INDEX
Negative Logits
moral
0.75
main
0.74
enlist
0.74
kids
0.71
lowered
0.70
زیاد
0.70
Murdock
0.69
dare
0.69
обяза
0.68
coward
0.68
POSITIVE LOGITS
もちろん
0.86
bicovariant
0.79
tileSize
0.77
িংস
0.74
மாற்ற
0.73
alternatively
0.73
ítica
0.72
('<0.71
涜
0.71
bandSize
0.71
Activations Density 0.006%