INDEX
Explanations
references to duality and comparisons between two entities
New Auto-Interp
Negative Logits
all
-0.19
242
-0.18
onda
-0.17
iera
-0.16
relationships
-0.16
fours
-0.16
various
-0.15
relationships
-0.15
airs
-0.14
friendships
-0.14
POSITIVE LOGITS
-two
0.32
two
0.29
两个
0.28
respectively
0.26
åĪĨåĪ«
0.26
two
0.26
Two
0.25
beide
0.25
obou
0.25
beiden
0.24
Activations Density 0.423%