INDEX
Explanations
phrases indicating ongoing actions or statuses
New Auto-Interp
Negative Logits
ester
-0.16
arc
-0.15
ioso
-0.15
ارس
-0.14
unately
-0.14
pest
-0.14
ört
-0.14
neh
-0.14
beit
-0.14
yster
-0.14
POSITIVE LOGITS
finally
0.24
back
0.21
again
0.20
Finally
0.19
ready
0.18
among
0.18
Finally
0.17
final
0.17
today
0.16
now
0.16
Activations Density 0.111%