Pinned Repositories
17.835-project
6.867-final-project
Based on the VoxCeleb dataset.
fff
FFF
personal-website
Tony Wang's personal website.
prompt-injection-interp
score_sde_pytorch_lib
Turns score_sde_pytorch into a pip installable library.
slurm-jax
Experiments with running JAX on multiple nodes in a slurm environment.
supercloud-tools
Scripts, configs, and binaries helpful for working on MIT supercloud.
vizFB
A graph vizualization web application for your Facebook friends.
ed1d1a8d's Repositories
ed1d1a8d/fff
FFF
ed1d1a8d/prompt-injection-interp
ed1d1a8d/slurm-jax
Experiments with running JAX on multiple nodes in a slurm environment.
ed1d1a8d/personal-website
Tony Wang's personal website.
ed1d1a8d/score_sde_pytorch_lib
Turns score_sde_pytorch into a pip installable library.
ed1d1a8d/supercloud-tools
Scripts, configs, and binaries helpful for working on MIT supercloud.
ed1d1a8d/ann-connectomics
Connectome analysis of PyTorch models.
ed1d1a8d/CLIP
Contrastive Language-Image Pretraining
ed1d1a8d/adversarial-attacks-pytorch
PyTorch implementation of adversarial attacks [torchattacks]
ed1d1a8d/DescribeDistributionalDifferences
Code for preprint: Summarizing Differences between Text Distributions with Natural Language
ed1d1a8d/DM-Improves-AT
Fork of code for the paper "Better Diffusion Models Further Improve Adversarial Training" (ICML 2023)
ed1d1a8d/edm
Elucidating the Design Space of Diffusion-Based Generative Models (EDM)
ed1d1a8d/EfficientNet-PyTorch-SAT
A PyTorch implementation of EfficientNet and EfficientNetV2 (coming soon!)
ed1d1a8d/hofvarpnir-dockerfiles
ed1d1a8d/inverse-scaling-eval-pipeline
Improved pipeline for running different sized GPT models and plotting the results
ed1d1a8d/jsonl-viewer
ed1d1a8d/KataGo
GTP engine and self-play learning in Go
ed1d1a8d/mal-css
Custom CSS for MyAnimeList
ed1d1a8d/metrics
Machine learning metrics for distributed, scalable PyTorch applications.
ed1d1a8d/micromamba-releases
Micromamba executables mirrored from conda-forge as Github releases
ed1d1a8d/mnist_challenge
A challenge to explore adversarial robustness of neural networks on MNIST.
ed1d1a8d/music
For musical transcriptions.
ed1d1a8d/nngp-ntk-simple
Code demonstrating how to compute NNGP and NT kernels for simple neural network architectures.
ed1d1a8d/pii-acdc
ed1d1a8d/robustbench
RobustBench: a standardized adversarial robustness benchmark [NeurIPS'21 Benchmarks and Datasets Track]
ed1d1a8d/scaling
Studying scaling laws in artificial and biological intelligence.
ed1d1a8d/skypilot
SkyPilot: Run LLMs, AI, and Batch jobs on any cloud. Get maximum savings, highest GPU availability, and managed execution—all with a simple interface.
ed1d1a8d/SmoothAdversarialTraining
Smooth Adversarial Training
ed1d1a8d/TAP
TAP: An automated jailbreaking method for black-box LLMs
ed1d1a8d/white-box
Tools for understanding how transformer predictions are built layer-by-layer