INDEX
Explanations
sweet foods or desserts
references to desserts and related food items
New Auto-Interp
Negative Logits
seed
-0.73
olds
-0.68
atory
-0.67
Frames
-0.67
¾
-0.65
ndra
-0.62
head
-0.62
CLE
-0.62
Anim
-0.61
reen
-0.61
POSITIVE LOGITS
=>
0.83
terday
0.78
avorite
0.75
xual
0.74
dessert
0.73
Mons
0.72
mud
0.72
essert
0.70
++++
0.70
emonium
0.69
Activations Density 0.016%