README: drop backward-op note

README.md

@@ -13,8 +13,7 @@ Forward Hadamard transform CUDA kernel, packaged for the
 [`kernels`](https://huggingface.co/docs/kernels/index) library.
 
 `fp32` / `fp16` / `bf16`, last dim from `1` up to `32768` (zero-padded to the
-next power of two internally). No backward op — `H` is symmetric, so apply
-the same kernel a second time.
+next power of two internally).
 
 ## Use