INDEX
Explanations
terms related to blindness and its effects
New Auto-Interp
Negative Logits
ãĥĨãĥ«
-0.15
Bridges
-0.15
aved
-0.14
iban
-0.14
inker
-0.14
cean
-0.14
coni
-0.14
undry
-0.14
avig
-0.13
ãĤ«ãĥ«
-0.13
POSITIVE LOGITS
fold
0.44
blind
0.31
blind
0.29
spot
0.28
-fold
0.27
Blind
0.27
spots
0.26
ness
0.25
folded
0.25
spot
0.25
Activations Density 0.007%