INDEX
Explanations
verbs associated with physical actions and movement
New Auto-Interp
Negative Logits
icus
-0.17
uddle
-0.15
pas
-0.14
afa
-0.14
æĤ
-0.14
ppelin
-0.14
uin
-0.14
ward
-0.13
cous
-0.13
yme
-0.13
POSITIVE LOGITS
into
0.21
_into
0.20
onto
0.18
0.17
onto
0.17
into
0.17
Into
0.17
chet
0.16
-alist
0.16
İ
0.16
Activations Density 0.176%