[sdiy] Ge transistor tester weird results

[cheater cheater]

Hi all,
I recently bought and tested a few Germanium transistors in my small
semiconductor tester and I got weird behaviors.

I have this transistor tester:

https://www.amazon.de/ARCELI-LCR-T4-Transistor-Kapazit%C3%A4t-Induktivit%C3%A4t/dp/B07J2Q4VY9/

As you can see, it's a simple device that has a ZIF socket. It will
auto-find the device's pinout and can figure out what sort of device
you put in it. For a transistor it will calculate the Vf and the hFE.

When I test a new Si transistor, say a BC558, it will auto-find the
BCE pins and show a schematic diagram of a PNP transistor. The hFE
might drift between tests a little, up to 1%.

I also tested the 20 AF306's I got, which are Ge transistors, with hFE
min 10, and they have the same TO-92 package as the BC558. First of
all, the schematic shown always shows the transistor in parallel with
a diode. I assume that's because the Vf is ~0.3V, and the tester
assumes Si, so it guesses there's a diode there.

Second of all, the hFE on those drifts a LOT between tests. As I keep
on repeatedly testing, it will drift up or down by maybe 10-20%, first
quickly and then slowly. The direction of drift is always the same for
the same transistor, but it will be different between transistors. As
the hFE drifts, the Vf drifts with it, too. If I take a small pause of
several seconds, the hFE will drift back close to its initial value,
but if I resume rapid test runs, it will drift away to its "final"
value more readily.

In addition to that, out of the AF306's I tested, there are some in
the hFE 10-40 range, some (fewer) in the 40-100 range, and then there
was one that's hFE 100, one that's around 250, and a couple that's
400-500. I don't know how to explain the ones that are hFE > 100.
What's going on here? Are they really that strong? Are they broken?
Also worth mentioning at hFE 500 the tester gets confused and assumes
the device is a P-JFET.

My guess is that once the tester figures out this is a transistor it
will just test hFE with a DC voltage and a very small current. So
given that is the case, and that a lot of the Germanium transistors
are in the right range, I don't think that the very high hFE
measurement has to be off... but I don't know what might be happening.

My main worry is that those transistors are broken and if I put them
in a reasonable circuit they'll try to output so much power they'll go
up in smoke. Or maybe it's something more subtle, like a lot of noise
being read by the ADC as high output current, or something like that.

How does one explain the weird drift and the super high hFE?