INDEX
Explanations
the word "was" across various contexts in the text
there was no/a/little/ample
New Auto-Interp
Negative Logits
is
-0.85
Is
-0.59
is
-0.59
Is
-0.57
are
-0.51
está
-0.49
является
-0.48
trabaja
-0.47
està
-0.47
的是
-0.46
POSITIVE LOGITS
bootstrapcdn
0.59
Rüyada
0.56
noqa
0.54
epam
0.52
っけ
0.52
Chwiliwch
0.51
FormData
0.49
outube
0.49
newData
0.49
0.47
Activations Density 0.015%