I’ve imported a large 2 hr session but want to break it into smaller 5/6 minute songs. However the memory of the larger files seem to follow each smaller song. How can I alleviate this?

Hi. Perhaps try to bounce to clip after each split?
All the best.

Ok, I’ll give that a try

You can save each split as a project and should see each's audio file be smaller.

I believe that is the issue here Each split retains the whole clip/tracks info.
Bounce to clip only preserves the audio selected/in question.
I might be wrong though
Edit: For using eaxh song as a separate project, the above idea is better.

There's few ways that I've used to split up live recordings...First make sure you back up the original ,,, twice..
 
If you have the 2 hours of audio when you "save as"  to that folder each song  can use the original audio without duplicating it, so the folder should not grow in size by anything worth worrying about. 
 
 
 
Method one.
You can set Sonar to highlight to the right or the left of a split. I set mine to highlight to the right.  
Now split between song one and two and delete the tracks to the right of the split. 
Ctrl/shift/A to unselect. 
SAVE AS and name the song. 
DO NOT check the "copy audio with project " box.. You don't need to if you stay in the same folder. What you'll end up with is  CWP icons one for each song and only 1 audio folder. The CWP's  will  each reference the audio in the original folder.  
Carry on with next song by using "undo" to bring back the whole file. 
 
 
Method Two   
Go through whole file and make your split points and tag with markers. 
Now Copy and paste each song to a new Project. 
You have 2 choices here when saving. 
You can leave it in the original folder and use the original audio folder. Or ( what I like doing) you can create a whole new folder for each song and then you will use the "copy Audio" box to create an audio folder in each song folder. Only the audio from that song will be put in the audio folder so when your done you can get rid of the original and archive it away. 
 
 
 
 
 
 

Worth mentioning is what I also do is set up the mix and all my plug ins before I split. 
Generally live recordings require the same treatment for all songs. After all unless the recording was post FOH and the sound person was always changing things most live recordings are static. 
Once you get song 1 dialed in the rest are going to be the same other than boosting solos or messing with harmonies. 

https://www.youtube.com/watch?v=m1w2buVhtvk
 
Bounce to clips is the trick. This video is a bit dated, but it might help.
 
This is not how I do it today.
Today, I split on either side of the song in the big project and add a marker at the points of the split.
Then I select all the clips.
Bounce to clips.
Then I use ctrl+shift+page up and page down to move between the markers and ht S to split again.
Next, you select all the clips and ctrl C then paste them into the new project.
 
You can go into the audio folder and should see all the files are the same size and are not too big.
 
Now, if you normalize all the tracks, sometimes you end up with one really big wave file. I'm not sure why and when this happens, but it can be confusing if you are expecting lots of tracks.
 
 

Thank you all. I forgot to mention I have preexisting songs already created. The live recording was to get the audience ,ambience, and lead singers energy
. So I have to figure out how to make that happen.