INDEX
Explanations
descriptive phrases ending in punctuation
New Auto-Interp
Negative Logits
↵
1.25
.;
0.98
("/");0.88
ise
0.79
ive
0.78
szint
0.77
rukt
0.77
»;
0.75
"/
0.75
gebn
0.73
POSITIVE LOGITS
While
0.89
Rather
0.87
Nevertheless
0.83
Despite
0.77
Ultimately
0.74
Therefore
0.74
Yet
0.71
bref
0.71
Nonetheless
0.68
Modern
0.67
Activations Density 0.323%