INDEX
Explanations
instances of emotional contrast and complex sentence structures
New Auto-Interp
Negative Logits
komp
-0.17
æģµ
-0.15
wer
-0.14
asel
-0.14
дод
-0.14
.tool
-0.14
604
-0.13
ÑĢоб
-0.13
ador
-0.13
esters
-0.13
POSITIVE LOGITS
Cove
0.17
second
0.15
bane
0.15
whereas
0.15
versus
0.14
arris
0.14
Grimm
0.14
pling
0.14
Ú
0.14
;
0.14
Activations Density 0.098%