INDEX
Explanations
conjunctions and their associated expressions
New Auto-Interp
Negative Logits
fathers
-0.20
fats
-0.20
-0.19
feared
-0.19
fears
-0.19
-0.19
France
-0.18
/files
-0.17
français
-0.17
fearful
-0.17
POSITIVE LOGITS
unf
0.21
endforeach
0.18
Mother
0.17
fo
0.16
Mother
0.16
0.16
FO
0.15
Ĥæķ°
0.15
FR
0.15
mother
0.15
Activations Density 0.166%