pub fn argmax_bf16(
gpu: &dyn GpuBackend,
kernel: KernelHandle,
logits: DevicePtr,
out: DevicePtr,
vocab_size: u32,
stream: u64,
) -> Result<()>Expand description
2026-09-25: Index of the largest of vocab_size BF16 logits, written as one
u32 to out, by a single block.