Popular repositories Loading

1. llm-adversarial-attacks llm-adversarial-attacks Public

   Adversarial attacks on LLMs, for influencing outputs of hidden layer linear probes and steering generations.

   Python 1

2. markov-transformers markov-transformers Public

   Unsupervised decoding of hidden markov models from transformer residual streams (for transformers also trained on HMM data and acting as optimal predictors of the HMM)

   Python 1

3. neural-optimal-control neural-optimal-control Public

   End-To-End differentiable optimal control with neural controllers

   Jupyter Notebook

4. unsupervised-representation-translation unsupervised-representation-translation Public

   Multimodal Unsupervised Machine Translation and Unsupervised Representation Learning for high-level model interpretability.

   Python

5. simple-stories simple-stories Public

   A dataset of 10k short and simple bilingual (English/German) stories and associated scripts for sampling new stories.

   Python