INDEX
Explanations
names related to people or places, particularly ones with the string "ry" in them
occurrences of the substring "ry"
New Auto-Interp
Negative Logits
assets
-0.64
insula
-0.61
ajor
-0.59
KNOWN
-0.58
owered
-0.58
validation
-0.57
IR
-0.56
hest
-0.56
hetically
-0.56
parenting
-0.56
POSITIVE LOGITS
stal
1.45
stals
1.29
stall
1.08
pter
1.02
gian
0.87
ng
0.85
croft
0.84
akov
0.82
wall
0.79
ansk
0.77
Activations Density 0.036%