INDEX
Explanations
proper names, particularly the name "Bert"
mentions of the name "Bert" and related contexts
New Auto-Interp
Negative Logits
¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯
-0.71
portation
-0.67
ansas
-0.67
ntax
-0.67
othal
-0.66
xual
-0.65
exclusive
-0.63
Effects
-0.63
mainland
-0.61
attendant
-0.60
POSITIVE LOGITS
ucci
0.98
rics
0.95
Bert
0.93
rand
0.93
ardo
0.93
ric
0.90
Seym
0.88
oro
0.88
illon
0.86
cher
0.85
Activations Density 0.021%