Alternatives and similar repositories for dangerous-capability-evaluations

Users that are interested in dangerous-capability-evaluations are comparing it to the libraries listed below

• anthropics / sleeper-agents-paper
  Contains random samples referenced in the paper "Sleeper Agents: Training Robustly Deceptive LLMs that Persist Through Safety Training".
  ☆111Updated last year
• METR / task-standard
  METR Task Standard
  ☆156Updated 6 months ago
• METR / RE-Bench
  ☆95Updated 3 months ago
• 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
• google-deepmind / mishax
  ☆136Updated 4 months ago
• UKGovernmentBEIS / inspect_evals
  Collection of evals for Inspect AI
  ☆198Updated this week
• EleutherAI / elk
  Keeping language models honest by directly eliciting knowledge encoded in their activations.
  ☆209Updated last week
• aypan17 / machiavelli
  ☆137Updated 2 weeks ago
• redwoodresearch / interp
  Redwood Research's transformer interpretability tools
  ☆14Updated 3 years ago
• anthropics / toy-models-of-superposition
  Notebooks accompanying Anthropic's "Toy Models of Superposition" paper
  ☆127Updated 2 years ago
• AsaCooperStickland / situational-awareness-evals
  Measuring the situational awareness of language models
  ☆37Updated last year
• goodfire-ai / r1-interpretability
  Open source interpretability artefacts for R1.
  ☆157Updated 3 months 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"
  ☆72Updated last year
• meg-tong / sycophancy-eval
  datasets from the paper "Towards Understanding Sycophancy in Language Models"
  ☆86Updated last year
• METR / public-tasks
  ☆99Updated 4 months ago
• anthropics / evals
  ☆287Updated last year
• UKGovernmentBEIS / control-arena
  ControlArena is a collection of settings, model organisms and protocols - for running control experiments.
  ☆80Updated this week
• emergent-misalignment / emergent-misalignment
  ☆182Updated 4 months ago
• callummcdougall / sae_vis
  Create feature-centric and prompt-centric visualizations for sparse autoencoders (like those from Anthropic's published research).
  ☆207Updated 7 months ago
• ApolloResearch / e2e_sae
  Sparse Autoencoder Training Library
  ☆54Updated 3 months ago
• thestephencasper / everything-you-need
  we got you bro
  ☆36Updated last year
• METR / vivaria
  Vivaria is METR's tool for running evaluations and conducting agent elicitation research.
  ☆103Updated last week
• haizelabs / bijection-learning
  ☆24Updated 9 months ago
• callummcdougall / sae_visualizer
  ☆28Updated last year
• ConsequentAI / fneval
  Functional Benchmarks and the Reasoning Gap
  ☆88Updated 10 months ago
• safety-research / safety-tooling
  Inference API for many LLMs and other useful tools for empirical research
  ☆61Updated this week
• TransluceAI / observatory
  A toolkit for describing model features and intervening on those features to steer behavior.
  ☆195Updated 8 months ago
• ApolloResearch / insider-trading
  ☆52Updated last year
• JoshEngels / MultiDimensionalFeatures
  Code for reproducing our paper "Not All Language Model Features Are Linear"
  ☆77Updated 8 months ago
• justinchiu / openlogprobs
  Extract full next-token probabilities via language model APIs
  ☆247Updated last year