INDEX
Explanations
Names and locations
This neuron activates on personal names (proper names of individuals) in the text.
New Auto-Interp
Negative Logits
opro
-0.07
_SHARE
-0.06
situación
-0.06
オ
-0.06
SPE
-0.06
iframe
-0.06
Seriously
-0.06
abella
-0.06
ší
-0.06
plaza
-0.06
POSITIVE LOGITS
akıl
0.06
剑
0.06
CHANGE
0.06
Rodgers
0.06
ponent
0.06
MOVED
0.06
.dense
0.06
ขนาด
0.06
lux
0.06
mohli
0.06
Activations Density 0.187%