HypotheSAEs: hypothesizing interpretable relationships in text datasets using sparse autoencoders. https://arxiv.org/abs/2502.04382

Superpipe - optimized LLM pipelines for structured data

Github Pages template based upon HTML and Markdown for personal, portfolio-based websites.

"A Discrete Variational Recurrent Topic Model without the Reparametrization Trick" (NeurIPS 2020)

Facebook AI Research Sequence-to-Sequence Toolkit written in Python.

TensorFlow code and pre-trained models for BERT

CAP database scripts.

Deep universal probabilistic programming with Python and PyTorch

🍔

A little word cloud generator in Python

Minimalist Jekyll Template, dark and light themes