INDEX
Explanations
references to Islamic terms and figures
New Auto-Interp
Negative Logits
BM
-0.22
LM
-0.22
WM
-0.21
McM
-0.21
TM
-0.20
CIM
-0.20
Clem
-0.19
McMahon
-0.19
LM
-0.18
STM
-0.18
POSITIVE LOGITS
am
0.70
ам
0.56
ams
0.54
amam
0.50
amd
0.48
ama
0.48
amt
0.48
amm
0.48
aml
0.46
ami
0.46
Activations Density 0.108%