INDEX
Explanations
Being indoors
This neuron detects terms and phrases associated with extreme social withdrawal or reclusive lifestyles (e.g. “hikikomori,” “shut-in,” “NEET”).
New Auto-Interp
Negative Logits
接
-0.07
.Cdecl
-0.06
.students
-0.06
Sanders
-0.06
Tank
-0.06
_BC
-0.06
552
-0.06
ジュ
-0.06
messenger
-0.06
Europe
-0.06
POSITIVE LOGITS
ζει
0.07
sluts
0.06
/ip
0.06
-- ↵
0.06
}}↵
0.06
affen
0.06
değişiklik
0.06
крок
0.06
ültür
0.06
liquidity
0.06
Activations Density 0.042%