marcelbtec/transformer-visualizer

The inner working of a transformer

vimarsh/repro-mimomamba-matrix-valued-attention

Space

Repro - MIMOMamba: From Scalar Duality to Matrix-Valued Attention

keppy/mathmare

small math vision language model

hnmr293/llama-viz

The attention map viewer for LLaMA models.

womenflyplanes/moa-attention-verified-mullin

The first attention kernel proven minimal BEFORE code was written

apple/embedding-atlas

Embedding Atlas is a tool that provides interactive visualizations for large embeddings. It allows…

lamalunderscore/manipulatte

Extensible library for different manipulations of the self-attention mechanism.