A little birthday cake for someone

a JAX + NNX implementation of Google's Titan NN. Transformer short-term attention + neural long-term memory with test-time updates

A new updated version of BwETA which uses JAX/Flax trained on TPU and has most of the advance features used by huge CLMs

Maximal Update Parametrization (μP) with Flax & Optax.

Kedro project of time-series-forecast of the multivariate weather using TSMixer.

A dump to all things LLMs, CUDA, DL frameworks, Tokenization Methods, and latest stuff I’m building/researching with

Unofficial flax plugin manager will help you add plugins so you dont have to search for them :)

SQA scottish Science Baccalaureate of Christopher Rae. Improving the preformance of distributed data parallelism of low wifi bandwidths

A tiny (8 line) parameter tape implementation.

Jax, Flax, examples (ImageClassification, SemanticSegmentation, and more...)

Comparisons between PyTorch and JAX (Flax)

A re-implementation of Callum McDougall's "Transformer from Scratch" using JAX (Flax).

fcpe jax flax version

Neural implicit digital elevation model

Unofficial JAX/Flax implementation of Pyramid Vision Transformer: A Versatile Backbone for Dense Prediction without Convolutions.

Simple flax engine unit testing