INDEX
Explanations
references to quantitative data related to membership or inventory in various contexts
New Auto-Interp
Negative Logits
illions
-0.16
anders
-0.16
coni
-0.15
ahlen
-0.14
/trunk
-0.14
wherever
-0.14
duo
-0.13
oxide
-0.13
/etc
-0.13
nie
-0.13
POSITIVE LOGITS
altogether
0.35
ãģĨãģ¡
0.27
total
0.24
spread
0.24
overall
0.21
plus
0.21
alto
0.20
total
0.19
spread
0.19
overall
0.19
Activations Density 0.243%