biondizzle a0ff8a3278 fix: transpose checkpoint block scales (N,K_sf)→(K_sf,N) for bridge ...

The bridge's assemble_scales_3d_side expects (K_sf, N) input and
transposes to (N, K_sf) internally before swizzling. The checkpoint
stores scales as (N, K_sf). Without this transpose, the kernel was
reading completely wrong scale data — cosine dropped to 0.713.

Also fixed dual global scale normalization: after transpose, gate/up
are along dim 1 (columns), not dim 0 (rows).

2026-05-16 03:43:30 +00:00

..

layertest.py

fix: transpose checkpoint block scales (N,K_sf)→(K_sf,N) for bridge

2026-05-16 03:43:30 +00:00

requirements.txt

test: add standalone layer 0 comparison test (no vLLM, no Docker)

2026-05-16 02:13:18 +00:00

run_test.sh

fix: use setup.py install for CUTLASS extension build

2026-05-16 02:21:17 +00:00

test_b_layout.py

cleanup: move useful tests to tests/, nuke stale debug tests

2026-05-16 02:14:37 +00:00

test_cutedsl.py

fix: B tensor K-major strides, scale_b axis swap

2026-05-16 03:04:31 +00:00

test_quick_rand.py

cleanup: move useful tests to tests/, nuke stale debug tests

2026-05-16 02:14:37 +00:00

test_uniform_fp4.py

cleanup: move useful tests to tests/, nuke stale debug tests

2026-05-16 02:14:37 +00:00