INDEX
Explanations
references to fingers and related concepts
New Auto-Interp
Negative Logits
WARD
-0.17
greg
-0.17
urr
-0.16
tl
-0.16
Ñħов
-0.16
cou
-0.15
AMPL
-0.15
roscope
-0.14
anium
-0.14
uxe
-0.14
POSITIVE LOGITS
/thumb
0.27
prints
0.24
finger
0.24
Finger
0.23
lings
0.21
tip
0.21
finger
0.21
tip
0.19
-tip
0.19
Lakes
0.18
Activations Density 0.014%