INDEX
Explanations
mentions of the color orange
references to the color orange
New Auto-Interp
Negative Logits
Ö¼
-0.97
arnaev
-0.81
76561
-0.78
fare
-0.76
ebin
-0.75
sonian
-0.75
raltar
-0.75
rets
-0.74
adr
-0.72
rolet
-0.72
POSITIVE LOGITS
peel
1.18
juice
1.13
Blossom
1.04
fruits
0.91
orange
0.87
Peel
0.86
Juice
0.84
berry
0.83
oranges
0.82
fruit
0.81
Activations Density 0.009%