INDEX
Explanations
references to gardens and gardening
New Auto-Interp
Negative Logits
nces
-0.78
esty
-0.75
channelAvailability
-0.68
acial
-0.68
sem
-0.67
imaru
-0.67
ufact
-0.66
phas
-0.65
ilitary
-0.63
ictional
-0.63
POSITIVE LOGITS
hire
0.90
fruit
0.87
garden
0.85
gardens
0.85
hose
0.80
arden
0.77
Gn
0.76
flower
0.75
keeper
0.75
Garden
0.75
Activations Density 0.019%