Forward's donut
The formula in Forward's classic paper is $$G=-\frac{d}{dt}\left(\frac{\eta N T r^2}{4\pi R^2}\right )$$ where $NT$ is the total mass current ($N$ windings of pipes carrying a heavy liquid in a spiral around the torus - here we will use the donut mass) and $\eta=3.73\cdot 10^{-26}$ m/kg. So plugging in your numbers, $r=0.03$ m, $R=0.05$ m and we assume $v(t)=0.03 a t$ kg m/s for some acceleration $a$ I get $3.2057\cdot 10^{-29}a$ , so to get "antigravity" we need $a=3.1194\cdot10^{29}$ m/s$^2$. That donut better be pretty indestructible.
(The acceleration is actually in principle physically possible, just a few orders of magnitude above electrons in wakefield accelerators, way below the Planck acceleration).
Tajmar's donut
One somewhat similar calculation can be found in Tajmar, M. (2010). Homopolar artificial gravity generator based on frame-dragging. Acta Astronautica, 66(9-10), 1297-1301. For a pair of rotating disks he states the field at the center as $B_g=(4G/c^2)mr\omega$ where $m$ is the disk mass, $r$ their radius and $\omega$ their angular velocity. The prefactor is $4G/c^2\approx 3\cdot 10^{-27}$.
Note that the gravitomagnetic field acts on a particle with mass only if it is moving, just as a magnetic field will only affect moving charges. The force is at right angles to the field and velocity, and proportional to the speed. This is why there has to be an accelerating flow around the torus in Forward's paper: had it been constant there would have been a constant gravitomagnetic field, and there would not have been any acceleration of particles inside the torus.
Tajmar suggests having a cabin moving at constant velocity along a hallway with the spinning disks to provide a velocity. However, the final model in the paper has a ring-shaped cabin surrounded by two rings of spinning disks that themselves spin around the centre. This way one can enjoy artificial gravity without having to move. While this model is a bit donut-shaped it doesn't correspond to a plausible motion of donut dough.
