adamjermyn / toy_model_interpretability

Related projects ⓘ

Alternatives and complementary repositories for toy_model_interpretability

• TomFrederik / unseal
  Mechanistic Interpretability for Transformer Models
  ☆49Updated 2 years ago
• understanding-search / maze-transformer
  This repo is built to facilitate the training and analysis of autoregressive transformers on maze-solving tasks.
  ☆24Updated 2 months ago
• redwoodresearch / interp
  Redwood Research's transformer interpretability tools
  ☆12Updated 2 years ago
• ApolloResearch / e2e_sae
  Sparse Autoencoder Training Library
  ☆26Updated last week
• jbloomAus / DecisionTransformerInterpretability
  Interpreting how transformers simulate agents performing RL tasks
  ☆69Updated last year
• mukobi / welfare-diplomacy
  General-Sum variant of the game Diplomacy for evaluating AIs.
  ☆23Updated 7 months ago
• bilal-chughtai / rep-theory-mech-interp
  ☆24Updated last year
• andyljones / boardlaw
  Scaling scaling laws with board games.
  ☆40Updated last year
• EleutherAI / w2s
  ☆18Updated last month
• koayon / atp_star
  PyTorch and NNsight implementation of AtP* (Kramar et al 2024, DeepMind)
  ☆13Updated 6 months ago
• neale / avoiding-side-effects
  Code for reproducing the results from the paper Avoiding Side Effects in Complex Environments
  ☆12Updated 3 years ago
• alexander-turner / attainable-utility-preservation
  ☆12Updated 3 years ago
• google-deepmind / dangerous-capability-evaluations
  ☆44Updated last month
• kanishkg / synchromesh
  Get language models to generate responses in a specific format reliably. Open source implementation of Synchromesh: Reliable code generat…
  ☆24Updated 8 months ago
• HumanCompatibleAI / leela-interp
  Code for "Evidence of Learned Look-Ahead in a Chess-Playing Neural Network"
  ☆14Updated 5 months ago
• google-deepmind / cartesian-frames
  A formalisation of Cartesian Frames, a perspective on embedded agency, in the HOL theorem prover.
  ☆19Updated 2 years ago
• thestephencasper / everything-you-need
  we got you bro
  ☆32Updated 3 months ago
• wattenberg / superposition
  Code associated to papers on superposition (in ML interpretability)
  ☆24Updated 2 years ago
• Sea-Snell / grokking
  unofficial re-implementation of "Grokking: Generalization Beyond Overfitting on Small Algorithmic Datasets"
  ☆61Updated 2 years ago
• Mech-Interp / PySvelte
  A library for bridging Python and HTML/Javascript (via Svelte) for creating interactive visualizations
  ☆13Updated 6 months ago
• timaeus-research / devinterp
  Tools for studying developmental interpretability in neural networks.
  ☆74Updated this week
• saprmarks / feature-circuits
  ☆102Updated last month
• anthropics / toy-models-of-superposition
  Notebooks accompanying Anthropic's "Toy Models of Superposition" paper
  ☆95Updated 2 years ago
• AsaCooperStickland / situational-awareness-evals
  Measuring the situational awareness of language models
  ☆33Updated 8 months ago
• cognitiveailab / neurosymbolic
  A neurosymbolic T5 agent for playing text games, from the EACL 2023 paper "Behavior Cloned Transformers are Neurosymbolic Reasoners"
  ☆19Updated last year
• taufeeque9 / codebook-features
  Sparse and discrete interpretability tool for neural networks
  ☆53Updated 8 months ago
• JoshEngels / SAE-Dark-Matter
  Code for our paper "Decomposing The Dark Matter of Sparse Autoencoders"
  ☆13Updated 2 weeks ago
• danielmamay / grokking
  Implementation of OpenAI's 'Grokking: Generalization Beyond Overfitting on Small Algorithmic Datasets' paper.
  ☆34Updated last year
• UFO-101 / auto-circuit
  A library for efficient patching and automatic circuit discovery.
  ☆30Updated last month
• samacqua / LARC
  Language-annotated Abstraction and Reasoning Corpus
  ☆78Updated last year