Update model card for new model version

README.md

@@ -29,8 +29,12 @@ it were only 31. Not all butterfly and moth species can be determined based on i
 
 ## Training Details
 
-The model was trained on the EuroHPC supercomputers LUMI, hosted by CSC (Finland) and the LUMI consortium through a EuroHPC Regular Access call. 
-Training was parallelized using the Pytorch DDP framework and the Hugging Face Accelerate library. More information on model training can be found here (link to publication) and scripts are availble here (link to github). 
+The first model version ([a9ba52f](https://huggingface.co/RikeB/MaxViT_butterfly_identification/commit/a9ba52fd730b165b8fcd58265076fa82e763ebc2)) was trained on the EuroHPC supercomputer LUMI, hosted by CSC (Finland) and the LUMI consortium through a EuroHPC Regular Access call.
+The second model version ([2371ff8](https://huggingface.co/RikeB/MaxViT_butterfly_identification/commit/2371ff89263bcc0319117d2a6e452e1dc5a82b4e)) was trained on the EuroHPC supercomputer LEONARDO, hosted by CINECA (Italy) and the LEONARDO consortium, also through a EuroHPC Regular Access call.
+
+Training was parallelized using the Pytorch DDP framework and the Hugging Face Accelerate library.
+More information on model training can be found in the publications below.
+Scripts are availble on [GitHub](https://github.com/AndiLindner/butterfly_identification). 
 
 ### Training Data
 
@@ -40,7 +44,7 @@ Species with less than 50 images were excluded from training. The final dataset
 
 ## Citation 
 
-The model was trained in the context of a data paper in which the butterfly and moth images dataset it was trained on was published: 
+The first model version was trained in the context of a data paper in which the butterfly and moth images dataset it was trained on was published: 
 
 ```bibtex
 @Article{Barkmannetal2025a,
@@ -64,3 +68,4 @@ url={https://doi.org/10.1038/s41597-025-05708-z}
 }
 ```
 
+Another publication is in preparation.