In an n-p-n transistor (operating in the active region):

1. The emitter-base junction is forward biased, injecting a large number of electrons into the base.
2. The base is made extremely thin and lightly doped.
3. This ensures that only a tiny fraction (typically $<2\%$) of these electrons recombine with holes in the base.
4. Most electrons ($>98\%$) diffuse across the base and are 'collected' by the reverse-biased collector junction, resulting in a large collector current ($I_C\approx I_E$).