Output of our reference model. Corner plot and posterior distributions (pink) of the four key parameters in our reference model: SFMS slope and normalization at log(M_★/M_⊙) = 10.5 (ms_slope, ms_norm) and the slope and normalization of the intrinsic scatter (a, b). The contours show the 1σ, 1.5σ, and 2σ confidence levels. We adopted truncated normal distributions to model the SFRs distribution in our Bayesian model (see Sect. 3.3). In purple we show the posterior distributions after running the Bayesian model on the robust sample imposing a Gaussian prior (mean = 1 and σ = 0.02) on ms_slope, referred to as “fixed ms_slope” in the legend. Above each panel we report medians with uncertainties (16th and 84th percentiles), those with asterisks refer to fixed ms_slope. The inset panels on the right show the intrinsic scatter distributions at two stellar masses, computed combining the posterior distributions for a and b with Eq. (2), and the σ_int evolution with stellar mass for the reference model and the model with a fixed ms_slope.