INDEX
Explanations
references to the word "this."
New Auto-Interp
Negative Logits
æĹ©
-0.19
early
-0.19
recently
-0.17
early
-0.17
Early
-0.16
liest
-0.15
lately
-0.15
Early
-0.15
recent
-0.15
erville
-0.15
POSITIVE LOGITS
month
0.21
year
0.18
month
0.17
decade
0.17
obre
0.16
year
0.15
week
0.15
same
0.15
.year
0.15
Qu
0.14
Activations Density 0.017%