INDEX
Explanations
phrases or words related to flowers
instances and mentions of flowers
New Auto-Interp
Negative Logits
ugu
-0.73
eport
-0.73
nesota
-0.70
unden
-0.69
*/(
-0.69
rome
-0.68
gregation
-0.67
resy
-0.67
ahon
-0.66
urai
-0.66
POSITIVE LOGITS
bloom
1.02
arranging
0.96
flowers
0.94
blo
0.92
buds
0.92
Gardens
0.90
gardens
0.89
pots
0.89
flower
0.88
fruit
0.87
Activations Density 0.036%