INDEX
Explanations
time and dates
numbers followed by punctuation
New Auto-Interp
Negative Logits
<unused628>
0.42
<unused1917>
0.42
<unused558>
0.40
स्क
0.37
<unused231>
0.37
0
0.36
A
0.36
<unused1728>
0.36
<unused444>
0.36
<unused602>
0.35
POSITIVE LOGITS
en
0.48
at
0.48
е
0.44
about
0.43
digamos
0.41
،
0.39
durch
0.38
ene
0.38
from
0.38
eksi
0.38
Activations Density 3.251%