From Nasa’s site:
The computer was tasked with receiving a large command load at the same time it was engaged in compressing previous science data. The main computer responded precisely as it was programmed to do, by entering safe mode and switching to the backup computer.
Thirty observations were lost during the three-day recovery period, representing less than one percent of the total science that the New Horizons team hoped to collect between July 4 and July 16.
https://www.nasa.gov/feature/nasa-s-new-horizons-on-track-for-pluto-flyby
More dramatic writeup extracted from the book Chasing New Horizons:
It was going to take an enormous amount of work to get the spacecraft back on schedule for the flyby. First they had to get it out of “safe mode” — the state the spacecraft goes into when it senses a problem, where every noncritical system is shut off. But there was much more to do to restore the flyby than just that. All the computer files that had been meticulously uploaded since December to support the coming exploration of Pluto would have to be reloaded to the spacecraft before the flyby operations could begin. This would be weeks of work under normal circumstances; but they didn’t have weeks, they had ten days until New Horizons reached Pluto and only three days until the start of the critical data taking for closest approach, when all of the most valuable scientific observations would be made.
Hat-tip to @sohkamyung