wesg52

AI-Poker

annotated-transformer

http://nlp.seas.harvard.edu/2018/04/03/attention.html

llm-context-neurons

Find context neurons in Pythia models.

mats_sae_training

Training Sparse Autoencoders on Language Models

pysindy-miosr

A package for the sparse identification of nonlinear dynamical systems from data

sindy_mio_paper

Code for Learning Sparse Nonlinear Dynamics via Mixed Integer Optimization

sparse-probing-paper

Sparse probing paper full code.

universal-neurons

Universal Neurons in GPT2 Language Models

UrbanAnalytics

Web application built on census data for academics and policy makers

world-models

Extracting spatial and temporal world models from LLMs

wesg52/world-models

Extracting spatial and temporal world models from LLMs

wesg52/sparse-probing-paper

Sparse probing paper full code.

wesg52/universal-neurons

Universal Neurons in GPT2 Language Models

wesg52/sindy_mio_paper

Code for Learning Sparse Nonlinear Dynamics via Mixed Integer Optimization

wesg52/llm-context-neurons

Find context neurons in Pythia models.

wesg52/AI-Poker

wesg52/mats_sae_training

Training Sparse Autoencoders on Language Models

wesg52/pysindy-miosr

A package for the sparse identification of nonlinear dynamical systems from data

wesg52/UrbanAnalytics

Web application built on census data for academics and policy makers

wesg52/annotated-transformer

http://nlp.seas.harvard.edu/2018/04/03/attention.html

wesg52/clusterrepo

Fork this repository for 15.S60 assignment 1 task 2.

wesg52/covid19-modeling

Coronavirus agent based modeling

wesg52/Genetic-Knapsack

Implementing genetic algorithms to solve 0-1 knapsack

wesg52/Kmedians

Benchmarking Kmedians for Redistricting

wesg52/Mikey-Midi

A python tutorial motivated by music

wesg52/MNIST

Using different neural network architectures to solve MNIST classification

wesg52/NOIE

Neural Open Information Extraction CS 6740 Final Project

wesg52/Project-Euler

Solutions to Project Euler problems

wesg52/sae_vis

Create feature-centric and prompt-centric visualizations for sparse autoencoders (like those from Anthropic's published research).

wesg52/TransformerLens

A library for mechanistic interpretability of GPT-style language models

wesg52/wesg52.github.io

Personal website