INDEX
Explanations
instances of the name "Harry" in various contexts
New Auto-Interp
Negative Logits
yar
-0.16
ory
-0.15
ese
-0.15
aci
-0.15
ermal
-0.15
GB
-0.15
gb
-0.15
éł
-0.14
ferr
-0.14
ittle
-0.14
POSITIVE LOGITS
hausen
0.30
Potter
0.25
pot
0.20
Pot
0.18
.nlm
0.18
POT
0.17
_pot
0.16
Styles
0.16
uki
0.16
ette
0.15
Activations Density 0.004%