pub const W8A16_GEMM_M16_N_TILE_WIDE: u32 = 64;Expand description
2026-09-25: The wide instantiation’s N tile (w8a16_gemm_m16_n64, kernel
M16_N_TILE_WIDE), selected for the dense FFN by
METRALE_FFN_M16_TC_NTILE=64. It halves the CTA count for a given N.