pub fn launch_dense_matrices(
gpu: &dyn GpuBackend,
x: &[f32],
gate: DevicePtr,
up: DevicePtr,
down: DevicePtr,
gate_dtype: u32,
down_dtype: u32,
hidden: usize,
inter: usize,
beta: f32,
linear_beta: f32,
stream: u64,
) -> Result<Vec<f32>>Expand description
2026-09-25: Runs the gate/up SiTU kernel and the down kernel on explicit device weights. Dtype tags: 0 = FP32, 1 = BF16.