INDEX
Explanations
phrases expressing disapproval or condemnation
expressions of condemnation or unacceptability
New Auto-Interp
Negative Logits
stash
-0.70
climbs
-0.68
atonin
-0.67
exits
-0.66
mysteriously
-0.66
selected
-0.65
dominate
-0.65
Likes
-0.63
slips
-0.63
transitions
-0.63
POSITIVE LOGITS
hypocrisy
1.07
hypocritical
0.91
disgrace
0.90
disrespect
0.86
unacceptable
0.85
arrogance
0.85
unethical
0.84
despicable
0.83
disrespectful
0.82
hypocr
0.82
Activations Density 0.632%