INDEX
Negative Logits
detections
0.46
মে
0.43
することができる
0.41
ensuring
0.40
OBSERV
0.40
还是很
0.39
warning
0.39
warnings
0.38
erkennen
0.38
konnten
0.38
POSITIVE LOGITS
blindly
0.80
underestimate
0.76
needlessly
0.70
rely
0.68
rush
0.68
knowingly
0.66
섣
0.64
jeopardize
0.62
confuse
0.61
jeopard
0.61
Activations Density 0.038%