INDEX
Explanations
phrases related to laziness
instances of the word "sleazy."
New Auto-Interp
Negative Logits
interrupted
-0.84
eous
-0.74
cake
-0.74
ipolar
-0.72
rition
-0.70
uates
-0.67
iates
-0.67
alone
-0.66
oulos
-0.66
iate
-0.66
POSITIVE LOGITS
sle
1.13
azy
0.99
igh
0.85
uth
0.81
ight
0.80
asure
0.79
pload
0.78
eker
0.75
Sle
0.73
iness
0.73
Activations Density 0.004%