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
  ☆26Updated 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
  ☆63Updated last year
• XuchanBao / behavioral-self-awareness
  ☆34Updated 10 months ago
• neelnanda-io / Crosscoders
  ☆58Updated last year
• METR / RE-Bench
  ☆128Updated 3 months ago
• amack315 / unsupervised-steering-vectors
  ☆36Updated last year
• callummcdougall / sae_visualizer
  ☆29Updated last year
• redwoodresearch / alignment_faking_public
  ☆82Updated 3 months ago
• UKGovernmentBEIS / control-arena
  ControlArena is a collection of settings, model organisms and protocols - for running control experiments.
  ☆148Updated this week
• ApolloResearch / e2e_sae
  Sparse Autoencoder Training Library
  ☆56Updated 8 months ago
• ndif-team / nnterp
  Unified access to Large Language Model modules using NNsight
  ☆74Updated last week
• nrimsky / LM-exp
  LLM experiments done during SERI MATS - focusing on activation steering / interpreting activation spaces
  ☆100Updated 2 years ago
• UFO-101 / auto-circuit
  A library for efficient patching and automatic circuit discovery.
  ☆85Updated 2 weeks ago
• LoryPack / LLM-LieDetector
  Code for the ICLR 2024 paper "How to catch an AI liar: Lie detection in black-box LLMs by asking unrelated questions"
  ☆71Updated last year
• safety-research / safety-tooling
  Inference API for many LLMs and other useful tools for empirical research
  ☆98Updated 3 weeks ago
• METR / task-standard
  METR Task Standard
  ☆170Updated 11 months ago
• steering-vectors / steering-vectors
  Steering vectors for transformer language models in Pytorch / Huggingface
  ☆137Updated 10 months ago
• google-deepmind / dangerous-capability-evaluations
  ☆64Updated this week
• annahdo / implementing_activation_steering
  A collection of different ways to implement accessing and modifying internal model activations for LLMs
  ☆21Updated last year
• ArthurConmy / Automatic-Circuit-Discovery
  ☆264Updated last year
• science-of-finetuning / diffing-toolkit
  A toolkit that provides a range of model diffing techniques including a UI to visualize them interactively.
  ☆50Updated this week
• EleutherAI / delphi
  Delphi was the home of a temple to Phoebus Apollo, which famously had the inscription, 'Know Thyself.' This library lets language models …
  ☆236Updated last week
• science-of-finetuning / crosscoder_learning
  Modified to support crosscoder training.
  ☆25Updated this week
• anthropics / sleeper-agents-paper
  Contains random samples referenced in the paper "Sleeper Agents: Training Robustly Deceptive LLMs that Persist Through Safety Training".
  ☆124Updated last year
• nrimsky / CAA
  Steering Llama 2 with Contrastive Activation Addition
  ☆206Updated last year
• explanare / ravel
  Evaluate interpretability methods on localizing and disentangling concepts in LLMs.
  ☆57Updated 2 months ago
• ApolloResearch / apd
  Attribution-based Parameter Decomposition
  ☆33Updated 7 months ago
• adamkarvonen / SAEBench
  ☆142Updated 2 weeks ago
• FlyingPumba / InterpBench
  A benchmark for mechanistic discovery of circuits in Transformers
  ☆16Updated last year