This is a great question, and congratulations on deciding to enter a game jam and start making something! Here are some things I think are good to bear in mind for jamming:

You May Not Finish

I think it took me at least a few game jams before I managed to finish an entry and submit it. Finishing is easily the hardest part of a game jam, because it requires you to think of a small enough idea, do the idea, and deal with any of the things that go wrong. It's really tough! But it's also the main skill you practice in game jams and you do get better fast. Just go in knowing you might not make it this time.

Your Idea Is Too Big

Whatever idea you have for your jam, it's too big. This is a universal truth. Every time you make your idea smaller, your life gets easier, but it will always be too big. I recommend starting out with an idea for a game that takes place on a single screen. It's a massive constraint, but it really helps focus your thoughts on a few ideas and avoids some key game features that sound simple but actually take a lot of time.

Finish On Day One

Something I started doing pretty fast was finishing my game on the first day of a two-day jam (or halfway through a jam of a different length). By "finishing" I mean I had implemented all of the systems and controls I wanted, and the game was sort of feature complete. This meant the second half of the time could be used for creating content (like levels), art, and polish. It also helps make your ideas even smaller because you have even less time to make the core of it!

This is a really good way to ensure you don't neglect the other parts of a game, like polishing, and get to practice a bit of everything. I felt happier about those game jam games after I was done.

You Can Use Anything

You said you'd downloaded Unity - that's great! I use Unity every day. Learning a tool during a game jam is hard mode, however, and learning a 'big' tool like Unity is even harder. I personally would recommend trying a tool that has less moving parts. Puzzlescript is a great way to make puzzle games fast, Ren'Py and Bitsy are good for making story-driven games or exploration/narrative games.

If you feel comfortable with code, I would highly recommend p5.js and/or p5.play, which are great for rapid prototyping. Game Maker is a classic for a reason and pretty good (although a big tool to learn, possibly, just better than Unity). Game Boy Studio is also excellent and lightweight.

I don't want to say 'don't use Unity' because you could just use the weekend as a learning experience for Unity! But be aware that you may not finish something as a result.

Don't Overwork

Jams are pretty good at encouraging this these days, but my advice would be don't spend more than six hours a day on your jam game. I made games I was really proud of in just eight hours (admittedly they had no art, it was just coloured squares) but crunching on a jam is absolutely the worst thing. Go take a walk, buy an ice cream, sit in the park a bit. We can only make good games if we're healthy and happy.

Good luck!