A decimal number is stored as a sign, a fraction, and a shift, in a fixed number of bits. That fixed width is why some decimals are rounded and why a huge one can become infinity.
About 1 min of reading
Probe the limits of a float.
Print a tenth sum to see the rounding.
Before you run it, say out loud what it will print. Then open the answer.
Multiply past the largest float.
Before you run it, say out loud what it will print. Then open the answer.
Print a third and see it cut off.
Before you run it, say out loud what it will print. Then open the answer.
Add a tiny number to a huge one.
Before you run it, say out loud what it will print. Then open the answer.
Read the first example one line at a time. A line that prints shows what appears on screen; the rest set things up quietly.
A float has a fixed number of fraction bits, so it carries a fixed count of digits of accuracy. Adding a tiny number to a huge one can change nothing at all.
For exact money or counts, keep whole numbers. Floats are for measured quantities where a tiny error is acceptable, which is most science and graphics.
print(1e308)
print(1e-308)A float is a sign, a fraction, and an exponent that shifts it. The exponent is what lets one width hold tiny and huge numbers.
print(0.1 + 0.2)A tenth has no exact binary fraction, so the nearest one that fits is stored. The sum shows that rounding, which is the same fact from the values subject.
print(1e308 * 10)Past the largest float, multiplying gives infinity. The exponent runs out of bits, and the value has no room.
print(1 / 3)A third is an endless binary fraction cut off at the width. The stored value is the closest that fits, not the true third.
Each one is defined here on this site. Open the entry for the full definition and the other words it points at.