Alternatives and similar repositories for Grokking

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

• anthropics / toy-models-of-superposition
  Notebooks accompanying Anthropic's "Toy Models of Superposition" paper
  ☆132Updated 3 years ago
• mcleish7 / arithmetic
  Code to reproduce "Transformers Can Do Arithmetic with the Right Embeddings", McLeish et al (NeurIPS 2024)
  ☆198Updated last year
• goodfire-ai / r1-interpretability
  Open source interpretability artefacts for R1.
  ☆165Updated 8 months ago
• google-deepmind / dangerous-capability-evaluations
  ☆64Updated 3 weeks ago
• google-deepmind / neural_networks_solomonoff_induction
  Learning Universal Predictors
  ☆81Updated last year
• neelnanda-io / Neuroscope
  Accompanying codebase for neuroscope.io, a website for displaying max activating dataset examples for language model neurons
  ☆13Updated 2 years ago
• google-deepmind / mishax
  ☆150Updated 4 months ago
• JoshEngels / MultiDimensionalFeatures
  Code for reproducing our paper "Not All Language Model Features Are Linear"
  ☆83Updated last year
• Silent-Zebra / twisted-smc-lm
  ☆31Updated 9 months ago
• neelnanda-io / 1L-Sparse-Autoencoder
  ☆132Updated 2 years ago
• ApolloResearch / e2e_sae
  Sparse Autoencoder Training Library
  ☆56Updated 8 months ago
• EleutherAI / elk-generalization
  Investigating the generalization behavior of LM probes trained to predict truth labels: (1) from one annotator to another, and (2) from e…
  ☆28Updated last year
• anthropics / sleeper-agents-paper
  Contains random samples referenced in the paper "Sleeper Agents: Training Robustly Deceptive LLMs that Persist Through Safety Training".
  ☆123Updated last year
• KihoPark / LLM_Categorical_Hierarchical_Representations
  ☆112Updated 10 months ago
• KindXiaoming / BIMT
  Brain-Inspired Modular Training (BIMT), a method for making neural networks more modular and interpretable.
  ☆174Updated 2 years ago
• likenneth / othello_world
  Emergent world representations: Exploring a sequence model trained on a synthetic task
  ☆197Updated 2 years ago
• Sea-Snell / grokking
  unofficial re-implementation of "Grokking: Generalization Beyond Overfitting on Small Algorithmic Datasets"
  ☆81Updated 3 years ago
• athms / mad-lab
  A MAD laboratory to improve AI architecture designs 🧪
  ☆135Updated last year
• timaeus-research / devinterp
  Tools for studying developmental interpretability in neural networks.
  ☆119Updated last week
• METR / RE-Bench
  ☆127Updated 2 months ago
• thestephencasper / everything-you-need
  we got you bro
  ☆36Updated last year
• aadityasingh / icl-dynamics
  ☆24Updated 8 months ago
• JasonGross / guarantees-based-mechanistic-interpretability
  ☆17Updated 3 weeks ago
• oripress / AlgoTune
  AlgoTune is a NeurIPS 2025 benchmark made up of 154 math, physics, and computer science problems. The goal is write code that solves each…
  ☆80Updated this week
• AllanYangZhou / universal_neural_functional
  ☆54Updated last year
• callummcdougall / sae_vis
  Create feature-centric and prompt-centric visualizations for sparse autoencoders (like those from Anthropic's published research).
  ☆236Updated last year
• harmonic-ai / datasets
  Harmonic Datasets
  ☆52Updated last year
• ckkissane / crosscoder-model-diff-replication
  Open source replication of Anthropic's Crosscoders for Model Diffing
  ☆63Updated last year
• KhoomeiK / complexity-scaling
  gzip Predicts Data-dependent Scaling Laws
  ☆34Updated last year
• tilde-research / sieve
  Applying SAEs for fine-grained control
  ☆25Updated last year