scienceetonnante / grokking

Related projects ⓘ

Alternatives and complementary repositories for grokking

• etalab-ia / franceservices-backend
  Backend ressources for Albert. Albert is a conversational agent that uses official French data sources to answer administrative agents qu…
  ☆116Updated this week
• Sohl-Dickstein / fractal
  The boundary of neural network trainability is fractal
  ☆161Updated 9 months ago
• lbourdois / pytorch-Deep-Learning-Notebooks-in-French
  Répertoire des traductions françaises des notebooks d'Alfredo Canziani du cours Deep Learning de la NYU
  ☆31Updated last year
• AlignmentResearch / tuned-lens
  Tools for understanding how transformer predictions are built layer-by-layer
  ☆433Updated 5 months ago
• TransformerLensOrg / CircuitsVis
  Mechanistic Interpretability Visualizations using React
  ☆200Updated 4 months ago
• saprmarks / dictionary_learning
  ☆146Updated last month
• HoagyC / sparse_coding
  Using sparse coding to find distributed representations used by neural networks.
  ☆189Updated last year
• CoderPat / croissant-llm-training
  Repository containing the code for training the CroissantLLM
  ☆21Updated 9 months ago
• scienceetonnante / lenia
  Exploration of the Lenia continuous cellular automaton
  ☆75Updated 10 months ago
• ndif-team / nnsight
  The nnsight package enables interpreting and manipulating the internals of deep learned models.
  ☆407Updated this week
• BlackHC / neural_net_checklist
  ☆139Updated 3 months ago
• jbloomAus / SAELens
  Training Sparse Autoencoders on Language Models
  ☆487Updated this week
• EleutherAI / sae-auto-interp
  ☆109Updated this week
• EleutherAI / sae
  Sparse autoencoders
  ☆347Updated this week
• johnmarktaylor91 / torchlens
  Package for extracting and mapping the results of every single tensor operation in a PyTorch model in one line of code.
  ☆485Updated last week
• tournesol-app / tournesol
  Free and open source code of the https://tournesol.app platform. Meet the community on Discord https://discord.gg/WvcSG55Bf3
  ☆330Updated this week
• ironjr / grokfast
  Official repository for the paper "Grokfast: Accelerated Grokking by Amplifying Slow Gradients"
  ☆516Updated 4 months ago
• serre-lab / Lens
  LENS Project
  ☆42Updated 9 months ago
• stanislavfort / dissect-git-re-basin
  Replicating and dissecting the git-re-basin project in one-click-replication Colabs
  ☆36Updated 2 years ago
• deel-ai / deel-torchlip
  Build and train Lipschitz-constrained networks: PyTorch implementation of 1-Lipschitz layers. For TensorFlow/Keras implementation, see ht…
  ☆27Updated last week
• ai-safety-foundation / sparse_autoencoder
  Sparse Autoencoder for Mechanistic Interpretability
  ☆191Updated 4 months ago
• timaeus-research / devinterp
  Tools for studying developmental interpretability in neural networks.
  ☆77Updated last week
• mlfoundations / task_vectors
  Editing Models with Task Arithmetic
  ☆431Updated 10 months ago
• Sea-Snell / grokking
  unofficial re-implementation of "Grokking: Generalization Beyond Overfitting on Small Algorithmic Datasets"
  ☆64Updated 2 years ago
• CentreSecuriteIA / BELLS
  Benchmarks for the Evaluation of LLM Supervision
  ☆28Updated this week
• callummcdougall / sae_vis
  Create feature-centric and prompt-centric visualizations for sparse autoencoders (like those from Anthropic's published research).
  ☆161Updated last month
• KellerJordan / Muon
  Muon optimizer for neural networks: >30% extra sample efficiency, <3% wallclock overhead
  ☆125Updated this week
• mcleish7 / arithmetic
  Code to reproduce "Transformers Can Do Arithmetic with the Right Embeddings", McLeish et al (NeurIPS 2024)
  ☆181Updated 5 months ago
• apartresearch / interpretability-starter
  🧠 Starter templates for doing interpretability research
  ☆63Updated last year
• epfml / llm-baselines
  nanoGPT-like codebase for LLM training
  ☆75Updated this week