INDEX
Explanations
punctuation marks as separators
New Auto-Interp
Negative Logits
identified
0.41
identified
0.40
gist
0.40
пок
0.39
شک
0.38
tay
0.37
anillos
0.37
Decree
0.35
ടുത്തു
0.35
unwa
0.34
POSITIVE LOGITS
sign
0.61
marks
0.54
знак
0.54
notation
0.53
マーク
0.48
segno
0.48
(')0.48
mark
0.47
işaret
0.47
符
0.47
Activations Density 0.058%