INDEX
Explanations
words related to a specific name
references to a specific individual, Judge Persky
New Auto-Interp
Negative Logits
AKING
-0.84
yrinth
-0.83
nant
-0.78
ctors
-0.75
Benghazi
-0.74
eals
-0.72
iannopoulos
-0.71
ablo
-0.69
LESS
-0.69
RAFT
-0.69
POSITIVE LOGITS
Pers
1.11
istent
0.97
istence
0.96
merga
0.88
Pers
0.87
defic
0.86
challeng
0.84
perspect
0.82
mosqu
0.81
pect
0.80
Activations Density 0.012%