INDEX
Explanations
mentions of the name "Andrew" in various contexts
New Auto-Interp
Negative Logits
eland
-0.19
ular
-0.18
uing
-0.15
برد
-0.15
innamon
-0.15
BERT
-0.15
jezd
-0.15
ris
-0.14
imo
-0.14
ernen
-0.14
POSITIVE LOGITS
thal
0.16
afür
0.15
son
0.15
edor
0.14
ship
0.14
anken
0.14
538
0.14
-desc
0.13
Globals
0.13
matic
0.13
Activations Density 0.008%