warp.pow# warp.pow(x: Float, y: Float) → Float# Kernel Differentiable Compute x raised to the power of y.