Investigating the generalization behavior of LM probes trained to predict truth labels: (1) from one annotator to another, and (2) from easy questions to hard
☆28May 23, 2024Updated last year
Alternatives and similar repositories for elk-generalization
Users that are interested in elk-generalization are comparing it to the libraries listed below
Sorting:
- A python sdk for LLM finetuning and inference on runpod infrastructure☆19Feb 16, 2026Updated last week
- A library for mechanistic anomaly detection☆22Jan 9, 2025Updated last year
- Code for my NeurIPS 2024 ATTRIB paper titled "Attribution Patching Outperforms Automated Circuit Discovery"☆47May 31, 2024Updated last year
- A collection of different ways to implement accessing and modifying internal model activations for LLMs☆20Oct 18, 2024Updated last year
- Stampy's copy of Alignment Research Dataset scraper☆13Dec 26, 2025Updated 2 months ago
- ☆33Jun 4, 2025Updated 8 months ago
- 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…☆13Jan 26, 2025Updated last year
- ☆21Jun 22, 2025Updated 8 months ago
- Decoder only transformer, built from scratch with PyTorch☆33Oct 22, 2023Updated 2 years ago
- Erasing concepts from neural representations with provable guarantees☆243Jan 27, 2025Updated last year
- Documenting large text datasets 🖼️ 📚☆14Dec 17, 2024Updated last year
- A benchmark for mechanistic discovery of circuits in Transformers☆16Dec 15, 2024Updated last year
- Repository with sample code using Apollo's suggested engineering practices☆15Dec 16, 2024Updated last year
- Codebase for Linguistic Collapse: Neural Collapse in (Large) Language Models [NeurIPS 2024] [arXiv:2405.17767]☆18Apr 14, 2025Updated 10 months ago
- Keeping language models honest by directly eliciting knowledge encoded in their activations.☆217Updated this week
- Understanding how features learned by neural networks evolve throughout training☆41Oct 24, 2024Updated last year
- Minimum Description Length probing for neural network representations☆20Jan 28, 2025Updated last year
- Pile Deduplication Code☆18May 15, 2023Updated 2 years ago
- URL downloader supporting checkpointing and continuous checksumming.☆19Nov 29, 2023Updated 2 years ago
- ☆150Dec 30, 2025Updated 2 months ago
- Evaluate interpretability methods on localizing and disentangling concepts in LLMs.☆57Oct 30, 2025Updated 4 months ago
- ☆396Aug 21, 2025Updated 6 months ago
- Code for Preventing Language Models From Hiding Their Reasoning, which evaluates defenses against LLM steganography.☆25Jan 26, 2024Updated 2 years ago
- Sparse Autoencoder for Mechanistic Interpretability☆292Jul 20, 2024Updated last year
- Delphi was the home of a temple to Phoebus Apollo, which famously had the inscription, 'Know Thyself.' This library lets language models …☆243Updated this week
- A blog on AI, personal development, and living a good life.☆35Updated this week
- ☆30Aug 2, 2024Updated last year
- Simple and scalable tools for data-driven pretraining data selection.☆29Jun 9, 2025Updated 8 months ago
- Designing a Dashboard for Transparency and Control of Conversational AI, https://arxiv.org/abs/2406.07882☆33Oct 7, 2025Updated 4 months ago
- [NeurIPS 2024 Spotlight] Code and data for the paper "Finding Transformer Circuits with Edge Pruning".☆66Aug 15, 2025Updated 6 months ago
- ☆209Oct 14, 2025Updated 4 months ago
- ☆28Feb 27, 2025Updated last year
- Code for the ICLR 2024 paper "How to catch an AI liar: Lie detection in black-box LLMs by asking unrelated questions"☆71Jun 19, 2024Updated last year
- A MATLAB app to interactively navigate Ryze Tello drone, read navigation data, process image data and produce equivalent MATLAB code. Thi…☆13Oct 22, 2025Updated 4 months ago
- Accompanying code for "Boosted Prompt Ensembles for Large Language Models"☆30Apr 13, 2023Updated 2 years ago
- Plan✕ is a platform for creating and publishing digital planning services☆17Updated this week
- ☆265Jan 12, 2026Updated last month
- ☆31Mar 27, 2025Updated 11 months ago
- Create feature-centric and prompt-centric visualizations for sparse autoencoders (like those from Anthropic's published research).☆247Updated this week