diff --git a/README.md b/README.md
index a016660..b4928b1 100644
--- a/README.md
+++ b/README.md
@@ -2,7 +2,7 @@
## Brain Decodes Deep Nets
PyTorch Implementation
-### [📄Paper](https://arxiv.org/abs/2312.01280) | [🌐Website](https://huzeyann.github.io/brain-decodes-deep-nets) | [](https://drive.google.com/file/d/158tm5AvCc346SDM1WqxanRjqcJFh5k5X/view?usp=sharing]
+### [📄Paper](https://arxiv.org/abs/2312.01280) | [🌐Website](https://huzeyann.github.io/brain-decodes-deep-nets) | [](https://drive.google.com/file/d/158tm5AvCc346SDM1WqxanRjqcJFh5k5X/view?usp=sharing)
We visualize pre-trained vision models by mapping them onto the brain, thus exposing their hidden inside. Visualization is a by-product of brain encoding model: predict brain fMRI measurements in response to images.