fp8_twin_bytes

spark_model::weight_loader::fp8_residency

Function fp8_twin_bytes 

Source 
pub fn fp8_twin_bytes(n: usize, k: usize) -> usize
Expand description

Bytes one FP8 [K, N] transposed twin costs: the E4M3 bytes plus the transposed [K/128, N/128] FP32 block-scale grid. Mirrors Fp8Weight::transpose_for_gemm (weight_map/quantized.rs:643/:660).