INDEX
Explanations
the preposition "to" as it relates to various actions and intentions
New Auto-Interp
Negative Logits
edi
-0.17
edin
-0.16
oward
-0.15
ede
-0.15
zbo
-0.14
kara
-0.14
alt
-0.14
wing
-0.14
obi
-0.14
zt
-0.13
POSITIVE LOGITS
bear
0.21
ument
0.16
olis
0.16
fruition
0.16
bear
0.15
earer
0.15
çĨĬ
0.15
Bear
0.15
uluk
0.15
_dot
0.14
Activations Density 0.024%