Files
nvfp4-megamoe-kernel/dsv4/kernels/attention
biondizzle aff208fb4c D5a: Fix LSE formula - lse = ln(row_sum) + row_max * ln(2)
row_max is in scale_log2 domain, need to convert to natural log domain.
attn_max = row_max * ln(2), so lse = ln(row_sum) + row_max * ln(2).
2026-05-23 21:15:14 +00:00
..
2026-05-23 20:08:31 +00:00