INDEX
Explanations
proper nouns related to people's names
the repeated mention of the name "Mann."
New Auto-Interp
Negative Logits
inyl
-0.77
Emir
-0.71
Absent
-0.69
>>>>>>>>
-0.63
inition
-0.62
inement
-0.61
CLS
-0.60
Ferry
-0.60
FISA
-0.58
dfx
-0.58
POSITIVE LOGITS
olini
1.02
ageddon
0.97
ibal
0.93
ucci
0.91
fred
0.90
ifest
0.90
emonic
0.90
ificent
0.90
heim
0.89
ella
0.89
Activations Density 0.037%