Create README.md

README.md

@@ -0,0 +1,19 @@
+---
+license: apache-2.0
+---
+# SAeUron: Interpretable Concept Unlearning in Diffusion Models with Sparse Autoencoders
+
+[Paper on arXiv](https://arxiv.org/abs/2501.18052) |
+[GitHub repo](https://github.com/cywinski/SAeUron)
+
+The repository contains Sparse Autoencoders trained in our work for blocks `up.1.1` and `up.1.2`.
+
+After cloning our GitHub repo, you can use them as follows:
+```python
+from SAE.sae import Sae
+
+device = "cuda"
+hookpoint = "unet.up_blocks.1.attentions.1"
+
+sae = Sae.load_from_hub("bcywinski/SAeUron", hookpoint=hookpoint, device=device)
+```