INDEX
Explanations
proper nouns, specifically names containing the sequence "van"
instances of the word "van" as well as variations of the name "Van" and "Von"
New Auto-Interp
Negative Logits
reps
-0.70
commanding
-0.69
align
-0.67
tamp
-0.67
handlers
-0.66
codes
-0.64
linebackers
-0.62
physicians
-0.62
counselors
-0.62
Fib
-0.62
POSITIVE LOGITS
van
4.44
von
1.99
Van
1.62
vana
1.60
va
1.53
ivan
1.51
van
1.50
vin
1.43
vo
1.35
vas
1.34
Activations Density 0.008%