Labels on y-axis do not match plotted values in scalars view

Just stumbled over something similar as this old issue. So I created a minimal example to reproduce the effect:

# tensorboard_y_axis_labels_issue.py

import keras
import numpy as np
from keras.layers import Input, Dense
from keras.models import Model

inputs = Input(shape=(100,))
x = Dense(80, activation='tanh')(inputs)
predictions = Dense(10, activation='softmax')(x)
model = Model(inputs=inputs, outputs=predictions)
model.compile(loss='categorical_crossentropy', optimizer='nadam')

x_train = np.random.rand(10000, 100)
y_train = np.random.rand(10000, 10)

model.fit(
    x=x_train,
    y=y_train,
    epochs=1000,
    callbacks=[keras.callbacks.TensorBoard(log_dir='tensorboard_logs')])

Screenshot of effect:

From the labels of the y-axis I'd expect the highlighted point to be > 11.5 instead of 11.48.

Command line:

python3 tensorboard_y_axis_labels_issue.py
tensorboard --logdir tensorboard_logs

Used versions:

$ pip3 list | grep tensor
tensorboard              1.12.2
tensorflow               1.12.0

$ inxi -S
System:    Host: tobias-Z170-HD3P Kernel: 4.13.0-38-generic x86_64 (64 bit) Desktop: MATE 1.18.0
           Distro: Linux Mint 18.2 Sonya

$ python3 --version
Python 3.6.8

Run the provided Python reproduction and open the generated logs with the TensorBoard Scalars view to confirm the y-axis mismatch. Trace the Scalars view's axis-labeling behavior and add a regression test if the relevant test location can be identified; done means plotted values and y-axis labels agree, including the highlighted point.