INDEX
Explanations
design and technical tools
This neuron detects capitalized technical acronyms and initialisms (especially security, medical imaging, AI, and enterprise software terms).
New Auto-Interp
Negative Logits
0.63
I
0.52
۔
0.46
。
0.45
ului
0.40
।
0.40
។
0.39
asia
0.39
kaan
0.39
decentral
0.38
POSITIVE LOGITS
на
1.02
on
0.74
ו
0.73
و
0.72
to
0.69
ي
0.67
ح
0.65
ান
0.63
2
0.62
ف
0.62
Activations Density 0.909%