Pinned Loading

1. TransformerLensOrg/TransformerLens TransformerLensOrg/TransformerLens Public

   A library for mechanistic interpretability of GPT-style language models

   Python 1.7k 314

2. self-organized-control self-organized-control Public

   To explore how complex and stable motor control can emerge from neurons, we designed neural cellular automata to robustly control a cart-pole agent.

   Python 19 3

3. swap-graphs swap-graphs Public

   An implementation of input swap graphs. A tool to discover the role of neural network components with causal interventions.

   Python 7 2

4. causal-checker causal-checker Public

   Python 6 2

5. non-robust-feature-transfer non-robust-feature-transfer Public

   I tried to transfer the non-robust features of a secret image into a geniuine image to create an adversarial example recognized as the secret image by CNN but looks like the genuine.

   Jupyter Notebook 1

6. esn-vs-lstm-on-cross-situationnal-learning esn-vs-lstm-on-cross-situationnal-learning Public

   Python 4 3