In total, in an unmitigated epidemic, we would predict approximately 510,000 deaths in GB and 2.2 million in the US, not accounting for the potential negative effects of health systems being overwhelmed on mortality.

In this “unmitigated” means no action is taken to slow the spread of the disease.

The whole paper with the original estimates is literally about the likely numbers of deaths for different mitigation strategies, like doing nothing, having voluntary self-quarantine, having vulnerable populations self-quarantine, and closing schools and universities.  

People telling you that the early models were wildly inaccurate are either ignorant or are intentionally misleading you for their own benefit.