INDEX
Negative Logits
“
0.69
them
0.64
**
0.63
‘
0.61
works
0.61
Working
0.60
Them
0.59
<sup>
0.58
ฯ
0.56
Subscribe
0.56
POSITIVE LOGITS
<unused1855>
0.94
<unused1724>
0.93
<unused338>
0.92
<unused1766>
0.92
<unused321>
0.91
<unused416>
0.90
<unused311>
0.90
<unused2066>
0.90
arder
0.89
sampah
0.88
Activations Density 0.040%