INDEX
Explanations
adverbs of admission or qualification
New Auto-Interp
Negative Logits
...
0.65
..."
0.65
Inner
0.64
...";
0.64
؛
0.63
نا
0.62
Configurer
0.62
和
0.61
inner
0.61
หว
0.60
POSITIVE LOGITS
perhaps
1.75
admittedly
1.50
indeed
1.43
perhaps
1.42
maybe
1.33
basically
1.32
actually
1.29
quizá
1.28
talvez
1.28
appunto
1.28
Activations Density 0.020%