gevfit repeatability of T for 12.5 years. Calculating return level of extreme temperature using Extrem Events Analysis

Hello everyone, here is a problem. Data since 1965 to 2022, I average the moving average over five days. After that I do the following
data = load('RAW.txt');
disp(data(1:5,:))
monthV = data(:,1);
figure(7)
plot(monthV,data(:,3),'r');
temp = data(data(:,3)~=0,:);
temp = temp(:,3);
returnT = [2;10;12.5;25;50;100];
returnP = 1-1./returnT;

parmhatT = gevfit(temp);
v1 = gevinv(returnP,parmhatT(1),parmhatT(2),parmhatT(3));
returnpV = 1.1:100;
returnTp = 1-1./returnpV;
figure(8)
returnLevel_T = gevinv(returnTp,parmhatT(1),parmhatT(2),parmhatT(3));
plot((returnpV),returnLevel_T,'k',returnT,v1,'k*');hold on
%ylim([23 31]); grid on
title('Temperature Maxima');
xlabel('Return period [years]');
ylabel('Return level [deg F]');
grid on
But MatLab give next error:
Warning: Maximum likelihood estimation did not converge. Iteration limit exceeded.
> In gevfit (line 119)
In Obobthennie_Raspredekenie (line 69)
Example data:
01.01.1965 0:00 0.08
02.01.1965 0:00 0.54
03.01.1965 0:00 0.62
04.01.1965 0:00 -0.22
05.01.1965 0:00 -2.06
06.01.1965 0:00 -5.36
07.01.1965 0:00 -7.16
08.01.1965 0:00 -9.08
09.01.1965 0:00 -11.64
10.01.1965 0:00 -13.38
11.01.1965 0:00 -13.3
12.01.1965 0:00 -14.58
13.01.1965 0:00 -14.82
I need to know the repeatability of T for 12.5 years.
What am I doing wrong?