Update README.md

README.md

@@ -1,8 +1,22 @@
 ---
 license: mit
 datasets:
+- ILSVRC/imagenet-1k
 - mlfoundations/datacomp_small
 base_model:
-- laion/CLIP-ViT-g-14-laion2B-s12B-b42K
+- laion/CLIP-ViT-g-14-laion2B-s34B-b88K
 ---
-ViT-H OpenCLIP initialized from https://huggingface.co/laion/CLIP-ViT-g-14-laion2B-s12B-b42K. The Text encoder is finetuned with the FARE loss using the charmer attack with $\rho=50$ and $k=1$.
+
+Model Initialized from `laion/CLIP-ViT-g-14-laion2B-s34B-b88K`. The text encoder is finetuned with LEAF at $k=1$ with $\rho=50$.
+
+To load this model use:
+
+```python
+from transformers import CLIPProcessor, CLIPModel
+
+model_name = "LEAF-CLIP/OpenCLIP-ViT-g-rho50-k1"
+processor_name = "laion/CLIP-ViT-g-14-laion2B-s34B-b88K"
+
+model = CLIPModel.from_pretrained(model_name)
+processor = CLIPProcessor.from_pretrained(processor_name)
+```