INDEX
Explanations
gently tracing or fixing gaze
New Auto-Interp
Negative Logits
screams
0.95
screaming
0.94
scream
0.89
screamed
0.88
shouting
0.82
咆
0.82
scre
0.80
vomiting
0.77
wept
0.76
monolithic
0.75
POSITIVE LOGITS
involuntarily
0.90
softly
0.87
playfully
0.85
lightly
0.85
slightly
0.82
graze
0.82
tracing
0.79
Trace
0.77
légèrement
0.77
gently
0.76
Activations Density 0.399%