fix website link

README.md

@@ -15,7 +15,7 @@ However, diffusion models require noisy input images, which destroys information
 
 
 We introduce CleanDIFT, a novel method to extract noise-free, timestep-independent features by enabling diffusion models to work directly with clean input images. 
-The approach is efficient, training on a single GPU in just 30 minutes. We publish these models alongside our paper ["CleanDIFT: Diffusion Features without Noise"](https://compvis.github.io/CleanDIFT/).
+The approach is efficient, training on a single GPU in just 30 minutes. We publish these models alongside our paper ["CleanDIFT: Diffusion Features without Noise"](https://compvis.github.io/cleandift/).
 
 We provide checkpoints for Stable Diffusion 1.5 and Stable Diffusion 2.1.