INDEX
Explanations
references to time and context in a narrative
New Auto-Interp
Negative Logits
رÙħز
-0.15
utura
-0.15
iveau
-0.14
laps
-0.14
acias
-0.14
marque
-0.14
achsen
-0.14
rrha
-0.14
baugh
-0.14
ewire
-0.14
POSITIVE LOGITS
premiere
0.17
bombard
0.16
ξε
0.16
tors
0.16
session
0.16
ennes
0.16
nth
0.15
walk
0.15
Premiere
0.15
safari
0.15
Activations Density 0.229%