INDEX
Explanations
phrases that indicate time or duration
New Auto-Interp
Negative Logits
lectron
-0.15
avit
-0.15
enton
-0.14
.lazy
-0.14
ỳ
-0.14
668
-0.14
iska
-0.13
šti
-0.13
alon
-0.13
acid
-0.13
POSITIVE LOGITS
Intr
0.16
VML
0.16
eso
0.15
Arth
0.14
Åĵ
0.14
Tort
0.14
mere
0.14
udas
0.14
tetas
0.14
edx
0.14
Activations Density 0.041%