INDEX
Explanations
references to being in front of various audiences or groups
New Auto-Interp
Negative Logits
era
-0.15
ogl
-0.15
Front
-0.15
enticate
-0.15
erals
-0.15
ancia
-0.15
.front
-0.14
loff
-0.14
_front
-0.14
ix
-0.14
POSITIVE LOGITS
cameras
0.25
camera
0.25
-camera
0.23
Cameras
0.20
Camera
0.20
quam
0.18
Camera
0.18
eyes
0.18
closed
0.18
camera
0.18
Activations Density 0.036%