Pinned Repositories
apush-textbook
A website I built to host pdf scans of the textbook for my AP US History class.
ARENA_3.0
autointerp
baulab
bhc
case_study
m3
Code and models I wrote for the Mathworks Math Modeling Challenge.
nnterface
sae-auto-interp
nnsight
The nnsight package enables interpreting and manipulating the internals of deep learned models.
cadentj's Repositories
cadentj/case_study
cadentj/nngine
cadentj/nnterface
cadentj/apush-textbook
A website I built to host pdf scans of the textbook for my AP US History class.
cadentj/ARENA_3.0
cadentj/autointerp
cadentj/baulab
cadentj/bhc
cadentj/boston
Some visualizations and stakeholder maps I built while consulting for the City of Boston.
cadentj/stuff
cadentj/cogstuff
cadentj/cogworks
My team's code for the MIT CogWorks course in the summer of 22.
cadentj/crosscoders
Open source replication of Anthropic's Crosscoders for Model Diffing
cadentj/demo
cadentj/dictionary_learning
cadentj/edge-attribution-patching
Code for my NeurIPS 2024 ATTRIB paper titled "Attribution Patching Outperforms Automated Circuit Discovery"
cadentj/fact_localization
cadentj/gt
cadentj/ndif
cadentj/ndif-website
cadentj/nnsight
The nnsight package enables interpreting and manipulating the internals of deep learned models.
cadentj/nnsight-docs
cadentj/nvim
A launch point for your personal nvim configuration
cadentj/oai_autointerp
cadentj/pytorchviz
A small package to create visualizations of PyTorch execution graphs
cadentj/qwang
cadentj/receipt-bot
Small python bot/gui that connects to your email inbox and reads college emails to send receipts!
cadentj/sae_vis
Create feature-centric and prompt-centric visualizations for sparse autoencoders (like those from Anthropic's published research).
cadentj/transformers_notebooks
cadentj/trl
Train transformer language models with reinforcement learning.