INDEX
Explanations
phrases that indicate a comparison or similarity
New Auto-Interp
Negative Logits
WriteTagHelper
-0.76
DockStyle
-0.58
OGND
-0.54
يتيمه
-0.50
aarrggbb
-0.48
__*/
-0.48
ջ
-0.48
مرئيه
-0.46
+#+
-0.45
Contribution
-0.44
POSITIVE LOGITS
applies
1.68
geldt
1.43
apply
1.41
Applies
1.25
Apply
1.23
apply
1.17
Apply
1.10
applying
1.06
applying
1.03
APPLY
1.03
Activations Density 0.236%