Grimoire
Tree — 01
attention/2
normalisation/1
position/1
sparse/1

Architectures

The shapes a forward pass can take. Attention and its cheaper cousins, how position is injected when the operation itself is permutation-invariant, where normalisation goes and what breaks when it goes elsewhere, and the conditional-compute blocks that buy parameters without buying FLOPs.

Five entries. One is a stub and is marked as such — read them as open questions rather than answers.

Entries
01.1
Mixture-of-Experts Routing
Top-k gate over expert FFNs with auxiliary load-balancing loss.
O(k·d²)
01.2
RMSNorm
Re-scaling without re-centring; LayerNorm minus the mean subtraction.
O(n·d)
01.3
Rotary Position Embedding
Relative position injected as a per-pair rotation in complex subspaces.
O(n·d)
01.4
Scaled Dot-Product Attention
Content-based retrieval over key–value pairs with 1/√d_k logit scaling.
O(n²·d)
01.5
Sliding-Window Attentionstub
Local band mask; receptive field grows linearly with depth.
O(n·w·d)