flashinfer.cake_rmsnorm_silu_gate_train.cake_rmsnorm_silu_gate

flashinfer.cake_rmsnorm_silu_gate_train.cake_rmsnorm_silu_gate(x: Tensor, z: Tensor, weight: Tensor, eps: float = 1e-06) → Tensor

Differentiable RMSNorm(x) * weight * SiLU(z) over the last dimension (128): CakeRMSNormSiLUGateFunction applied to the arguments.