INDEX
Explanations
the word "But" to indicate a contrasting idea or situation
phrases that begin with "But."
New Auto-Interp
Negative Logits
comm
-0.68
complete
-0.64
classified
-0.62
2018
-0.61
due
-0.61
approximately
-0.60
unknown
-0.60
generally
-0.60
inst
-0.59
completed
-0.59
POSITIVE LOGITS
But
2.90
However
2.22
Yet
2.14
BUT
1.99
But
1.92
but
1.88
Nevertheless
1.85
And
1.82
Then
1.78
Nonetheless
1.65
Activations Density 0.038%