collin-burns / discovering_latent_knowledge
☆263Updated last year
Alternatives and similar repositories for discovering_latent_knowledge:
Users that are interested in discovering_latent_knowledge are comparing it to the libraries listed below
- ☆214Updated 6 months ago
- Keeping language models honest by directly eliciting knowledge encoded in their activations.☆197Updated last week
- Tools for understanding how transformer predictions are built layer-by-layer☆481Updated 10 months ago
- Mechanistic Interpretability Visualizations using React☆235Updated 3 months ago
- Create feature-centric and prompt-centric visualizations for sparse autoencoders (like those from Anthropic's published research).☆192Updated 3 months ago
- ☆114Updated 7 months ago
- Using sparse coding to find distributed representations used by neural networks.☆226Updated last year
- ☆121Updated last year
- ☆66Updated 4 months ago
- ☆197Updated last year
- Erasing concepts from neural representations with provable guarantees☆226Updated 2 months ago
- ☆90Updated last month
- Steering Llama 2 with Contrastive Activation Addition☆134Updated 10 months ago
- ☆80Updated this week
- Delphi was the home of a temple to Phoebus Apollo, which famously had the inscription, 'Know Thyself.' This library lets language models …☆165Updated last week
- A library for efficient patching and automatic circuit discovery.☆62Updated last month
- Sparsify transformers with SAEs and transcoders☆499Updated this week
- Steering vectors for transformer language models in Pytorch / Huggingface☆91Updated last month
- ☆264Updated last month
- ☆155Updated last week
- Code for my NeurIPS 2024 ATTRIB paper titled "Attribution Patching Outperforms Automated Circuit Discovery"☆30Updated 10 months ago
- ☆173Updated last year
- Extract full next-token probabilities via language model APIs☆237Updated last year
- Sparse Autoencoder for Mechanistic Interpretability☆234Updated 8 months ago
- Training Sparse Autoencoders on Language Models☆695Updated last week
- ☆61Updated 2 years ago
- ☆264Updated 9 months ago
- Emergent world representations: Exploring a sequence model trained on a synthetic task☆178Updated last year
- datasets from the paper "Towards Understanding Sycophancy in Language Models"☆73Updated last year
- Code to reproduce "Transformers Can Do Arithmetic with the Right Embeddings", McLeish et al (NeurIPS 2024)☆187Updated 10 months ago