alan-cooney / transformer-from-scratch
Decoder only transformer, built from scratch with PyTorch
☆25Updated 10 months ago
Related projects: ⓘ
- Create feature-centric and prompt-centric visualizations for sparse autoencoders (like those from Anthropic's published research).☆133Updated last month
- Using sparse coding to find distributed representations used by neural networks.☆162Updated 10 months ago
- ☆110Updated 3 weeks ago
- Mechanistic Interpretability Visualizations using React☆175Updated 2 months ago
- Sparse Autoencoder for Mechanistic Interpretability☆173Updated 2 months ago
- Steering vectors for transformer language models in Pytorch / Huggingface☆52Updated 2 months ago
- ☆174Updated 4 months ago
- Training Sparse Autoencoders on Language Models☆367Updated this week
- ☆99Updated 10 months ago
- Sparse autoencoders☆297Updated last week
- ☆75Updated this week
- ☆91Updated last month
- Steering Llama 2 with Contrastive Activation Addition☆83Updated 3 months ago
- ☆54Updated last week
- The nnsight package enables interpreting and manipulating the internals of deep learned models.☆356Updated this week
- ☆33Updated 3 months ago
- This repository collects all relevant resources about interpretability in LLMs☆230Updated last week
- ☆10Updated 2 months ago
- Code and results accompanying the paper "Refusal in Language Models Is Mediated by a Single Direction".☆76Updated 3 weeks ago
- Tools for studying developmental interpretability in neural networks.☆69Updated this week
- ☆267Updated 2 months ago
- Emergent world representations: Exploring a sequence model trained on a synthetic task☆165Updated last year
- Understand and test language model architectures on synthetic tasks.☆156Updated 4 months ago
- Resources for skilling up in AI alignment research engineering. Covers basics of deep learning, mechanistic interpretability, and RL.☆185Updated 7 months ago
- ☆246Updated 6 months ago
- Improving Alignment and Robustness with Circuit Breakers☆124Updated 2 months ago
- LLM experiments done during SERI MATS - focusing on activation steering / interpreting activation spaces☆73Updated last year
- ☆24Updated 5 months ago
- ☆274Updated this week
- Code for reproducing our paper "Not All Language Model Features Are Linear"☆57Updated last week