INDEX
Explanations
HTML color codes and attributes related to table formatting
New Auto-Interp
Negative Logits
-black
-0.17
geb
-0.17
enger
-0.16
é»ij
-0.15
black
-0.15
pitch
-0.15
/black
-0.15
BCM
-0.14
801
-0.14
berries
-0.14
POSITIVE LOGITS
White
0.21
White
0.20
Whites
0.20
WHITE
0.18
white
0.17
whites
0.16
whiteColor
0.16
white
0.16
.White
0.16
arrass
0.15
Activations Density 0.008%