INDEX
Explanations
characteristics related to furniture and product reviews
New Auto-Interp
Negative Logits
compass
-0.19
Compass
-0.17
ÑĨин
-0.16
Russo
-0.15
touching
-0.15
ç½²
-0.14
touch
-0.14
utton
-0.14
æĻ´
-0.14
stras
-0.14
POSITIVE LOGITS
cov
0.34
cou
0.31
cov
0.30
Cov
0.29
_CO
0.29
_cov
0.29
cob
0.27
Cob
0.27
cob
0.27
cover
0.24
Activations Density 0.045%