My guess is that this seems to happen when the locked mouse cursor reaches the edge of the screen and the engine flips it to the opposite side of the screen. At this time, the cursor lags at 144 hertz. Perhaps the implementation of this functionality was carried out only at 60 hertz? But did not check for 144 hertz?
I tried to ignore the jumps, the lag is much less, but still noticeable.
The best result was given by using the previous value, but the lag is still present.