seitzdom's repositories
custom_jvp_vjp_tutorials
A-Z Tutorials on how to implement custom vjps and jvps in JAX and PyTorch
burn
Burn - A Flexible and Comprehensive Deep Learning Framework in Rust
diffrax
Numerical differential equation solvers in JAX. Autodifferentiable and GPU-capable. https://docs.kidger.site/diffrax/
jax
Composable transformations of Python+NumPy programs: differentiate, vectorize, JIT to GPU/TPU, and more
levanter
Legible, Scalable, Reproducible Foundation Models with Named Tensors and Jax
LLMs-from-scratch
Implementing a ChatGPT-like LLM from scratch, step by step
PiPPy
Pipeline Parallelism for PyTorch
dominikandreasseitz
My personal webpage
dominikandreasseitz.github.io
My personal webpage
dspy
DSPy: The framework for programming—not prompting—foundation models
llm.c
LLM training in simple, raw C/CUDA
rag-demystified
An LLM-powered advanced RAG pipeline built from scratch
tknzr
Tokenization using Rust
txfm
JAX Transformer implementations