is there a way to improve the precision of floats? i saw someone talking about how it sometimes can take 10.00001 and change it to 9.999998 and cause a wrong answer ive been wondering is there a way to maybe limit the float to only one zero so it wont happen?

but why is it happening?

Your solution being too slow is not a kata issue.
If inputs are not presented to you on failure you can print them yourself. See this and this paragraph of troubleshooting guide to see how.

wat?

if the problem with my code is execution time out (over 12000ms) how can i see the tests that caused it?

Yes, just that.

so i just need to make it more efficient?

Your solution being too slow is not a kata issue.

Your solution produces correct resutls, that's why it passes some tests. However, it's not efficient enough, that's why it passes only 3 tests out of 50.