Docs/Editor

Slash menu

Type / inside a note to insert any block.

~/Research/Transformers.md

Transformers

Sequence models that replace recurrence with attention.

Built on . Introduced in [@vaswani2017].

Self-attention

Every token attends to every other token in one step:

Attention(Q,K,V) = softmax(QK⊤ / √dk) V

See the paper, page 3:

LocalOffline readyMarkdown 60 words · saved

Keep typing to filter: /ta finds Table. ↑ ↓ to move, Enter to insert, Esc to close, Tab to jump between placeholders.

There are about a hundred blocks: text, lists, Callouts, Math, research and study templates, media, tasks and dates. Pin favourites with ★. Every item is also in Mod+K.

Edit this page on GitHub