INDEX
Explanations
words related to furniture, specifically couches and sofas
references to couches and sofas
New Auto-Interp
Negative Logits
iod
-0.82
arts
-0.74
Selected
-0.72
Thomson
-0.70
atically
-0.68
ith
-0.68
Ga
-0.67
tical
-0.67
etsk
-0.65
atic
-0.63
POSITIVE LOGITS
couch
3.82
sofa
2.71
Couch
2.17
mattress
1.48
porch
1.39
rug
1.32
cush
1.29
treadmill
1.27
carpet
1.25
recl
1.25
Activations Density 0.011%