tilde-research / sieveLinks
Applying SAEs for fine-grained control
☆25Updated last year
Alternatives and similar repositories for sieve
Users that are interested in sieve are comparing it to the libraries listed below
Sorting:
- ☆147Updated 3 months ago
- Open source replication of Anthropic's Crosscoders for Model Diffing☆63Updated last year
- Sparse Autoencoder Training Library☆56Updated 8 months ago
- Delphi was the home of a temple to Phoebus Apollo, which famously had the inscription, 'Know Thyself.' This library lets language models …☆234Updated last week
- ☆83Updated 2 weeks ago
- Create feature-centric and prompt-centric visualizations for sparse autoencoders (like those from Anthropic's published research).☆233Updated last year
- ☆36Updated last year
- Open source interpretability artefacts for R1.☆165Updated 8 months ago
- A tiny easily hackable implementation of a feature dashboard.☆15Updated 2 months ago
- ☆29Updated last year
- Attribution-based Parameter Decomposition☆33Updated 6 months ago
- Code for reproducing our paper "Not All Language Model Features Are Linear"☆83Updated last year
- ☆37Updated 10 months ago
- Code to reproduce "Transformers Can Do Arithmetic with the Right Embeddings", McLeish et al (NeurIPS 2024)☆198Updated last year
- ☆58Updated last year
- Notebooks accompanying Anthropic's "Toy Models of Superposition" paper☆131Updated 3 years ago
- Universal Neurons in GPT2 Language Models☆31Updated last year
- A library for efficient patching and automatic circuit discovery.☆84Updated 5 months ago
- ☆132Updated 2 years ago
- ☆112Updated 10 months ago
- Official repo for Learning to Reason for Long-Form Story Generation☆73Updated 8 months ago
- Engine for collecting, uploading, and downloading model activations☆24Updated 8 months ago
- The simplest, fastest repository for training/finetuning medium-sized GPTs.☆181Updated 6 months ago
- Steering vectors for transformer language models in Pytorch / Huggingface☆135Updated 10 months ago
- nanoGPT-like codebase for LLM training☆113Updated last month
- Investigating the generalization behavior of LM probes trained to predict truth labels: (1) from one annotator to another, and (2) from e…☆28Updated last year
- Modified to support crosscoder training.☆25Updated 2 months ago
- ☆25Updated 2 years ago
- Minimal (400 LOC) implementation Maximum (multi-node, FSDP) GPT training☆132Updated last year
- ☆23Updated 11 months ago