Figure 1: Artist's impression of an exoplanet.
Mass and radius are two of the most important properties that define a planet. However, in the effort to detect planets, usually only the mass or radius of the planet is measured. As a result, there is a need to predict a planet's radius based on its mass or a planet's mass based on its radius. Chen & Kipping (2016) present a forecasting model based on a probabilistic mass-radius relation from a sample of 316 objects with well-constrained masses and radii. The objects span nine orders-of-magnitude in mass, from dwarf planets to low-mass stars. They are classified into 4 classes - Terran worlds (i.e. Earth is in this category), Neptunian worlds, Jovian worlds and stars. With this classification, dwarf planets are simply low-mass Terran worlds and brown dwarfs are simply high-mass Jovian worlds.
In the model by Chen & Kipping (2016), there is a transition in the mass-radius relation at ~2.0 times the mass of Earth. This transition marks the divide between solid Terran worlds and gas-rich Neptunian worlds. What this means for solid Super-Earths is that they are expected to have masses not much greater than the mass of Earth (i.e. within ~2.0 times the mass of Earth).
Figure 2: The mass-radius relation from dwarf planets to low-mass stars. Chen & Kipping (2016)
There appears to be no change in the mass-radius relation from Jupiter-mass planets to brown dwarfs. Based only on their mass and radius, brown dwarfs can be seen as high-mass Jovian worlds. Also, there appears to be no change in the mass-radius relation from dwarf planets to Earth-mass planets, indicating that based on mass and radius alone, dwarf planets are simply low-mass Terran worlds. The transition from Neptunian worlds to Jovian worlds occurs at ~0.4 times the mass of Jupiter. With this classification, Saturn is close to being the largest Neptunian world.
Reference:
Chen & Kipping (2016), "Probabilistic Forecasting of the Masses and Radii of Other Worlds", arXiv:1603.08614 [astro-ph.EP]