INDEX
Negative Logits
ing
-0.52
es
-0.52
<eos>
-0.51
ker
-0.49
g
-0.48
ers
-0.45
er
-0.45
ent
-0.45
vern
-0.45
ently
-0.45
POSITIVE LOGITS
Majefty
1.17
^(@)
1.08
Theſe
1.05
Anſ
1.04
itſelf
1.01
Efq
1.00
myſelf
0.98
ſelf
0.96
doubtnut
0.95
Houſe
0.93
Activations Density 0.520%