INDEX
Explanations
instances of the word "hug" and related variations
instances of the word "hug."
New Auto-Interp
Negative Logits
piracy
-0.70
ourses
-0.64
Punk
-0.61
Procedure
-0.61
Edison
-0.60
umer
-0.59
secondary
-0.58
piracy
-0.57
Proced
-0.57
DoS
-0.57
POSITIVE LOGITS
hug
1.06
hugs
1.02
eness
1.01
Hug
0.97
eson
0.91
wrap
0.90
hugged
0.90
goodbye
0.88
glers
0.87
gers
0.86
Activations Density 0.008%