INDEX
Explanations
words with special characters, likely associated with emphasis or quotes
instances of the character "âĢ" and its variations
New Auto-Interp
Negative Logits
Marble
-0.73
typew
-0.68
Wad
-0.67
grips
-0.66
Reprodu
-0.64
division
-0.64
range
-0.63
Doodle
-0.63
guiActiveUnfocused
-0.63
raft
-0.63
POSITIVE LOGITS
ł
1.09
against
1.06
ij
1.01
certain
0.99
sure
0.99
IJ
0.98
¹
0.98
Ĵ
0.95
ª
0.95
ı
0.93
Activations Density 0.222%