It looks like Thursday and Friday will be near or below freezing, but the expected high Saturday is 48F, and 56F on Sunday. Chance of 0.05" of rain overnight Sat-Sun.
This is a 10-day weather model that I use all the time:
http://wxmaps.org/pix/lgagfs.png (NYC)
Where it has lines for a date (like 24 Nov), that is GMT midnight, so the line for 24 Nov is 7 or 8pm on the 23rd. Look at the 2m Temp line, and the 3-hr precipitation bars at the bottom.
If anyone is interested, here are the 10-day models for selected cities in the US and Canada:
http://wxmaps.org/pix/meteograms
One thing I use a lot in the summer is the Lifted Index (which has an upside-down scale). When the Lifted Index goes negative, there is a chance of pop-up showers. Interestingly, NYC goes to zero Monday around dark in the middle of that big rainstorm.
Finally, it's a model for *weather*. So it's subject to being very, very wrong. The images refresh twice a day around 8am and 8pm eastern time (~noon and midnight GMT). The rain forecast typically changes ("becomes more accurate") three days out. Sometimes it accurately forecasts stuff over a week away, but sometimes it says it won't rain tomorrow and it actually drizzles all day.