INDEX
Explanations
exclamations expressing strong emotions or reactions
exclamatory punctuation and expressions of strong emotions
New Auto-Interp
Negative Logits
atories
-0.75
aldi
-0.74
liest
-0.70
ator
-0.68
reen
-0.68
unker
-0.67
path
-0.66
sten
-0.64
ouf
-0.64
lining
-0.63
POSITIVE LOGITS
!!!!!
1.25
!!!
1.13
!!
1.09
?!
1.05
@#
0.95
??
0.95
:-)
0.93
!/
0.93
!!!!
0.90
?:
0.90
Activations Density 0.015%