INDEX
Explanations
proper names
the occurrence of the substring "iz" in various contexts
New Auto-Interp
Negative Logits
quartered
-0.86
hitch
-0.64
horns
-0.63
respawn
-0.62
contrasting
-0.61
whistle
-0.61
heast
-0.59
ridges
-0.59
riott
-0.59
Norn
-0.59
POSITIVE LOGITS
arro
1.41
ombie
1.13
enegger
1.01
hou
1.01
hao
0.97
umi
0.89
akura
0.88
abal
0.87
irtual
0.84
ymes
0.84
Activations Density 0.031%