INDEX
Explanations
expressions related to visual perception and spatial understanding
New Auto-Interp
Negative Logits
äch
-0.15
linger
-0.15
orno
-0.14
ronics
-0.14
iales
-0.14
abb
-0.14
emer
-0.13
iente
-0.13
heat
-0.13
.ibatis
-0.13
POSITIVE LOGITS
similarly
0.29
Similarly
0.29
Similarly
0.28
Likewise
0.24
Lik
0.24
similar
0.23
likewise
0.21
same
0.20
similar
0.19
analog
0.18
Activations Density 0.162%