It looks like the slowest paths of the transcendental functions were improved by a lot. But how often do these paths get used? The article doesn't say so the performance benefits may be insignificant.
They don't get used very often. I don't know how much though because that would require running the functions for the entire domain of inputs which would be years for the univariate functions and pretty much forever for pow. If you want anecdotal data then I have that: I didn't hit the slow path of log for the entire week I threw different inputs at it. pow hit quite a few in an hour of running but even that was about 1 in 1000 or less. However, since there is no pattern to determine which interval of inputs will have more slow path hitting inputs, you can never say that only 0.1% of the inputs use the slow paths. Consequently, I cannot really give an average case picture based on these improvements.
That said (as I've mentioned elsewhere), my main motive of writing the blog post was to share the methods, which I believe are more interesting than the results themselves. The results are not satisfactory IMO since there is still a lot more work to be done.