INDEX
Explanations
conjunctions and transition words
New Auto-Interp
Negative Logits
0
0.39
НИ
0.38
ח
0.37
be
0.36
!";
0.35
sembles
0.34
ixed
0.34
кти
0.34
set
0.33
or
0.33
POSITIVE LOGITS
Furthermore
0.49
Furthermore
0.47
Consequently
0.46
Interestingly
0.46
Although
0.43
mivel
0.43
Because
0.43
चूंकि
0.43
Moreover
0.42
Integrating
0.41
Activations Density 0.023%