INDEX
Explanations
references to Muhammad or Mohammed
Mohammed and variations
New Auto-Interp
Negative Logits
Banjar
-0.43
copy
-0.42
Lark
-0.35
[
-0.35
kyard
-0.35
Copy
-0.34
cấp
-0.34
Legacy
-0.34
Kall
-0.34
pin
-0.34
POSITIVE LOGITS
Mohammed
2.27
Mohammed
2.17
Muhammed
1.63
Mohammad
1.58
Mohammad
1.48
Muhammad
1.46
Mohamed
1.42
hammed
1.39
Muhammad
1.33
Mohamed
1.31
Activations Density 0.002%