INDEX
Explanations
This neuron activates on business- or organization-related nouns—especially terms like “Management,” “Services,” or company/person names.
New Auto-Interp
Negative Logits
619
-0.07
Gio
-0.06
>tag
-0.06
µ
-0.06
aseña
-0.06
Glover
-0.06
flour
-0.06
Glory
-0.06
IDX
-0.06
Fo
-0.06
POSITIVE LOGITS
and
0.12
And
0.10
and
0.10
&
0.09
And
0.09
AND
0.09
άν
0.08
いて
0.08
-and
0.08
и
0.08
Activations Density 0.080%