INDEX
Explanations
mentions of the name "Tom" and variations associated with it
New Auto-Interp
Negative Logits
iaux
-0.17
uben
-0.16
unning
-0.16
uary
-0.15
vet
-0.15
weet
-0.15
loi
-0.14
stav
-0.14
licht
-0.14
sut
-0.14
POSITIVE LOGITS
atoes
0.22
asso
0.21
mas
0.21
islav
0.20
ás
0.19
Tom
0.19
tom
0.18
cat
0.18
kins
0.17
ma
0.17
Activations Density 0.032%