[context33] - add batch tiling (N_REPEAT=500) for GPU utilisation

Browse Source

Tile the single sentence 500× to form a (500, T, 40) batch so matrix ops
are (500, 168) instead of (1, 168). Note: with identical rows the gradient
is mathematically equivalent to batch=1; real speedup requires diverse data.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

...

This commit is contained in:

jens

2026-05-31 20:03:03 +02:00

co-authored by Claude Sonnet 4.6

parent bbc03d37e1

commit 586803856f

1 changed files with 104 additions and 123 deletions

Show all changes Ignore whitespace when comparing lines Ignore changes in amount of whitespace Ignore changes in whitespace at EOL

Download Patch File Download Diff File Expand all files Collapse all files

context33.ipynb

+104 -123

File diff suppressed because one or more lines are too long