INDEX
Explanations
interrogative phrases asking about personal situations or conditions
New Auto-Interp
Negative Logits
iddle
-0.17
engu
-0.15
apas
-0.15
326
-0.15
utton
-0.15
owa
-0.15
annie
-0.15
uthor
-0.15
reich
-0.14
оваÑĢ
-0.14
POSITIVE LOGITS
iere
0.15
/to
0.14
iel
0.14
elin
0.14
ůst
0.14
ValueType
0.13
endo
0.13
je
0.13
ForResource
0.13
æij
0.13
Activations Density 0.040%