INDEX
Explanations
phrases indicating repetition or recurrence of events
New Auto-Interp
Negative Logits
repeatedly
-0.18
final
-0.18
constantly
-0.18
finally
-0.17
iders
-0.17
still
-0.16
Scoped
-0.15
twice
-0.14
continually
-0.14
any
-0.14
POSITIVE LOGITS
another
0.23
another
0.23
Another
0.19
addition
0.19
Another
0.19
æ·»
0.19
installment
0.19
Addition
0.17
ovnÄĽ
0.17
additions
0.17
Activations Density 0.116%