INDEX
Explanations
references to small or diminutive objects or concepts
New Auto-Interp
Negative Logits
den
-0.49
gra
-0.47
ggen
-0.46
↵↵
-0.46
Gra
-0.46
glVertex
-0.46
(>
-0.44
utama
-0.43
la
-0.42
↵
-0.42
POSITIVE LOGITS
Small
1.34
tiny
1.31
SMALL
1.31
small
1.30
small
1.29
Small
1.27
SMALL
1.24
Tiny
1.24
Tiny
1.17
smallest
1.15
Activations Density 0.417%