOcean Gradient GPU virtual machines with just a single click. Zero configuration with optimized deployments.
- KV cache compression via sparse coding☆18Oct 26, 2025Updated 10 months ago
- ☆32May 27, 2026Updated 3 months ago
- tugHall: a simulator of cancer cell evolution based on the hallmarks of cancer, linked to the mutational states of tumor-related genes. T…☆13Dec 11, 2023Updated 2 years ago
- ☆38Apr 30, 2024Updated 2 years ago
- This is the code repository for "Uncovering Safety Risks of Large Language Models through Concept Activation Vector"☆49Oct 13, 2025Updated 10 months ago
- Implementation of the BatchTopK activation function for training sparse autoencoders (SAEs)☆67Jul 24, 2025Updated last year
- Sparse Autoencoder Training Library☆58May 1, 2025Updated last year
- TDMS 2.0 support for F# and C#☆13Dec 26, 2022Updated 3 years ago
- Sparsify transformers with SAEs and transcoders☆739Updated this week
- Managed Database hosting by DigitalOcean • AdPostgreSQL, MySQL, MongoDB, Kafka, Valkey, and OpenSearch available. Automatically scale up storage and focus on building your apps.
- Competing Genetic Algorithm to find profitable Trading Strategies on a financial market☆14May 29, 2019Updated 7 years ago
- ☆26Jun 22, 2025Updated last year
- Code for the arXiv preprint "Answer, Assemble, Ace: Understanding How Transformers Answer Multiple Choice Questions"☆15Aug 2, 2025Updated last year
- Create feature-centric and prompt-centric visualizations for sparse autoencoders (like those from Anthropic's published research).☆273Feb 27, 2026Updated 6 months ago
- Efficient Dictionary Learning with Switch Sparse Autoencoders (SAEs)☆25Dec 1, 2024Updated last year
- ☆13Dec 4, 2024Updated last year
- ☆33Nov 28, 2024Updated last year
- ☆149Aug 4, 2024Updated 2 years ago
- A benchmark for mechanistic discovery of circuits in Transformers☆18Dec 15, 2024Updated last year
- Bare Metal GPUs on DigitalOcean Gradient AI • AdPurpose-built for serious AI teams training foundational models, running large-scale inference, and pushing the boundaries of what's possible.
- ☆16Feb 21, 2025Updated last year
- ASCETIC (Agony-baSed Cancer EvoluTion InferenCe) is a novel framework for the inference of a set of statistically significant temporal pa…☆12Apr 18, 2025Updated last year
- Providing the answer to "How to do patching on all available SAEs on GPT-2?". It is an official repository of the implementation of the p…☆14Jan 26, 2025Updated last year
- ☆12Apr 19, 2022Updated 4 years ago
- ☆428Aug 21, 2025Updated last year
- Code for the paper "Large Language Models Share Representations of Latent Grammatical Concepts Across Typologically Diverse Languages" (N…☆17Apr 13, 2025Updated last year
- ☆11Apr 28, 2024Updated 2 years ago