INDEX
Explanations
prepositions
This neuron specializes in picking out modifier words—particularly adverbs and prepositional markers such as “primarily,” “during,” and “to” that signal emphasis, extent, or temporal context.
New Auto-Interp
Negative Logits
�
-0.07
۱۰
-0.06
'o
-0.06
/task
-0.06
_bins
-0.06
۶
-0.06
_ORDER
-0.06
致
-0.06
One
-0.06
6
-0.06
POSITIVE LOGITS
أغ
0.07
าข
0.06
Lat
0.06
azes
0.06
ΔE
0.06
>n
0.06
Dez
0.06
}}</
0.06
/component
0.06
ROSS
0.06
Activations Density 0.088%