INDEX
Explanations
the name "Martin" across various contexts
New Auto-Interp
Negative Logits
osu
-0.17
eren
-0.16
cats
-0.16
748
-0.15
kettle
-0.15
IXEL
-0.15
anning
-0.15
ằng
-0.15
eker
-0.14
ilton
-0.14
POSITIVE LOGITS
Luther
0.30
ique
0.28
elli
0.24
IQUE
0.21
borough
0.21
iqu
0.19
ho
0.18
aira
0.18
/software
0.17
iano
0.16
Activations Density 0.010%