INDEX
Explanations
no question/certainty
The neuron detects occurrences of the word “question” (especially in phrases like “no question” or “never a question”) indicating rhetorical certainty.
New Auto-Interp
Negative Logits
кул
-0.06
With
-0.06
mart
-0.06
wang
-0.06
racked
-0.06
novembre
-0.06
海
-0.06
anders
-0.06
attorney
-0.06
Individuals
-0.06
POSITIVE LOGITS
associations
0.07
Lib
0.06
ativo
0.06
mor
0.06
Leo
0.06
Sprite
0.06
่าง
0.06
akeup
0.06
fst
0.06
manners
0.06
Activations Density 0.085%