pub fn ssm_replay_row_bytes(qkvz_elems: usize, nv: usize) -> usizeExpand description
2026-09-25: Bytes of one cached verify row of GDN inputs, per SSM layer: the
deinterleaved qkvz row (qkvz_elems BF16, the rows of
ConvGdnArgs::deinterleaved) plus the gate/beta row (nv * 2 FP32, the
rows of ConvGdnArgs::gates_buf).