INDEX
Explanations
frequent conjunctions and subjects in sentences
New Auto-Interp
Negative Logits
eton
-0.17
thon
-0.17
eam
-0.16
опол
-0.16
rzy
-0.15
ae
-0.15
eacher
-0.15
imd
-0.14
agate
-0.14
apol
-0.14
POSITIVE LOGITS
å²
0.14
CCC
0.14
zig
0.14
unci
0.14
obb
0.14
setId
0.13
ãĥªãĥ¼
0.13
lâu
0.13
Peb
0.13
ĩnh
0.13
Activations Density 0.110%