INDEX
Explanations
intensifiers and superlatives
New Auto-Interp
Negative Logits
that
0.82
the
0.64
this
0.63
it
0.62
which
0.62
they
0.60
when
0.58
That
0.58
that
0.57
those
0.55
POSITIVE LOGITS
extrêmement
0.65
매우
0.61
estremamente
0.60
非常に
0.57
খুবই
0.57
extremamente
0.56
انتہائی
0.53
très
0.52
duże
0.52
excellent
0.51
Activations Density 0.071%