Courtesy of J.H. in comments, it looks like the Russian casualty data is probably fake. They were surprisingly high, but as Will Baird pointed out, quite plausible considering what we know about the number of high-intensity engagements so far.
The problem is that even if 2,000 combat deaths so far is plausible, a rather lower number is more probable. That is why I was surprised when the report first surfaced.
Best to assume for the time being that the report is fake. That said, Forbes has not yet retracted the report. But Forbes may never retract it: the story showed up in one of their blogs and was sourced to a Russian news aggregator. Until they do, some doubt about the report will remain ... but as of now, the evidence is that Russia did not inadvertently (or advertently) publish an estimate of its casualties in Ukraine.