INDEX
Explanations
concluding transitional words
New Auto-Interp
Negative Logits
downside
0.32
afield
0.30
ंदे
0.30
قابل
0.30
possibly
0.29
”،
0.29
çoit
0.29
нила
0.28
rollover
0.27
দেশিক
0.27
POSITIVE LOGITS
Therefore
0.61
Essentially
0.58
Unlike
0.54
Think
0.50
Consequently
0.50
When
0.49
Instead
0.49
Поэтому
0.48
Thus
0.48
Specifically
0.48
Activations Density 0.208%