xtrain

Files

Gahow Wang f38beb0346 test: flash finite-diff grad-check uses single-tile clean regime

Match the trusted composed grad-check dims (seq=5<FA_TILE); the multi-tile
online-softmax path is gated by flash_bwd_matches_composed_bwd (seq=40),
sharper than finite-diff on the near-zero grads a long softmax produces.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>

2026-06-17 23:16:20 +08:00

xtrain-autodiff

test: flash finite-diff grad-check uses single-tile clean regime

2026-06-17 23:16:20 +08:00

xtrain-cuda

cuda: fused flash-attention kernel (fwd + flash-style bwd)

2026-06-17 23:10:25 +08:00

xtrain-distributed

test+bins: flash grad-check, flash==composed, PyTorch parity, --flash flag

2026-06-17 23:10:39 +08:00

xtrain-model

test+bins: flash grad-check, flash==composed, PyTorch parity, --flash flag