Sizes in ancient texts sometimes used hyperbole e.g. "as many as the grains of sand on the shore" is a typical example. But also a factor in ancient languages is that they used homonyms just as the English language does, and distinguished differences from context. (Example: "row" has several meanings in English, I can think of 3 immediately and there may be more.)

To take a biblical example, "אלפ" means thousand in ancient Hebrew. But it originally meant "head" or "head of an ox", and in context regarding numbers of people in the 12 tribes it also meant "head of the household".

So in Numbers 2 where you see the listing of the peoples, you might see a modern translation like: "The leader of the people of Judah is Nahshon son of Amminadab. His division numbers 74,600." Then: "The leader of the people of Issachar is Nethanel son of Zuar. His division numbers 54,400." And so on.

But if you add up all the numbers you get: 74,600+54,400+57,400=186,400; 46,500+59,300+45,650=151,450; 40,500+32,200+35,400=108,100; 62,700+41,500+53,400=157,600. Then 186,400+151,450+108,100+157,600=603,550. However there is something strange. If you then go right to numbers 3:43 it says: "The total number of firstborn males a month old or more, listed by name, was 22,273". So really -- if there were 603,500 men able to fight, but between them all they only had 22,273 firstborn sons? Something is wrong.

One attempt at handling this is to assume in context that Judah consisted of 74 households of 600 men, and Issachar had 54 households of 400 men, and so on. This approach results in 598 households of 5,550 men. But this results in too small of a number of men, because they couldn't have 22,273 firstborn sons. Or if it was 22 households with 273 firstborn sons, that is too few firstborn sons for 5,550 men.

So no matter how you slice it, the numbers are messed up. A clue exists in the name of the book (Numbers) but no one has I think ever figured out an explanation that satisfies everyone.