INDEX
Explanations
mentions of a specific name or family name "Ker" with varying activation values depending on the context
mentions of the name "Kerley."
New Auto-Interp
Negative Logits
hips
-0.73
defe
-0.72
ãĥł
-0.67
ol
-0.64
legraph
-0.64
OIL
-0.64
ians
-0.64
ERY
-0.64
EMENT
-0.64
Ń·
-0.61
POSITIVE LOGITS
mit
1.10
bledon
0.85
ble
0.84
sten
0.82
isine
0.81
lik
0.79
locked
0.76
rolet
0.76
locks
0.74
bled
0.73
Activations Density 0.061%