Update model card (step 28149)

README.md

@@ -16,8 +16,8 @@ Multi-modal Variational Autoencoder for text embedding transformation using geom
 
 - **Fusion Strategy**: cantor
 - **Latent Dimension**: 768
-- **Training Steps**: 27,524
-- **Best Loss**: 0.1849
+- **Training Steps**: 28,149
+- **Best Loss**: 0.1847
 
 ## Architecture