INDEX
Explanations
Nothing else, Nothing phone, Anything V
New Auto-Interp
Negative Logits
menos
0.50
weniger
0.44
anteriores
0.43
fewer
0.42
น้อย
0.42
mindre
0.41
moins
0.40
мень
0.39
refleks
0.39
меньше
0.39
POSITIVE LOGITS
enz
0.42
Bou
0.41
Function
0.40
Happened
0.40
hauptsächlich
0.39
hyth
0.39
Supplement
0.39
rist
0.38
gars
0.37
happening
0.37
Activations Density 0.001%