KDE for a logarithmic data in plotly

I'm trying to use plotly violin plot for a logarithmic data set with positive values only.

when I use the regular violin aka:

fig= go.Figure()
fig.add_trace(go.Violin(x=last_value, line_color=color, name=round(leverage, 3),spanmode='hard'))
fig.update_traces(orientation='h', side='positive', width=3, points=False)
fig.show()

I get :

adding the line

fig.update_xaxes(type='log')

I get

As the data is supposed to reflect probabilities a log transformation yields the following.

fig.add_trace(go.Violin(x=np.log10(last_value), line_color=color, name=round(leverage, 3),spanmode='hard'))

The problem is that I want the x-axis and hover data to reflect the actual data, not the log10 of the data. Is there any way to achieve this?

A short summed up example code :

last_value=np.load('data.npy')

fig=go.Figure()
fig.add_trace(go.Violin(x=last_value, spanmode='hard'))
fig.update_traces(orientation='h', side='positive', points=False)
fig.update_xaxes(type='log')
fig.show()

fig=go.Figure()
fig.add_trace(go.Violin(x=np.log10(last_value), spanmode='hard'))
fig.update_traces(orientation='h', side='positive', points=False)
fig.show()

the data file is uploaded here:

https://fastupload.io/ba8b4df5120a5ce8