You should almost certainly not tell them. While I have seen a certain amount of "friendly" post-review disclosure, even if this is accepted in your field, I think the circumstances here strongly weigh against it.
If the paper is currently under review, you should first let the editor know, and absolutely not tell your potential advisor. If you told your prospective advisor, it would create the appearance of a quid pro quo. No matter your intentions, if someone told me this while I was interviewing them, I would be shaken and question their trustworthiness.
If you have already reviewed the paper, and you gave it a good review, and the paper has been published, it would be less blatantly offensive. I understand the urge to say, "I became interested in your group because of this paper, which I refereed" - but this is still not a good idea. Simply showing your strong interest and understanding of that paper is better, and creates no impression of a quid pro quo.
The one circumstance in which I can imagine a lab being offended that you did not disclose your involvement is if you gave a notably harsh review to a paper. The key issue in this case would be finding out if your idea of what is important in a paper, or what methods are useful, might be critically different from your prospective advisor's. If you join the group, but then have major issues with how some experiments are conducted, your advisor might be bothered that you did not disclose that - whether or not this had its origin in a refereeing situation. Disclosing in this circumstance will almost certainly not improve your chances of being hired - no one likes harsh criticism - but it could tell you more about whether this is a lab you could work with.