Struct std::DiTBlock

Each block computes 6 modulation vectors (shift1,scale1,gate1, shift2,scale2,gate2) from a conditioning vector c via one linear layer, then runs a pre-LN self-attention sublayer and a pre-LN FFN sublayer, each modulated by its half of the 6 and gated additively into the residual stream (the "adaLN-Zero" block from the DiT paper, minus the SiLU the paper puts before WadaLN and minus the zero-init convention — both orthogonal to the block's wiring, which is what this implements and verifies):

mod = c @ WadaLN                                    [1, 6*dModel]
h = LN(x); h = h*(1+scale1)+shift1
x = x + gate1 * (MHA(h@Wq,h@Wk,h@Wv,heads) @ Wo)
h2 = LN(x); h2 = h2*(1+scale2)+shift2
x = x + gate2 * (relu(h2@W1+b1)@W2+b2)