INDEX
Explanations
legal and investigative language used in police reports
New Auto-Interp
Negative Logits
ë§ŀ
-0.07
vant
-0.06
بÙĪØ§Ø³Ø·Ø©
-0.06
/INFO
-0.06
kenin
-0.06
åĬ©
-0.06
ichtig
-0.06
assin
-0.06
ffi
-0.06
خبر
-0.06
POSITIVE LOGITS
instances
0.08
found
0.07
manipulation
0.07
records
0.07
material
0.07
finds
0.07
concerns
0.07
evidence
0.07
manip
0.07
åıijçݰ
0.07
Activations Density 0.039%