Stars
Track your Claude usage from the macOS menu bar. Visualize consumption over time, get smart plan upgrade/downgrade recommendations, and never miss a promo.
GPU-accelerated electrophysiology (EEG, iEEG, LFP) transforms for large batch jobs.
A framework for collecting and analyzing prediction market data, including the largest publicly available dataset of Polymarket and Kalshi market and trade data.
An open source library designed to provide community examples of Joint Embedding Predictive Architectures (JEPAs). It contains code and examples for learning representations from images, video, and…
Zoomable, animated scatterplots in the browser that scales over a billion points
An interactive NVIDIA-GPU process viewer and beyond, the one-stop solution for GPU process management.
A lightweight data processing framework built on DuckDB and 3FS.
Code and weights for the paper "Cluster and Predict Latents Patches for Improved Masked Image Modeling"
ViT Prisma is a mechanistic interpretability library for Vision and Video Transformers (ViTs).
🌐 Make websites accessible for AI agents. Automate tasks online with ease.
supporting pytorch FSDP for optimizers
gmix is an attempt to build a large language model using techniques from the field of lossless data compression
A Python toolbox for conformal prediction research on deep learning models, using PyTorch.
An explanation of patents from Neuralink in PPT format (some including Python Jupyter notebooks). For educational purposes only.
A library for mechanistic interpretability of GPT-style language models
A collection of pairing functions.
Learning in infinite dimension with neural operators.
[VLDB'24] Blitzcrank is to compress in-memory, OLTP databases. It introduces a new entropy coding algorithm named Delayed Coding.
A 1D analogue of the MNIST dataset for measuring spatial biases and answering Science of Deep Learning questions.
An implementation of 1D, 2D, and 3D positional encoding in Pytorch and TensorFlow
A simple way to keep track of an Exponential Moving Average (EMA) version of your Pytorch model
PyTorch code and models for V-JEPA self-supervised learning from video.
Minimal Implimentation of VCRec (2024) for collapse provention.
A concise but complete full-attention transformer with a set of promising experimental features from various papers
A scientific instrument for investigating latent spaces
Adapt AutoSAM to segment Arctic thermal infrared images into melt ponds, sea ice and ocean.