This is probably necrobump, but this thread has been a very rich and elightening resource that I feel I need to contribute.
Mid-way through the development of my Faraday60 I removed the TP1684 from an old HHKB and started reverse-engineering it. I opened it up and using a litography-specific microscope I could take a look at the internals; I suspect it is a ultra-low-capacitance precision peak detector -- it detects a voltage peak and returns its amplitude, using JFET inputs it can achieve very high response times with a very low input current. I also believe that some of its functioning depend on a bias adjustment of its pins.
As far as I know there are no integrated circuits that have the same functionality, which only tells us that perfectly replicating the original Topre circuitry is impossible unless we get our hans on some of these sensors, which I highly doubt we are able to do unless by taking it from sacrifice boards.