INDEX
Explanations
positive adjectives associated with warmth, friendliness, and hospitality
the word "warm" and its variations or context
New Auto-Interp
Negative Logits
argon
-0.81
IMAGES
-0.79
ONE
-0.70
DEV
-0.70
doms
-0.70
aurus
-0.69
issors
-0.69
UNCH
-0.69
gur
-0.68
IBLE
-0.67
POSITIVE LOGITS
achine
1.22
fuzz
1.02
warm
1.01
est
1.01
fuzzy
0.91
warmth
0.91
hearted
0.88
ening
0.88
welcome
0.88
blanket
0.85
Activations Density 0.016%