On line, on other websites, I have seen that equation as
a +b * exp (-2×(x-mu)**2)/(sigma**2))
Is the formula elsewhere incorrect? Or is it evaluating for a "different b" value of the same Gaussian curve peak of value "b"?