INDEX
Explanations
punctuation marks and their frequency or patterns in text
New Auto-Interp
Negative Logits
ETCH
-0.19
etch
-0.17
umb
-0.15
umn
-0.14
Rencontre
-0.14
Cres
-0.14
Flo
-0.14
äm
-0.14
endor
-0.14
Crunch
-0.14
POSITIVE LOGITS
yx
0.19
otron
0.16
hle
0.14
zew
0.14
eneg
0.14
Trojan
0.14
cobra
0.14
eÅŁ
0.14
side
0.14
aac
0.14
Activations Density 0.120%