Alternatives and similar repositories for sleeper-agents

Users that are interested in sleeper-agents are comparing it to the libraries listed below

• slavachalnev / SAE-TS
  Improving Steering Vectors by Targeting Sparse Autoencoder Features
  ☆24Updated last year
• EleutherAI / elk-generalization
  Investigating the generalization behavior of LM probes trained to predict truth labels: (1) from one annotator to another, and (2) from e…
  ☆28Updated last year
• ckkissane / crosscoder-model-diff-replication
  Open source replication of Anthropic's Crosscoders for Model Diffing
  ☆62Updated last year
• ApolloResearch / e2e_sae
  Sparse Autoencoder Training Library
  ☆55Updated 6 months ago
• UFO-101 / auto-circuit
  A library for efficient patching and automatic circuit discovery.
  ☆80Updated 4 months ago
• steering-vectors / steering-vectors
  Steering vectors for transformer language models in Pytorch / Huggingface
  ☆130Updated 9 months ago
• ApolloResearch / apd
  Attribution-based Parameter Decomposition
  ☆32Updated 5 months ago
• ArthurConmy / Automatic-Circuit-Discovery
  ☆255Updated last year
• FlyingPumba / InterpBench
  A benchmark for mechanistic discovery of circuits in Transformers
  ☆16Updated 11 months ago
• EleutherAI / delphi
  Delphi was the home of a temple to Phoebus Apollo, which famously had the inscription, 'Know Thyself.' This library lets language models …
  ☆228Updated this week
• METR / RE-Bench
  ☆119Updated last month
• saprmarks / feature-circuits
  ☆196Updated last month
• explanare / ravel
  Evaluate interpretability methods on localizing and disentangling concepts in LLMs.
  ☆56Updated last month
• UKGovernmentBEIS / control-arena
  ControlArena is a collection of settings, model organisms and protocols - for running control experiments.
  ☆129Updated this week
• safety-research / safety-tooling
  Inference API for many LLMs and other useful tools for empirical research
  ☆80Updated last week
• science-of-finetuning / diffing-toolkit
  A toolkit that provides a range of model diffing techniques including a UI to visualize them interactively.
  ☆41Updated this week
• nrimsky / LM-exp
  LLM experiments done during SERI MATS - focusing on activation steering / interpreting activation spaces
  ☆99Updated 2 years ago
• redwoodresearch / interp
  Redwood Research's transformer interpretability tools
  ☆14Updated 3 years ago
• nrimsky / CAA
  Steering Llama 2 with Contrastive Activation Addition
  ☆194Updated last year
• noanabeshima / matryoshka-saes
  ☆25Updated last year
• METR / task-standard
  METR Task Standard
  ☆168Updated 9 months ago
• annahdo / implementing_activation_steering
  A collection of different ways to implement accessing and modifying internal model activations for LLMs
  ☆19Updated last year
• KoyenaPal / future-lens
  Code and Data Repo for the CoNLL Paper -- Future Lens: Anticipating Subsequent Tokens from a Single Hidden State
  ☆20Updated last month
• neelnanda-io / 1L-Sparse-Autoencoder
  ☆132Updated 2 years ago
• ApolloResearch / deception-detection
  ☆25Updated 9 months ago
• redwoodresearch / alignment_faking_public
  ☆81Updated last month
• neelnanda-io / Crosscoders
  ☆57Updated last year
• JoshEngels / SAE-Dark-Matter
  Code for our paper "Decomposing The Dark Matter of Sparse Autoencoders"
  ☆24Updated 9 months ago
• TransformerLensOrg / CircuitsVis
  Mechanistic Interpretability Visualizations using React
  ☆302Updated 11 months ago
• callummcdougall / sae_vis
  Create feature-centric and prompt-centric visualizations for sparse autoencoders (like those from Anthropic's published research).
  ☆228Updated 11 months ago