We combine HST G102 & G141 NIR grism spectroscopy with HST/WFC3-UVIS , HST/WFC3-IR and Spitzer /IRAC [ 3.6 \mu m ] photometry to assemble a sample of massive ( log ( M _ { star } / M _ { \odot } ) \sim 11.0 ) and quenched galaxies at z \sim 1.5 .
Our sample of 41 galaxies is the largest with G102+G141 NIR spectroscopy for quenched sources at these redshifts .
In contrast to the local Universe , z \sim 1.5 quenched galaxies in the high-mass range have a wide range of stellar population properties .
We find their SEDs are well fitted with exponentially decreasing SFHs , and short star-formation time-scales ( \tau \leq 100 Myr ) .
Quenched galaxies also show a wide distribution in ages , between 1-4 Gyr .
In the ( u - r ) _ { 0 } -versus-mass space quenched galaxies have a large spread in rest-frame color at a given mass .
Most quenched galaxies populate the z \sim 1.5 red-sequence ( RS ) , but an important fraction of them ( 32 % ) have substantially bluer colors .
Although with a large spread , we find that the quenched galaxies on the RS have older median ages ( 3.1 Gyr ) than the quenched galaxies off the RS ( 1.5Gyr ) .
We also show that a rejuvenated SED can not reproduce the observed stacked spectra of ( the bluer ) quenched galaxies off the RS .
We derive the upper limit on the fraction of massive galaxies on the RS at z \sim 1.5 to be < 43 \% .
We speculate that the young quenched galaxies off the RS are in a transition phase between vigorous star formation at z > 2 and the z \sim 1.5 RS .
According to their estimated ages , the time required for quenched galaxies off the RS to join their counterparts on the z \sim 1.5 RS is of the order of \sim 1 Gyr .