Type a number and watch its float32/float64 sign, exponent, and mantissa bits light up. Tap bits to change the value live, see the exact stored decimal, and learn why 0.1+0.2 isn't 0.3 — denormals, Infinity, and NaN included.
IEEE754 Floating-Point Visualizer
Quick presets
Enter a number
Tap any bit to flip it — watch the value change live
Sign — positive (+)
Exponent (11) — field 1019 → unbiased E = -4
Mantissa (52) — 1.m form — the leading 1 is an implicit bit, not stored (52 bits)
value = +1 × 2^-4 × mantissa
Computers store fractions in binary. Decimal 0.1 is an infinite fraction in binary (0.0001100110011…₂), so it must be rounded to fit 52 (or 23) bits — like writing 1/3 as 0.333… on paper.
Try it: switch 0.1 to float32 — the error grows ~1000×