bartbussmann / BatchTopKLinks
Implementation of the BatchTopK activation function for training sparse autoencoders (SAEs)
☆41Updated last month
Alternatives and similar repositories for BatchTopK
Users that are interested in BatchTopK are comparing it to the libraries listed below
Sorting:
- ☆34Updated 5 months ago
- Sparse Autoencoder Training Library☆52Updated last month
- ☆101Updated 3 weeks ago
- Open source replication of Anthropic's Crosscoders for Model Diffing☆55Updated 7 months ago
- ☆95Updated 4 months ago
- ☆26Updated 2 years ago
- ☆18Updated last year
- Delphi was the home of a temple to Phoebus Apollo, which famously had the inscription, 'Know Thyself.' This library lets language models …☆185Updated last week
- Official implementation of FIND (NeurIPS '23) Function Interpretation Benchmark and Automated Interpretability Agents☆49Updated 9 months ago
- ☆57Updated this week
- ☆44Updated 7 months ago
- Sparse and discrete interpretability tool for neural networks☆63Updated last year
- Code for reproducing our paper "Not All Language Model Features Are Linear"☆75Updated 6 months ago
- [ICLR 2025] Monet: Mixture of Monosemantic Experts for Transformers☆69Updated this week
- Official repo for the paper "Weight-based Decomposition: A Case for Bilinear MLPs"☆21Updated 6 months ago
- Multi-Layer Sparse Autoencoders (ICLR 2025)☆22Updated 4 months ago
- ☆97Updated 4 months ago
- Official implementation of "BERTs are Generative In-Context Learners"☆28Updated 3 months ago
- ☆85Updated 10 months ago
- ☆19Updated 2 months ago
- ☆14Updated last year
- This repository contains the code used for the experiments in the paper "Fine-Tuning Enhances Existing Mechanisms: A Case Study on Entity…☆27Updated last year
- ☆121Updated last year
- Code for NeurIPS 2024 Spotlight: "Scaling Laws and Compute-Optimal Training Beyond Fixed Training Durations"☆74Updated 7 months ago
- ☆23Updated 4 months ago
- ☆28Updated 3 months ago
- ☆17Updated last year
- ☆12Updated last month
- Code for my NeurIPS 2024 ATTRIB paper titled "Attribution Patching Outperforms Automated Circuit Discovery"☆35Updated last year
- ☆37Updated last month