750 MHash/s would be far too high. No single GPU could achieve that, much less a (Fermi-based) Nvidia.
The M2075 appears to be closest to the GTX470 (since they likely share the same GF100 chip and have 448 CUDA cores). Thus, a good estimate would be 80-120 MHash/s, depending on core and shader clock speed.
There's actually a M2050 in the comparison chart (listed as 80 MHash/s). The M2075 also appears to be the same as the M2050, but with more RAM. See http://www.nvidia.com/docs/IO/105880/DS-Tesla-M-Class-Aug11.pdf