biondizzle 1330e2b2cf cleanup: remove debug prints, ready for testing ...

Current state:
- Token indices on CPU (avoids CuTeDSL GPU memory corruption)
- Scale assembly uses per-expert swizzle + scatter (matches reference)
- compute_activation_global_scales warmup gets ~0.97 cosine
- expert_offsets passed without leading 0 (matches pipeline)
- layertest + cudagraph_test pass

2026-05-17 08:30:41 +00:00

..

patches

fix: set _l1_activation_global_scale (with underscore) — attribute name mismatch

2026-05-17 03:35:20 +00:00

nvfp4_cutedsl.py

cleanup: remove debug prints, ready for testing

2026-05-17 08:30:41 +00:00