INDEX
Explanations
instances of the word 'go' and its variants, indicating motion or action
New Auto-Interp
Negative Logits
upon
-0.28
upon
-0.24
Upon
-0.21
Upon
-0.19
/up
-0.16
ëĪĦ
-0.15
oki
-0.15
ull
-0.15
upd
-0.14
üzerine
-0.14
POSITIVE LOGITS
own
0.28
-on
0.27
Own
0.23
Own
0.22
OWN
0.22
own
0.20
-On
0.20
ON
0.19
ON
0.18
owns
0.18
Activations Density 0.052%