EleutherAI / elk
Keeping language models honest by directly eliciting knowledge encoded in their activations.
☆197Updated this week
Alternatives and similar repositories for elk:
Users that are interested in elk are comparing it to the libraries listed below
- ☆263Updated last year
- Mechanistic Interpretability Visualizations using React☆233Updated 3 months ago
- Erasing concepts from neural representations with provable guarantees☆226Updated last month
- ☆260Updated 8 months ago
- ☆211Updated 5 months ago
- Create feature-centric and prompt-centric visualizations for sparse autoencoders (like those from Anthropic's published research).☆189Updated 3 months ago
- Mechanistic Interpretability for Transformer Models☆50Updated 2 years ago
- ☆26Updated 11 months ago
- ☆121Updated last year
- Delphi was the home of a temple to Phoebus Apollo, which famously had the inscription, 'Know Thyself.' This library lets language models …☆161Updated this week
- METR Task Standard☆144Updated last month
- Steering Llama 2 with Contrastive Activation Addition☆129Updated 10 months ago
- Notebooks accompanying Anthropic's "Toy Models of Superposition" paper☆117Updated 2 years ago
- A dataset of alignment research and code to reproduce it☆74Updated last year
- ☆113Updated 7 months ago
- datasets from the paper "Towards Understanding Sycophancy in Language Models"☆73Updated last year
- Resources for skilling up in AI alignment research engineering. Covers basics of deep learning, mechanistic interpretability, and RL.☆206Updated last year
- ☆61Updated 4 months ago
- Steering vectors for transformer language models in Pytorch / Huggingface☆90Updated last month
- Emergent world representations: Exploring a sequence model trained on a synthetic task☆177Updated last year
- Utilities for the HuggingFace transformers library☆66Updated 2 years ago
- Tools for understanding how transformer predictions are built layer-by-layer☆480Updated 9 months ago
- Using sparse coding to find distributed representations used by neural networks.☆224Updated last year
- ☆124Updated this week
- Extract full next-token probabilities via language model APIs☆233Updated last year
- Sparsify transformers with SAEs and transcoders☆494Updated this week
- A library for efficient patching and automatic circuit discovery.☆59Updated last month
- TART: A plug-and-play Transformer module for task-agnostic reasoning☆195Updated last year
- Mass-editing thousands of facts into a transformer memory (ICLR 2023)☆470Updated last year
- Vivaria is METR's tool for running evaluations and conducting agent elicitation research.☆83Updated this week