INDEX
Explanations
This neuron detects mentions of the Jainism religion (tokens like “Jain” and “ism”).
New Auto-Interp
Negative Logits
лечение
-0.07
머
-0.07
.Publish
-0.06
skyt
-0.06
attributes
-0.06
죽
-0.06
длин
-0.06
() ↵ ↵ ↵
-0.06
violated
-0.06
�
-0.06
POSITIVE LOGITS
cmpeq
0.07
Raymond
0.06
Lotus
0.06
brand
0.06
ham
0.06
аков
0.06
iteDatabase
0.06
、三
0.06
+B
0.06
Hose
0.06
Activations Density 0.002%