Pinned Loading

1. ndif-team/nnterp ndif-team/nnterp Public

   Unified access to Large Language Model modules using NNsight

   Python 121 12

2. science-of-finetuning/diffing-toolkit science-of-finetuning/diffing-toolkit Public

   A toolkit that provides a range of model diffing techniques including a UI to visualize them interactively.

   Python 88 24

3. tiny-activation-dashboard tiny-activation-dashboard Public

   A tiny easily hackable implementation of a feature dashboard.

   Jupyter Notebook 18 1

4. science-of-finetuning/crosscoder_learning science-of-finetuning/crosscoder_learning Public

   Forked from saprmarks/dictionary_learning

   Modified to support crosscoder training.

   Python 29 10

5. science-of-finetuning/sparsity-artifacts-crosscoders science-of-finetuning/sparsity-artifacts-crosscoders Public

   Code for the "Overcoming Sparsity Artifacts in Crosscoders to Interpret Chat-Tuning" paper.

   Jupyter Notebook 19 9

6. llm-lang-agnostic llm-lang-agnostic Public

   minimal code to reproduce results from Separating Tongue from Thought: Activation Patching Reveals Language-Agnostic Concept Representations in Transformers

   Jupyter Notebook 14 5