// neural architecture explainers

ML INDEX

interactive visualizations + technical deep dives 20 files
Gen 2 — canonical spec
Gen 2 canonical — 20 files
19
DNN Math Cheatsheet
linear algebra / activation / loss / backprop / transformer
20
CUDA + Tensor Cores
gpu architecture / warps / memory hierarchy / mixed precision
18
Activation Functions
sigmoid / relu / gelu / vanishing gradient
01
Attention Mechanism
transformers / qkv / multi-head / flash attn
02
Attention Mechanism II
transformers / query-key-value / extended
03
Backprop vs STDP
learning rules / bio-plausibility
04
Backpropagation
gradient descent / chain rule
05
Convolutional Networks
kernels / feature maps / receptive field
06
Connectome
network topology / graph structure
07
Diffusion Models
ddpm / score matching / latent diffusion
08
Embedding Geometry
vector space / superposition / homology
09
Embeddings + Vector Space
semantic geometry / dimensionality
10
Hopfield Networks
energy landscapes / associative memory / attention
11
LoRA Explainer
fine-tuning / low-rank adaptation
12
Neuron Comparison
bio vs artificial / morphology
13
Perceptron Walkthrough
code / step-by-step
14
Perceptron Viz
interactive / weights + bias
16
Mamba / SSM
state space models / selective scan / hippo
17
Prompt Engineering
asking nicely / vibes / linkedin certified
15
Reservoir Computing
echo state networks / dynamics