INDEX
Explanations
people and crowds across languages
New Auto-Interp
Negative Logits
Unfortunately
0.28
which
0.27
and
0.27
$
0.26
,
0.26
ionic
0.26
'
0.26
theoretical
0.25
Which
0.25
theorems
0.25
POSITIVE LOGITS
人们
0.34
ಜನರು
0.33
partecipanti
0.33
ప్రజ
0.32
الناس
0.31
люди
0.31
జరిగ
0.30
கூட்டம்
0.30
परिजन
0.30
నిర్ణ
0.30
Activations Density 0.000%