Regarding Noose Hoover thermostat relaxation time (Tdamp) for NPT cooling rate simulation

Dear all,
I have some small queries regarding Noose Hoover thermostat relaxation time (Tdamp) for NPT cooling rate simulation. I have used a time step is 5 fs and for Noose Hoover thermostat and barostat time constants are 1 ps and 5 ps in cooling rate simulation. I would highly appreciate if anyone gives suggestion about the effect of temperature as a function of time during cooling rate simulation. In my case up to the liquid state, temperature fluctuation is close to the bath temperature but when it starts transforming into an amorphous state initially temperature is increased for the small time and then start decreasing, finally stalled to the bath temperature. I don’t have any idea why temperature is increasing across the transformation region (from liquid to amorphous)? Is it because of Tdamp parameter?

Thank you very much for your time and consideration. I hope to hear from you soon.

Yours sincerely,

Nandlal Pingua