INDEX
Explanations
proper nouns, specifically names
references to the name "Ivan" and related entities
New Auto-Interp
Negative Logits
++++
-0.82
ULTS
-0.66
Polo
-0.63
IENT
-0.63
Scient
-0.63
wait
-0.62
Rept
-0.61
taboola
-0.61
+++
-0.60
Fight
-0.60
POSITIVE LOGITS
ovie
0.98
ovich
0.92
imov
0.84
owsky
0.81
ovic
0.80
andum
0.79
stadt
0.79
obi
0.76
achev
0.76
nai
0.75
Activations Density 0.092%