ckkissane / sae-transferLinks
Code to reproduce key results accompanying "SAEs (usually) Transfer Between Base and Chat Models"
☆11Updated 10 months ago
Alternatives and similar repositories for sae-transfer
Users that are interested in sae-transfer are comparing it to the libraries listed below
Sorting:
- A small package implementing some useful wrapping around nnsight☆13Updated this week
- Code to enable layer-level steering in LLMs using sparse auto encoders☆18Updated last month
- Evaluate interpretability methods on localizing and disentangling concepts in LLMs.☆47Updated 8 months ago
- Investigating the generalization behavior of LM probes trained to predict truth labels: (1) from one annotator to another, and (2) from e…☆27Updated last year
- A library for efficient patching and automatic circuit discovery.☆65Updated last month
- Multi-Layer Sparse Autoencoders (ICLR 2025)☆22Updated 3 months ago
- ☆9Updated 2 weeks ago
- ☆31Updated 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…☆11Updated 4 months ago
- Official Code for our paper: "Language Models Learn to Mislead Humans via RLHF""☆14Updated 7 months ago
- Sparse Autoencoder Training Library☆52Updated last month
- Attribution-based Parameter Decomposition☆24Updated this week
- ☆22Updated last year
- Code for the ICLR 2024 paper "How to catch an AI liar: Lie detection in black-box LLMs by asking unrelated questions"☆70Updated 11 months ago
- Code for reproducing our paper "Not All Language Model Features Are Linear"☆75Updated 6 months ago
- PyTorch and NNsight implementation of AtP* (Kramar et al 2024, DeepMind)☆18Updated 4 months ago
- https://footprints.baulab.info☆17Updated 8 months ago
- ☆12Updated last year
- Algebraic value editing in pretrained language models☆65Updated last year
- Open source replication of Anthropic's Crosscoders for Model Diffing☆55Updated 7 months ago
- ☆14Updated last year
- ☆23Updated 4 months ago
- Universal Neurons in GPT2 Language Models☆29Updated last year
- ☆84Updated 10 months ago
- ☆43Updated 6 months ago
- ☆44Updated last year
- Code for my NeurIPS 2024 ATTRIB paper titled "Attribution Patching Outperforms Automated Circuit Discovery"☆34Updated last year
- Experiments with representation engineering☆11Updated last year
- [NeurIPS 2024 Spotlight] Code and data for the paper "Finding Transformer Circuits with Edge Pruning".☆56Updated 2 months ago
- Code and Data Repo for the CoNLL Paper -- Future Lens: Anticipating Subsequent Tokens from a Single Hidden State☆18Updated last year