INDEX
Explanations
mentions of time durations related to months or years
New Auto-Interp
Negative Logits
ouz
-0.18
ouch
-0.15
dne
-0.14
cdr
-0.14
oyer
-0.14
ashi
-0.14
prites
-0.14
oya
-0.13
uss
-0.13
p
-0.13
POSITIVE LOGITS
amak
0.16
üre
0.15
ugins
0.15
jak
0.15
inki
0.15
kaar
0.14
ept
0.14
adele
0.14
igg
0.14
juan
0.14
Activations Density 0.045%