INDEX
Explanations
greetings and expressions of goodwill
New Auto-Interp
Negative Logits
latego
-0.59
kasarigan
-0.58
AsUp
-0.57
-0.56
ThroughAttribute
-0.56
esez
-0.56
ilmente
-0.55
indd
-0.55
icrous
-0.54
Plus
-0.53
POSITIVE LOGITS
welcome
1.41
Welcome
1.31
Welcome
1.25
Hello
1.21
Hi
1.19
welcome
1.19
hello
1.17
Hello
1.15
Hi
1.15
hi
1.10
Activations Density 0.313%