INDEX
Explanations
phrases related to legal trouble and imprisonment
New Auto-Interp
Negative Logits
rians
-0.73
phies
-0.73
Remastered
-0.70
ãĤ¤ãĥĪ
-0.69
xit
-0.68
Scal
-0.67
cs
-0.66
eu
-0.66
rouse
-0.62
ãĥ¤
-0.62
POSITIVE LOGITS
sentences
1.03
sentence
1.02
sentenced
1.01
sentencing
1.00
jailed
0.99
convicted
0.99
probation
0.94
prison
0.91
manslaughter
0.89
jail
0.89
Activations Density 0.045%