INDEX
Explanations
phrases indicating repetition or recurrence
New Auto-Interp
Negative Logits
tein
-0.72
allery
-0.69
uckland
-0.69
inth
-0.69
pload
-0.68
hent
-0.68
etheus
-0.67
mosp
-0.67
dor
-0.64
Moder
-0.64
POSITIVE LOGITS
throughout
1.06
thereafter
0.84
across
0.80
whenever
0.72
ciating
0.71
during
0.71
until
0.71
emanating
0.70
imaginable
0.70
occasions
0.70
Activations Density 0.039%