INDEX
Explanations
references to traffic safety and the effects of messaging on driver behavior
New Auto-Interp
Negative Logits
anim
-0.16
oga
-0.15
lt
-0.15
iÄĻ
-0.14
воÑĢ
-0.14
éĻ
-0.14
hana
-0.14
ichel
-0.14
uke
-0.14
ijd
-0.14
POSITIVE LOGITS
distracted
0.35
distraction
0.28
texting
0.27
distractions
0.27
text
0.25
distract
0.25
Text
0.24
drunk
0.23
text
0.23
texts
0.22
Activations Density 0.038%