INDEX
Explanations
mentions of the names 'Chelsea' and 'CNN' within the text
New Auto-Interp
Negative Logits
fully
-0.20
nes
-0.18
ìĦľëĬĶ
-0.17
ãģª
-0.17
mate
-0.16
bred
-0.16
ally
-0.15
ateur
-0.15
mates
-0.15
liqu
-0.15
POSITIVE LOGITS
yas
0.18
ught
0.16
issance
0.15
ye
0.15
enberg
0.15
iah
0.15
ÄĽt
0.15
ya
0.15
704
0.14
Braz
0.14
Activations Density 0.096%