INDEX
Explanations
Dante's Inferno and Divine Comedy
New Auto-Interp
Negative Logits
disconnected
0.41
isak
0.40
жением
0.39
ijos
0.39
rijk
0.39
ஷ
0.38
किशन
0.38
unaffected
0.37
outstanding
0.37
zość
0.37
POSITIVE LOGITS
Dante
1.34
Inferno
0.95
Infer
0.85
infer
0.78
Dant
0.75
Infer
0.67
Purg
0.64
Beatrice
0.62
infer
0.57
sinners
0.56
Activations Density 0.008%