INDEX
Explanations
specific first names like "Marcel", "Hendrick", and "Blanc"
references to specific individuals or names, particularly those related to Marcel
New Auto-Interp
Negative Logits
soType
-0.78
cv
-0.67
bill
-0.66
¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯
-0.66
harbor
-0.66
udder
-0.65
bags
-0.64
Jaw
-0.64
velt
-0.63
cean
-0.62
POSITIVE LOGITS
Marcel
0.86
Blanc
0.75
atile
0.73
Dek
0.73
imir
0.73
onnaissance
0.73
MAP
0.70
halluc
0.69
Mond
0.67
confir
0.67
Activations Density 0.017%