There are the usual differences between the GFS and NAM models – the NAM is colder and wetter- the GFS has much less precipitation and is a bit warmer in the upper atmosphere.
In recent years, the NAM has often over-predicted the amounts of snow with model runs more than a day in advance. So while the NAM is predicting several inches of snow in PHL, I think we can ignore it and go with the GFS.
The GFS model has temperatures falling in the upper atmosphere to support snow just as much of the precipitation winds down. It maintains surface temperatures that are above freezing until the precipitation shuts down. So conditions for snow and snow accumulation are very marginal with the GFS model. While a changeover to snow looks to occur about 2-3 PM, conditions for accumulation will not be optimal because the precip will be winding down. With warm and wet surface conditions, expect snow accumulations to be about 1 inch or so in the immediate Philadelphia area.
Traveling conditions will not be good late afternoon, but the precipitation ends about 7 PM.
Again, much of this will be RAIN, changing to snow 2-3 PM and not looking to accumulate very much before ending by 7PM.