INDEX
Explanations
phrases indicating measurement of distance or time
New Auto-Interp
Negative Logits
gså
-0.34
trap
-0.32
Trap
-0.31
ImageContext
-0.30
trap
-0.29
teacher
-0.28
short
-0.28
short
-0.27
dup
-0.26
tried
-0.26
POSITIVE LOGITS
遠
0.70
distant
0.69
remote
0.67
远
0.66
lontano
0.65
Distant
0.65
jauh
0.65
Remote
0.63
remot
0.63
distant
0.61
Activations Density 0.142%