Using uperscript exponents in Log axis

Hi, I know that with Bokeh>2.0 there is the possibility of having superscritp exponents in a log axis, however I have not been able to use them.
Here is an example using Holoviews, which accepts bokeh tick formatters for the axes:

import holoviews as hv
from bokeh.models import LogTickFormatter
hv.extension('bokeh')

hv.Curve([(0, 1e3), (1,2e7), (3,4e12)]).opts(logy=True, 
                                             yformatter=LogTickFormatter(use_superscript_exponents=True))

I get:

AttributeError: unexpected attribute 'use_superscript_exponents' to LogTickFormatter, possible attributes are js_event_callbacks, js_property_callbacks, name, subscribed_events, tags or ticker

I’m using bokeh version 2.0.2

Am I doing something wrong?

Thanks

There is no such functionality. The PR that was intended to bring it was closed: https://github.com/bokeh/bokeh/pull/9005

You can use this instead: https://stackoverflow.com/a/58895556/564509