INDEX
Explanations
mentions of specific people's names
New Auto-Interp
Negative Logits
essentiel
-0.51
savevideobot
-0.47
apropi
-0.47
compañ
-0.46
réc
-0.46
refroidissement
-0.45
syke
-0.45
apparence
-0.44
bandeira
-0.43
ganger
-0.43
POSITIVE LOGITS
(@
0.71
(@
0.54
@
0.51
@
0.47
writes
0.47
@+
0.46
writes
0.46
Writes
0.44
aka
0.43
AKA
0.43
Activations Density 0.404%