INDEX
Explanations
dating profiles
The neuron detects first-person descriptions of personal traits, interests, and relationship desires in dating-style profile text.
New Auto-Interp
Negative Logits
_MethodInfo
-0.07
einmal
-0.06
";↵↵
-0.06
unanswered
-0.06
doğal
-0.06
announcement
-0.06
말을
-0.06
らず
-0.06
izard
-0.06
.quest
-0.06
POSITIVE LOGITS
↵↵
0.06
τύ
0.06
वह
0.06
成本
0.06
خاص
0.06
vertical
0.06
Jan
0.06
برای
0.06
Alex
0.06
sill
0.06
Activations Density 0.031%