INDEX
Explanations
descriptions related to physical objects and furniture in a home
New Auto-Interp
Negative Logits
utenberg
-0.72
iyah
-0.70
uality
-0.67
abad
-0.67
hypers
-0.63
arians
-0.62
iferation
-0.61
iant
-0.61
ient
-0.59
ename
-0.58
POSITIVE LOGITS
cush
1.08
cushion
0.96
washer
0.82
sofa
0.81
tops
0.81
potato
0.81
mattress
0.79
sur
0.78
chairs
0.77
couch
0.76
Activations Density 8.311%