INDEX
Explanations
"be" followed by specific descriptors
New Auto-Interp
Negative Logits
séparation
0.32
സാധ
0.31
έχει
0.31
интересу
0.31
executionContext
0.30
往往
0.30
intended
0.29
ارض
0.29
सहारे
0.29
چھوٹے
0.29
POSITIVE LOGITS
careful
0.61
vigilant
0.61
mindful
0.60
proactive
0.60
cautious
0.54
friend
0.53
able
0.53
considerate
0.52
cheeky
0.49
attentive
0.49
Activations Density 0.051%