Alternatives and similar repositories for interpretability-starter

Users that are interested in interpretability-starter are comparing it to the libraries listed below

• TransformerLensOrg / CircuitsVis
  Mechanistic Interpretability Visualizations using React
  ☆299Updated 10 months ago
• thestephencasper / everything-you-need
  we got you bro
  ☆36Updated last year
• timaeus-research / devinterp
  Tools for studying developmental interpretability in neural networks.
  ☆114Updated 4 months ago
• anthropics / toy-models-of-superposition
  Notebooks accompanying Anthropic's "Toy Models of Superposition" paper
  ☆130Updated 3 years ago
• ArthurConmy / Automatic-Circuit-Discovery
  ☆252Updated last year
• callummcdougall / ARENA_2.0
  Resources for skilling up in AI alignment research engineering. Covers basics of deep learning, mechanistic interpretability, and RL.
  ☆231Updated 3 months ago
• callummcdougall / sae_vis
  Create feature-centric and prompt-centric visualizations for sparse autoencoders (like those from Anthropic's published research).
  ☆226Updated 11 months ago
• AlignmentResearch / tuned-lens
  Tools for understanding how transformer predictions are built layer-by-layer
  ☆541Updated 3 months ago
• neelnanda-io / 1L-Sparse-Autoencoder
  ☆131Updated 2 years ago
• EleutherAI / concept-erasure
  Erasing concepts from neural representations with provable guarantees
  ☆239Updated 9 months ago
• ARBORproject / arborproject.github.io
  ☆82Updated 8 months ago
• google-deepmind / mishax
  ☆143Updated 2 months ago
• collin-burns / discovering_latent_knowledge
  ☆281Updated last year
• EleutherAI / delphi
  Delphi was the home of a temple to Phoebus Apollo, which famously had the inscription, 'Know Thyself.' This library lets language models …
  ☆222Updated last week
• Sea-Snell / grokking
  unofficial re-implementation of "Grokking: Generalization Beyond Overfitting on Small Algorithmic Datasets"
  ☆79Updated 3 years ago
• EleutherAI / elk
  Keeping language models honest by directly eliciting knowledge encoded in their activations.
  ☆212Updated this week
• irregular-rhomboid / EAI-Math-Reading-Group
  Resources from the EleutherAI Math Reading Group
  ☆54Updated 8 months ago
• Prisma-Multimodal / ViT-Prisma
  ViT Prisma is a mechanistic interpretability library for Vision and Video Transformers (ViTs).
  ☆321Updated 3 months ago
• adamkarvonen / SAEBench
  ☆134Updated 3 weeks ago
• annahdo / implementing_activation_steering
  A collection of different ways to implement accessing and modifying internal model activations for LLMs
  ☆19Updated last year
• ckkissane / crosscoder-model-diff-replication
  Open source replication of Anthropic's Crosscoders for Model Diffing
  ☆59Updated last year
• nostalgebraist / transformer-utils
  Utilities for the HuggingFace transformers library
  ☆72Updated 2 years ago
• jbloomAus / SAEDashboard
  ☆79Updated last month
• saprmarks / dictionary_learning
  ☆361Updated 2 months ago
• ApolloResearch / e2e_sae
  Sparse Autoencoder Training Library
  ☆55Updated 6 months ago
• anthropics / PySvelte
  A library for bridging Python and HTML/Javascript (via Svelte) for creating interactive visualizations
  ☆198Updated 3 years ago
• bilal-chughtai / rep-theory-mech-interp
  ☆27Updated 2 years ago
• ApolloResearch / apd
  Attribution-based Parameter Decomposition
  ☆31Updated 5 months ago
• jacobdunefsky / transcoder_circuits
  ☆184Updated 11 months ago
• steering-vectors / steering-vectors
  Steering vectors for transformer language models in Pytorch / Huggingface
  ☆128Updated 8 months ago