INDEX
Explanations
references to bottles and their types
New Auto-Interp
Negative Logits
rael
-0.19
646
-0.16
876
-0.16
648
-0.16
upp
-0.15
roe
-0.15
erte
-0.15
enburg
-0.15
esModule
-0.15
è¨Ģ
-0.15
POSITIVE LOGITS
caps
0.21
cap
0.20
caps
0.20
opener
0.19
cap
0.19
-cap
0.19
-neck
0.19
brush
0.18
cork
0.17
Neck
0.17
Activations Density 0.007%