Alternatives and similar repositories for control-arena

Users that are interested in control-arena are comparing it to the libraries listed below

• METR / task-standard
  METR Task Standard
  ☆167Updated 9 months ago
• ArthurConmy / Automatic-Circuit-Discovery
  ☆253Updated last year
• safety-research / safety-tooling
  Inference API for many LLMs and other useful tools for empirical research
  ☆80Updated last week
• TransformerLensOrg / CircuitsVis
  Mechanistic Interpretability Visualizations using React
  ☆301Updated 11 months ago
• UKGovernmentBEIS / inspect_evals
  Collection of evals for Inspect AI
  ☆289Updated this week
• TransluceAI / observatory
  A toolkit for describing model features and intervening on those features to steer behavior.
  ☆214Updated last year
• redwoodresearch / alignment_faking_public
  ☆81Updated last month
• callummcdougall / sae_vis
  Create feature-centric and prompt-centric visualizations for sparse autoencoders (like those from Anthropic's published research).
  ☆227Updated 11 months ago
• timaeus-research / devinterp
  Tools for studying developmental interpretability in neural networks.
  ☆114Updated 4 months ago
• anthropics / sleeper-agents-paper
  Contains random samples referenced in the paper "Sleeper Agents: Training Robustly Deceptive LLMs that Persist Through Safety Training".
  ☆122Updated last year
• google-deepmind / dangerous-capability-evaluations
  ☆62Updated last month
• ai-safety-foundation / sparse_autoencoder
  Sparse Autoencoder for Mechanistic Interpretability
  ☆284Updated last year
• callummcdougall / ARENA_2.0
  Resources for skilling up in AI alignment research engineering. Covers basics of deep learning, mechanistic interpretability, and RL.
  ☆232Updated 3 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 …
  ☆225Updated last week
• TransluceAI / docent
  ☆63Updated last month
• Butanium / nnterp
  Unified access to Large Language Model modules using NNsight
  ☆59Updated last week
• emergent-misalignment / emergent-misalignment
  ☆226Updated 3 weeks ago
• steering-vectors / steering-vectors
  Steering vectors for transformer language models in Pytorch / Huggingface
  ☆129Updated 9 months ago
• METR / vivaria
  Vivaria is METR's tool for running evaluations and conducting agent elicitation research.
  ☆120Updated last week
• METR / public-tasks
  ☆106Updated this week
• safety-research / safety-examples
  ☆19Updated last week
• ndif-team / nnsight
  The nnsight package enables interpreting and manipulating the internals of deep learned models.
  ☆701Updated this week
• GraySwanAI / circuit-breakers
  Improving Alignment and Robustness with Circuit Breakers
  ☆242Updated last year
• EleutherAI / elk
  Keeping language models honest by directly eliciting knowledge encoded in their activations.
  ☆212Updated last week
• aypan17 / machiavelli
  ☆141Updated 3 months ago
• jacobdunefsky / transcoder_circuits
  ☆188Updated last year
• saprmarks / feature-circuits
  ☆195Updated last month
• apartresearch / interpretability-starter
  🧠 Starter templates for doing interpretability research
  ☆75Updated 2 years ago
• anthropics / evals
  ☆310Updated last year
• UFO-101 / auto-circuit
  A library for efficient patching and automatic circuit discovery.
  ☆80Updated 4 months ago