How to calculate error from the coefficients given by gaussian model (for 95% confidence bounds ) after fit ?

For eg , I got these parameters,

f(x) = a1*exp(-((x-b1)/c1)^2) Coefficients (with 95% confidence bounds): a1 = 0.8236 (0.3624, 1.285) b1 = 1003 (801.9, 1204) c1 = 391.3 (95.29, 687.4) Here b1 is time lag ,a1 is peak value and c1 is width

i want to find uncertainity in time lag i.e b1 . so error should be in form of say x+_del(x)