Add ViT-H (#2)

dinov2/models/vision_transformer.py

@@ -405,6 +405,20 @@ def vit_large(patch_size=16, num_register_tokens=0, **kwargs):
     return model
 
 
+def vit_huge(patch_size=16, num_register_tokens=0, **kwargs):
+    model = DinoVisionTransformer(
+        patch_size=patch_size,
+        embed_dim=1280,
+        depth=32,
+        num_heads=16,
+        mlp_ratio=4,
+        block_fn=partial(Block, attn_class=MemEffAttention),
+        num_register_tokens=num_register_tokens,
+        **kwargs,
+    )
+    return model
+
+
 def vit_giant2(patch_size=16, num_register_tokens=0, **kwargs):
     """
     Close to ViT-giant, with embed-dim 1536 and 24 heads => embed-dim per head 64