INDEX
Explanations
references to materials and their properties
New Auto-Interp
Negative Logits
åij¢
-0.16
zwar
-0.16
enÃŃ
-0.14
/*#__
-0.14
ooth
-0.14
utin
-0.14
IDI
-0.14
riot
-0.13
ving
-0.13
acci
-0.13
POSITIVE LOGITS
otherwise
1.86
Otherwise
1.60
otherwise
1.59
Otherwise
1.47
OTHERWISE
1.38
else
1.11
sonst
1.02
åIJ¦
0.99
else
0.92
else
0.87
Activations Density 0.548%