INDEX
Explanations
division
This neuron activates on occurrences of “div,” “divide,” or “division” (and their variants), i.e. mentions of the division operation.
New Auto-Interp
Negative Logits
ennis
-0.07
rawn
-0.07
Mehr
-0.06
далеко
-0.06
机构
-0.06
ткани
-0.06
jclass
-0.06
expansion
-0.06
AHL
-0.06
ثل
-0.06
POSITIVE LOGITS
onDelete
0.07
=msg
0.07
자는
0.07
_Osc
0.06
_settings
0.06
péri
0.06
inorder
0.06
Parking
0.06
(Mock
0.06
≦
0.06
Activations Density 0.013%