It says no such thing. The "large number" of underpaid men is not being compared against the number of underpaid women.
The article still contains these quotes, which clearly establish that the equity pay raises went disproportionately to men:
> The study, which disproportionately led to pay raises for thousands of men [...]
> In response to the study, Google gave $9.7 million in additional compensation to 10,677 employees for this year. Men account for about 69 percent of the company’s work force, but they received a higher percentage of the money.