Looped Transformers
Named by 5 kept items in this workspace.
Search every workspace for Looped Transformers. It matches the name as written, so it can find other things that share it.
The items that name it 5
Reuse layers across loops; Meta found full-model looping best at fixed parameters.
Run the input through transformer layers multiple times, scaling the model through depth rather than added weights.
All techniques in this workspace.