INDEX
Explanations
references to the term "static"
New Auto-Interp
Negative Logits
hoff
-0.87
ceans
-0.83
hof
-0.81
gard
-0.80
vest
-0.80
zees
-0.76
holes
-0.75
ador
-0.74
andals
-0.74
gdala
-0.74
POSITIVE LOGITS
analy
0.85
electricity
0.81
inline
0.72
emission
0.69
iple
0.66
element
0.66
wallpaper
0.65
cling
0.65
animation
0.65
barrier
0.64
Activations Density 0.016%