INDEX
Explanations
mentions of specific products and their attributes
New Auto-Interp
Negative Logits
lient
-0.16
kich
-0.15
idla
-0.15
оÑĢаз
-0.15
ÑĤаб
-0.15
kili
-0.15
engin
-0.14
idel
-0.14
inator
-0.14
Cannon
-0.14
POSITIVE LOGITS
Poss
0.34
possess
0.34
poss
0.31
poss
0.29
possessing
0.27
possesses
0.25
possession
0.25
Poss
0.24
having
0.24
possessed
0.23
Activations Density 0.025%