INDEX
Explanations
words related to abundance or fullness
New Auto-Interp
Negative Logits
ners
-0.16
dess
-0.16
keys
-0.15
ook
-0.15
wo
-0.15
pon
-0.15
atch
-0.15
933
-0.15
湯
-0.15
ot
-0.15
POSITIVE LOGITS
withd
0.22
滿
0.18
满
0.18
-full
0.17
à¹Ħà¸Ľ
0.15
'gc
0.15
acomment
0.15
aeda
0.15
ıydı
0.15
filled
0.14
Activations Density 0.048%