pub fn hc_pre_lowrank(
gpu: &dyn GpuBackend,
kernel: KernelHandle,
streams: DevicePtr,
w: &HcLowRank,
y_out: DevicePtr,
inj_out: DevicePtr,
scratch: DevicePtr,
num_tokens: u32,
hidden_size: u32,
hc_mult: u32,
norm_eps: f32,
stream: u64,
) -> Result<()>Expand description
Collapse the hc_mult streams to one, and emit the per-stream injection
weights the matching hc_post_lowrank needs.
streams [T, hc, H] -> y_out [T, H], inj_out [T, hc].