Top cited papers plot is incorrect after switching to Bokeh 2.3.2

[olegs]

[olegs]

Number of citations is treated as negative for some reason, so that log scaling doesn't apply.

Data itself:

year        id  ... paper_year count
0   1985   2857475  ...       1985     2
1   1986   2997788  ...       1985     4
2   1987   2997788  ...       1985     8
3   1988   2997788  ...       1985     5
4   1989   2997788  ...       1985     7
5   1990   2997788  ...       1985     6
6   1991   2777259  ...       1989     6
7   1992   2777259  ...       1989     5
8   1993   2777259  ...       1989     2
9   1994   2997788  ...       1985     3
10  1995  25969563  ...       2015     1
11  1996   2997788  ...       1985     2
12  1997   2777259  ...       1989     1
13  1998   2997788  ...       1985     5
14  1999   2857475  ...       1985     1
15  2000   2777259  ...       1989     2
16  2001   2777259  ...       1989     1
17  2002   2777259  ...       1989     2
18  2003   2777259  ...       1989     6
19  2004  12727904  ...       2003     5
20  2005  15533437  ...       2004     4
21  2006  12727904  ...       2003     6
22  2007  15533437  ...       2004     6
23  2008  17363581  ...       2007     7
24  2009   2777259  ...       1989     8
25  2010  17436311  ...       2007     8
26  2011   2777259  ...       1989     9
27  2012  21844894  ...       2012     9
28  2013  22438023  ...       2012    11
29  2014  24138928  ...       2013    33
30  2015  24138928  ...       2013    91
31  2016  24138928  ...       2013   138
32  2017  24138928  ...       2013   190
33  2018  24138928  ...       2013   221
34  2019  24138928  ...       2013   242
35  2020  24138928  ...       2013   276
36  2021  30669119  ...       2019     1

[olegs]

The issue is related with log scale y axis, when plotting on linear scale everything is okay.

[olegs]

Relevant open issue in Bokeh issue tracker: bokeh/bokeh#10511