wesg52 / universal-neuronsLinks
Universal Neurons in GPT2 Language Models
☆30Updated last year
Alternatives and similar repositories for universal-neurons
Users that are interested in universal-neurons are comparing it to the libraries listed below
Sorting:
- Sparse Autoencoder Training Library☆54Updated 5 months ago
- ☆23Updated 8 months ago
- Code for reproducing our paper "Not All Language Model Features Are Linear"☆80Updated 10 months ago
- ☆53Updated last year
- Gemstones: A Model Suite for Multi-Faceted Scaling Laws (NeurIPS 2025)☆29Updated last week
- Open source replication of Anthropic's Crosscoders for Model Diffing☆59Updated 11 months ago
- A library for efficient patching and automatic circuit discovery.☆77Updated 2 months ago
- Sparse and discrete interpretability tool for neural networks☆63Updated last year
- Delphi was the home of a temple to Phoebus Apollo, which famously had the inscription, 'Know Thyself.' This library lets language models …☆215Updated this week
- ☆106Updated 7 months ago
- ☆23Updated last year
- ☆33Updated 8 months ago
- Attribution-based Parameter Decomposition☆30Updated 3 months ago
- PyTorch and NNsight implementation of AtP* (Kramar et al 2024, DeepMind)☆19Updated 8 months ago
- Stanford NLP Python library for benchmarking the utility of LLM interpretability methods☆134Updated 3 months ago
- Code to reproduce key results accompanying "SAEs (usually) Transfer Between Base and Chat Models"☆12Updated last year
- ☆123Updated last year
- ☆85Updated last year
- ☆71Updated last month
- ☆91Updated last year
- Code for NeurIPS 2024 Spotlight: "Scaling Laws and Compute-Optimal Training Beyond Fixed Training Durations"☆84Updated 11 months ago
- Code for my NeurIPS 2024 ATTRIB paper titled "Attribution Patching Outperforms Automated Circuit Discovery"☆41Updated last year
- ☆20Updated last year
- Applying SAEs for fine-grained control☆24Updated 9 months ago
- ☆23Updated 10 months ago
- ☆45Updated last year
- Code to reproduce "Transformers Can Do Arithmetic with the Right Embeddings", McLeish et al (NeurIPS 2024)☆193Updated last year
- [NeurIPS 2024] Goldfish Loss: Mitigating Memorization in Generative LLMs☆92Updated 10 months ago
- A tiny easily hackable implementation of a feature dashboard.☆15Updated 2 weeks ago
- Language models scale reliably with over-training and on downstream tasks☆100Updated last year