INDEX
Explanations
the conjunction "or" in various contexts
New Auto-Interp
Negative Logits
ldr
-0.16
/goto
-0.15
odes
-0.15
utom
-0.15
γι
-0.15
istr
-0.15
lessly
-0.14
лод
-0.14
unsch
-0.14
esser
-0.14
POSITIVE LOGITS
merely
0.18
onto
0.17
/if
0.17
just
0.16
áo
0.14
otherwise
0.14
Ķ
0.14
obox
0.14
nan
0.14
ãģ©ãģĨ
0.14
Activations Density 0.044%