INDEX
Explanations
contrasting or alternative ideas
New Auto-Interp
Negative Logits
আবারো
0.76
सहित
0.75
视
0.75
nữa
0.69
सुम
0.69
weiteren
0.66
полный
0.65
üllen
0.65
িসহ
0.64
kó
0.63
POSITIVE LOGITS
Whereas
3.15
whereas
2.84
Whereas
2.75
whereas
2.62
Meanwhile
2.60
Conversely
2.55
Conversely
2.49
Meanwhile
2.39
conversely
2.32
hingegen
2.29
Activations Density 0.149%