When you say “which were run with a Pipeline execute”, may I assume you mean you want to validate a child pipeline that was fed from a “parent” pipeline?
If so, then the “Record Replay” snap is a great solution to this.
Your child pipeline should have a single open input somewhere in the pipeline where it receives the document stream from the parent. Put the Record Replay snap just before it and save the pipeline. Then go back and execute or validate the parent pipeline so that it feeds the child with a docstream. Even if the child fails, if the parent otherwise ran and gave it a docstream, you’re good. It will have saved this as a document with a unique filename in the project.

Now you can open the child pipeline in the SnapLogic Designer canvas and validate it. The replay snap will get its input from the file and feed it to the next snap, and now you can see what’s happening.
Note: If, however, you have a manually created or faked doc stream of your own, don’t try to put it in a file and have the replay snap read from it. That won’t work, because it has it’s own funky syntax for writing the JSON. Instead of the replay snap, use a JSON generator snap and put your “faked” or manually created documents there.
Does this help?