To fix overfitting on small generative datasets, you can use techniques like data augmentation, regularization, and early stopping.
Here is the code reference using tensorflow:

Hence, this approach combines data augmentation, dropout, L2 regularization, and early stopping to reduce overfitting.