A toolkit for describing model features and intervening on those features to steer behavior.
☆230Dec 12, 2025Updated 2 months ago
Alternatives and similar repositories for observatory
Users that are interested in observatory are comparing it to the libraries listed below
Sorting:
- The nnsight package enables interpreting and manipulating the internals of deep learned models.☆825Updated this week
- Delphi was the home of a temple to Phoebus Apollo, which famously had the inscription, 'Know Thyself.' This library lets language models …☆243Feb 23, 2026Updated last week
- Stanford NLP Python library for benchmarking the utility of LLM interpretability methods☆168Feb 22, 2026Updated last week
- Sparsify transformers with SAEs and transcoders☆696Feb 23, 2026Updated last week
- ☆28Nov 16, 2025Updated 3 months ago
- ☆36Apr 30, 2024Updated last year
- Open source interpretability artefacts for R1.☆171Apr 21, 2025Updated 10 months ago
- Multi-Layer Sparse Autoencoders (ICLR 2025)☆29Feb 6, 2026Updated 3 weeks ago
- ☆83Updated this week
- ☆83Feb 25, 2025Updated last year
- MishformerLens intends to be a drop-in replacement for TransformerLens that AST patches HuggingFace Transformers rather than implementing…☆10Oct 7, 2024Updated last year
- open source interpretability platform 🧠☆729Updated this week
- Create feature-centric and prompt-centric visualizations for sparse autoencoders (like those from Anthropic's published research).☆247Updated this week
- ☆45Feb 13, 2026Updated 2 weeks ago
- A library for mechanistic interpretability of GPT-style language models☆3,112Feb 23, 2026Updated last week
- ☆209Oct 14, 2025Updated 4 months ago
- Attribution-based Parameter Decomposition☆34Jun 11, 2025Updated 8 months ago
- Experiments in applying interpretability techniques to learned reward functions.☆10Dec 11, 2020Updated 5 years ago
- Repository for "Training Language Models To Explain Their Own Computations"☆21Dec 22, 2025Updated 2 months ago
- ☆36Jul 4, 2025Updated 7 months ago
- ☆23Jan 27, 2026Updated last month
- Code to reproduce key results accompanying "SAEs (usually) Transfer Between Base and Chat Models"☆13Jul 18, 2024Updated last year
- Keeping language models honest by directly eliciting knowledge encoded in their activations.☆217Feb 23, 2026Updated last week
- ☆19Sep 16, 2025Updated 5 months ago
- Interpreting Language Models with Contrastive Explanations (EMNLP 2022 Best Paper Honorable Mention)☆62May 12, 2022Updated 3 years ago
- Stanford NLP Python library for understanding and improving PyTorch models via interventions☆863Jan 29, 2026Updated last month
- Experimental LLM interface exploring new ways to use AI to improve human thinking☆19Updated this week
- [NeurIPS 2025 MechInterp Workshop - Spotlight] Official implementation of the paper "RelP: Faithful and Efficient Circuit Discovery in La…☆27Nov 3, 2025Updated 3 months ago
- Vivaria is METR's tool for running evaluations and conducting agent elicitation research.☆134Feb 15, 2026Updated 2 weeks ago
- NeuroSurgeon is a package that enables researchers to uncover and manipulate subnetworks within models in Huggingface Transformers☆43Feb 12, 2025Updated last year
- [NAACL'25 Oral] Steering Knowledge Selection Behaviours in LLMs via SAE-Based Representation Engineering☆73Jan 16, 2026Updated last month
- Code for steering and monitoring with concepts vectors in LLMs. https://arxiv.org/abs/2502.03708☆22Aug 10, 2025Updated 6 months ago
- ☆571Jul 19, 2024Updated last year
- ☆150Dec 30, 2025Updated 2 months ago
- HypotheSAEs: hypothesizing interpretable relationships in text datasets using sparse autoencoders. https://arxiv.org/abs/2502.04382☆77Updated this week
- ☆89Dec 18, 2025Updated 2 months ago
- Code and results accompanying the paper "Refusal in Language Models Is Mediated by a Single Direction".☆351Jun 13, 2025Updated 8 months ago
- Einsum with einops style variable names☆18May 16, 2024Updated last year
- ☆20Apr 10, 2025Updated 10 months ago