Hello,
I managed to utilize the following notebook for finetuning Gemma3-1b. I tried to run the same script utilizing Gemma3-13b, obtaining some errors during the finetune stage (notice I am using A100 as GPU and I didn't change any configuration besides the name of the model). My question is whether versions of Gemma3 with more parameters can be finetuned with this notebook or if, it is not supported, which tutorial/resource I can follow to kickstart the finetuning of Gemma3 13b.
Link to the tutorial used:
https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/community/model_garden/model_garden_gemma3_finetuning_on_vertex.ipynb