i don't know how they get to 0.1 resolution, i would guess that if it were an important data item that it would have a lower PID value. drcat posted these findings which you likely are already aware:
374 (D1 -10)/2 = SoC
374 D7/2 = C [Capacity]
373: (((D3 * 256) + D4) - 32768) * -0.01 = Pack Amps Out (#)
373: ((D5 * 256) + D6) * 0.1 = Pack Volts (#)
He is looking for the voltage reading of the capacitor rise in the MCU in order to spoof the CAN buss to solve the P1A15 DTC error. That value is in the data items list of the MCU, but maybe there is a separate buss between MCU and EVECU? such that it doesn't show on the OBDII port..?
He posted a great video of his setup in the DIY battery upgrade thread and has a github with code and examples, etc, see github/dkotin/can_2515 folder
Kolyandex posted about the first 2 bytes of PID 374:
1st byte is "Control SOC",
second byte is "Display SOC"
i take that to mean the control value to the EVECU and the value sent to the dash fuel gauge display.