INDEX
Explanations
conjunctions and transition phrases that indicate contrast or addition
New Auto-Interp
Negative Logits
ſelf
-1.04
itſelf
-1.00
themſelves
-0.97
Majefty
-0.92
*/;
-0.90
himſelf
-0.87
tvguidetime
-0.87
ſelves
-0.86
Portale
-0.85
بوابة
-0.84
POSITIVE LOGITS
But
0.67
I
0.62
And
0.62
And
0.59
But
0.58
-
0.54
但是
0.54
maybe
0.51
Maybe
0.51
it
0.47
Activations Density 0.191%