INDEX
Explanations
quotation marks followed by temporal words
New Auto-Interp
Negative Logits
dagens
-1.46
tonight
-1.36
today
-1.34
if
-1.32
Tonight
-1.31
currently
-1.26
目前
-1.23
its
-1.23
yesterday
-1.22
hopefully
-1.20
POSITIVE LOGITS
afterwards
1.41
after
1.38
después
1.34
當時
1.31
then
1.27
recuerdo
1.26
afterward
1.24
setelah
1.23
وكان
1.21
当時の
1.20
Activations Density 0.024%