INDEX
Explanations
the presence of various formatting tags or structural elements in the text
New Auto-Interp
Negative Logits
Jeografia
-0.52
RoomId
-0.52
colgante
-0.50
puerta
-0.47
enfans
-0.47
bénéfices
-0.47
Valentín
-0.46
Legături
-0.46
ensaft
-0.46
SBATCH
-0.46
POSITIVE LOGITS
<h2>
0.93
<h3>
0.86
<h1>
0.84
<h4>
0.84
<h5>
0.72
<h6>
0.57
/**
0.55
<bos>
0.51
__*/
0.47
Перейти
0.45
Activations Density 0.047%