This is an ensemble mean and not an individual model forecast so one of these could be influenced by member(s) who are drastically different than others, but the High over Easten Canada is weaker at 12Z Sat on the GEFS than it is on the GFS while the "50/50 Low" is also weaker, to the tune of 30Mb, on the GEFS. Would certainly make for a drastically different result. The low is 968 on the GFS and 990 on the GEFS. There is a cluster of lows around the top but speaking about the farther south displaced one.