Alternatives and similar repositories for evals

Users that are interested in evals are comparing it to the libraries listed below

• EleutherAI / elk
  Keeping language models honest by directly eliciting knowledge encoded in their activations.
  ☆209Updated this week
• METR / task-standard
  METR Task Standard
  ☆159Updated 7 months ago
• meg-tong / sycophancy-eval
  datasets from the paper "Towards Understanding Sycophancy in Language Models"
  ☆90Updated last year
• aypan17 / machiavelli
  ☆138Updated last month
• TransformerLensOrg / CircuitsVis
  Mechanistic Interpretability Visualizations using React
  ☆285Updated 8 months ago
• anthropics / ConstitutionalHarmlessnessPaper
  ☆240Updated 2 years ago
• collin-burns / discovering_latent_knowledge
  ☆276Updated last year
• justinchiu / openlogprobs
  Extract full next-token probabilities via language model APIs
  ☆247Updated last year
• AlignmentResearch / tuned-lens
  Tools for understanding how transformer predictions are built layer-by-layer
  ☆524Updated last month
• google-research / cascades
  Python library which enables complex compositions of language models such as scratchpads, chain of thought, tool use, selection-inference…
  ☆210Updated 3 months ago
• EleutherAI / concept-erasure
  Erasing concepts from neural representations with provable guarantees
  ☆233Updated 7 months ago
• anthropics / sleeper-agents-paper
  Contains random samples referenced in the paper "Sleeper Agents: Training Robustly Deceptive LLMs that Persist Through Safety Training".
  ☆112Updated last year
• callummcdougall / sae_vis
  Create feature-centric and prompt-centric visualizations for sparse autoencoders (like those from Anthropic's published research).
  ☆219Updated 8 months ago
• neelnanda-io / 1L-Sparse-Autoencoder
  ☆127Updated last year
• anthropics / toy-models-of-superposition
  Notebooks accompanying Anthropic's "Toy Models of Superposition" paper
  ☆129Updated 3 years ago
• METR / RE-Bench
  ☆98Updated 4 months ago
• nrimsky / LM-exp
  LLM experiments done during SERI MATS - focusing on activation steering / interpreting activation spaces
  ☆95Updated last year
• ArthurConmy / Automatic-Circuit-Discovery
  ☆240Updated 11 months ago
• METR / public-tasks
  ☆101Updated 5 months ago
• UKGovernmentBEIS / inspect_evals
  Collection of evals for Inspect AI
  ☆223Updated last week
• normster / llm_rules
  RuLES: a benchmark for evaluating rule-following in language models
  ☆231Updated 6 months ago
• jessicarumbelow / Backwards
  ☆83Updated last year
• TransluceAI / observatory
  A toolkit for describing model features and intervening on those features to steer behavior.
  ☆198Updated 10 months ago
• likenneth / othello_world
  Emergent world representations: Exploring a sequence model trained on a synthetic task
  ☆190Updated 2 years ago
• google-deepmind / dangerous-capability-evaluations
  ☆57Updated last month
• lukasberglund / reversal_curse
  ☆295Updated last year
• UKGovernmentBEIS / control-arena
  ControlArena is a collection of settings, model organisms and protocols - for running control experiments.
  ☆90Updated this week
• moirage / alignment-research-dataset
  A dataset of alignment research and code to reproduce it
  ☆77Updated 2 years ago
• hendrycks / ethics
  Aligning AI With Shared Human Values (ICLR 2021)
  ☆297Updated 2 years ago
• ofirpress / self-ask
  Code and data for "Measuring and Narrowing the Compositionality Gap in Language Models"
  ☆320Updated last year