INDEX
Explanations
visual elements related to urban landscapes and architectural features
New Auto-Interp
Negative Logits
Synopsis
-0.15
yoksa
-0.13
ležit
-0.13
γι
-0.13
agraph
-0.13
instead
-0.13
.Android
-0.12
illi
-0.12
569
-0.12
lector
-0.12
POSITIVE LOGITS
viewed
0.28
taken
0.25
seen
0.24
detail
0.23
detail
0.22
taken
0.21
view
0.21
showing
0.20
looking
0.20
©
0.20
Activations Density 0.196%