I've been running some tests with the ICM-20689 and it seems like its recording at ~1005Hz with the default SPI sampling frequency, rather than 1000 as stated in the data sheet. I've recorded for short periods of time as well as long periods of time, and there is always more samples than I would expect with a 1k sampling frequency. This effect is most pronounced when the experiment lasts for more than 1000seconds and I end up having more than 5 seconds of extra IMU data. Any thoughts on where things could be breaking down here?