INDEX
Explanations
terms related to weapons and military technology
New Auto-Interp
Negative Logits
uta
-0.15
é¹
-0.15
incinn
-0.15
836
-0.15
rette
-0.14
.Experimental
-0.14
oga
-0.14
åĽ
-0.14
otropic
-0.14
osing
-0.14
POSITIVE LOGITS
mith
0.21
adal
0.16
etu
0.16
398
0.15
linger
0.14
667
0.14
Fon
0.14
ovice
0.14
adding
0.13
Against
0.13
Activations Density 0.020%