INDEX
Explanations
The neuron flags first‐person expressions of affection, desire, or intent (e.g. “I want,” “I love,” “I can’t wait”) in romantic or caring dialogue.
New Auto-Interp
Negative Logits
manera
-0.07
olmaktadır
-0.07
kendine
-0.07
\Annotation
-0.06
weep
-0.06
vekili
-0.06
Ky
-0.06
sequelize
-0.06
encrypt
-0.06
stringWithFormat
-0.06
POSITIVE LOGITS
dia
0.08
ads
0.07
чої
0.07
pps
0.06
ㅎ
0.06
Systems
0.06
_CITY
0.06
_FOLDER
0.06
мін
0.06
.son
0.06
Activations Density 0.031%