INDEX
Explanations
references to the name "Bob."
New Auto-Interp
Negative Logits
cona
-0.16
åĿĽ
-0.15
iffe
-0.14
memberOf
-0.14
upert
-0.14
vapor
-0.14
izmet
-0.14
uations
-0.14
å¢
-0.14
/if
-0.14
POSITIVE LOGITS
bi
0.34
bie
0.34
bing
0.32
ble
0.30
cat
0.29
Dylan
0.28
bies
0.26
cats
0.26
bit
0.25
bin
0.25
Activations Density 0.006%