INDEX
Explanations
phrases related to intense emotions, particularly crying and distress
instances of crying or expressions of sadness
New Auto-Interp
Negative Logits
awaru
-0.79
ammy
-0.73
pend
-0.70
ositories
-0.70
iciency
-0.70
insula
-0.69
efficients
-0.67
icient
-0.67
velength
-0.66
nomine
-0.65
POSITIVE LOGITS
stals
1.04
baby
0.95
cry
0.87
cried
0.85
hovah
0.83
ogen
0.82
sis
0.82
aloud
0.81
pter
0.80
Cry
0.80
Activations Density 0.012%