Evaluating model score after inverting from logarithmic x+1

Question

I stacked with question about inverting the r-square model value after taking log1p(). So before logarithmic actions r-square was 0.59, after using log1p()