INDEX
Explanations
characters or symbols that are visually represented with different fonts or styles
instances of the character 'ĩ'
New Auto-Interp
Negative Logits
ters
-0.88
lisher
-0.76
rette
-0.73
kernels
-0.69
faced
-0.69
streak
-0.68
verts
-0.68
geons
-0.68
fet
-0.67
figured
-0.67
POSITIVE LOGITS
ا
0.92
ãģĨ
0.83
atic
0.79
ÑĮ
0.79
alian
0.78
andise
0.78
Į
0.78
į
0.77
åŃ
0.77
е
0.76
Activations Density 0.017%