INDEX
Explanations
references to milk and dairy products
New Auto-Interp
Negative Logits
ihar
-0.17
sembles
-0.17
295
-0.16
ucher
-0.14
nels
-0.14
eer
-0.14
anie
-0.14
Zuk
-0.14
ctor
-0.14
ì
-0.14
POSITIVE LOGITS
shake
0.42
maid
0.30
maids
0.28
aukee
0.26
weed
0.23
milk
0.22
bone
0.21
shed
0.21
toast
0.20
repl
0.20
Activations Density 0.007%