INDEX
Explanations
occurrences of the word 'the'
New Auto-Interp
Negative Logits
ſcher
-0.69
iſen
-0.68
iſchen
-0.68
queſto
-0.65
arangay
-0.65
majánló
-0.65
feroit
-0.64
iſten
-0.64
ſſel
-0.62
ghijklmnop
-0.62
POSITIVE LOGITS
about
1.02
ABOUT
0.87
About
0.82
about
0.82
About
0.79
ABOUT
0.71
عن
0.55
sobre
0.53
regarding
0.52
Bout
0.52
Activations Density 0.025%