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
  ☆302Updated 11 months ago
• ArthurConmy / Automatic-Circuit-Discovery
  ☆258Updated last year
• callummcdougall / sae_vis
  Create feature-centric and prompt-centric visualizations for sparse autoencoders (like those from Anthropic's published research).
  ☆231Updated 11 months ago
• timaeus-research / devinterp
  Tools for studying developmental interpretability in neural networks.
  ☆116Updated 5 months ago
• callummcdougall / ARENA_2.0
  Resources for skilling up in AI alignment research engineering. Covers basics of deep learning, mechanistic interpretability, and RL.
  ☆233Updated 3 months ago
• neelnanda-io / 1L-Sparse-Autoencoder
  ☆132Updated 2 years ago
• thestephencasper / everything-you-need
  we got you bro
  ☆36Updated last year
• AlignmentResearch / tuned-lens
  Tools for understanding how transformer predictions are built layer-by-layer
  ☆549Updated 3 months ago
• EleutherAI / elk
  Keeping language models honest by directly eliciting knowledge encoded in their activations.
  ☆214Updated this week
• EleutherAI / concept-erasure
  Erasing concepts from neural representations with provable guarantees
  ☆239Updated 10 months ago
• anthropics / toy-models-of-superposition
  Notebooks accompanying Anthropic's "Toy Models of Superposition" paper
  ☆130Updated 3 years ago
• collin-burns / discovering_latent_knowledge
  ☆283Updated last year
• ApolloResearch / apd
  Attribution-based Parameter Decomposition
  ☆32Updated 5 months ago
• ARBORproject / arborproject.github.io
  ☆83Updated 9 months ago
• google-deepmind / mishax
  ☆144Updated 2 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 …
  ☆228Updated last week
• nostalgebraist / transformer-utils
  Utilities for the HuggingFace transformers library
  ☆72Updated 2 years ago
• irregular-rhomboid / EAI-Math-Reading-Group
  Resources from the EleutherAI Math Reading Group
  ☆54Updated 9 months ago
• saprmarks / dictionary_learning
  ☆367Updated 3 months ago
• TomFrederik / unseal
  Mechanistic Interpretability for Transformer Models
  ☆53Updated 3 years ago
• ckkissane / crosscoder-model-diff-replication
  Open source replication of Anthropic's Crosscoders for Model Diffing
  ☆62Updated last year
• KihoPark / LLM_Categorical_Hierarchical_Representations
  ☆111Updated 9 months ago
• UKGovernmentBEIS / control-arena
  ControlArena is a collection of settings, model organisms and protocols - for running control experiments.
  ☆132Updated this week
• ai-safety-foundation / sparse_autoencoder
  Sparse Autoencoder for Mechanistic Interpretability
  ☆284Updated last year
• anthropics / PySvelte
  A library for bridging Python and HTML/Javascript (via Svelte) for creating interactive visualizations
  ☆198Updated 3 years ago
• steering-vectors / steering-vectors
  Steering vectors for transformer language models in Pytorch / Huggingface
  ☆130Updated 9 months ago
• ndif-team / nnsight
  The nnsight package enables interpreting and manipulating the internals of deep learned models.
  ☆707Updated last week
• redwoodresearch / Easy-Transformer
  ☆130Updated last year
• callummcdougall / sae_visualizer
  ☆29Updated last year
• yash-srivastava19 / arrakis
  Arrakis is a library to conduct, track and visualize mechanistic interpretability experiments.
  ☆31Updated 7 months ago