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
  ☆272Updated 7 months ago
• thestephencasper / everything-you-need
  we got you bro
  ☆35Updated last year
• anthropics / toy-models-of-superposition
  Notebooks accompanying Anthropic's "Toy Models of Superposition" paper
  ☆127Updated 2 years 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
• timaeus-research / devinterp
  Tools for studying developmental interpretability in neural networks.
  ☆100Updated last month
• ArthurConmy / Automatic-Circuit-Discovery
  ☆233Updated 10 months ago
• neelnanda-io / 1L-Sparse-Autoencoder
  ☆123Updated last year
• callummcdougall / ARENA_2.0
  Resources for skilling up in AI alignment research engineering. Covers basics of deep learning, mechanistic interpretability, and RL.
  ☆219Updated last year
• google-deepmind / mishax
  ☆134Updated 4 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 …
  ☆200Updated this week
• EleutherAI / concept-erasure
  Erasing concepts from neural representations with provable guarantees
  ☆231Updated 6 months ago
• ARBORproject / arborproject.github.io
  ☆81Updated 5 months ago
• irregular-rhomboid / EAI-Math-Reading-Group
  Resources from the EleutherAI Math Reading Group
  ☆53Updated 5 months ago
• AlignmentResearch / tuned-lens
  Tools for understanding how transformer predictions are built layer-by-layer
  ☆512Updated last year
• collin-burns / discovering_latent_knowledge
  ☆274Updated last year
• EleutherAI / elk
  Keeping language models honest by directly eliciting knowledge encoded in their activations.
  ☆209Updated last week
• adamkarvonen / SAEBench
  ☆107Updated 2 weeks ago
• ndif-team / nnsight
  The nnsight package enables interpreting and manipulating the internals of deep learned models.
  ☆619Updated this week
• ApolloResearch / apd
  Attribution-based Parameter Decomposition
  ☆28Updated last month
• ai-safety-foundation / sparse_autoencoder
  Sparse Autoencoder for Mechanistic Interpretability
  ☆257Updated last year
• ckkissane / crosscoder-model-diff-replication
  Open source replication of Anthropic's Crosscoders for Model Diffing
  ☆57Updated 9 months ago
• callummcdougall / sae_visualizer
  ☆28Updated last year
• saprmarks / dictionary_learning
  ☆324Updated 3 weeks ago
• google-deepmind / dangerous-capability-evaluations
  ☆55Updated 10 months ago
• nostalgebraist / transformer-utils
  Utilities for the HuggingFace transformers library
  ☆70Updated 2 years ago
• ApolloResearch / e2e_sae
  Sparse Autoencoder Training Library
  ☆54Updated 3 months ago
• TomFrederik / unseal
  Mechanistic Interpretability for Transformer Models
  ☆51Updated 3 years ago
• Prisma-Multimodal / ViT-Prisma
  ViT Prisma is a mechanistic interpretability library for Vision and Video Transformers (ViTs).
  ☆289Updated last week
• jbloomAus / SAEDashboard
  ☆62Updated this week
• likenneth / othello_world
  Emergent world representations: Exploring a sequence model trained on a synthetic task
  ☆186Updated 2 years ago