INDEX
Explanations
transformed words such as abbreviations or hyphenated terms
references to science fiction and fantasy genres
New Auto-Interp
Negative Logits
imentary
-0.81
irements
-0.70
mast
-0.70
enlarged
-0.66
redistributed
-0.65
contrace
-0.62
votes
-0.61
Preferences
-0.61
Comments
-0.61
nearest
-0.61
POSITIVE LOGITS
villains
0.98
ampunk
0.94
trope
0.93
Manga
0.92
acters
0.90
manga
0.90
adaptation
0.90
advertising
0.90
superheroes
0.90
villain
0.89
Activations Density 0.159%