pub fn softplus_gate_mul_head_broadcast(
gpu: &dyn GpuBackend,
kernel: KernelHandle,
input: DevicePtr,
gate: DevicePtr,
output: DevicePtr,
nq: u32,
hd: u32,
num_tokens: u32,
stream: u64,
) -> Result<()>Expand description
Per-head softplus gate multiply with broadcast over head_dim.