INDEX
Explanations
phrases indicating contrast or transition in ideas
New Auto-Interp
Negative Logits
cel
-0.16
.named
-0.15
619
-0.15
CEL
-0.14
unks
-0.14
£i
-0.14
Tomorrow
-0.14
ishes
-0.14
Daily
-0.14
Tomorrow
-0.14
POSITIVE LOGITS
ìĿ´ë²Ī
0.51
lần
0.36
again
0.36
desta
0.35
again
0.31
this
0.31
ÑĨÑĮого
0.29
this
0.28
ä»Ĭå¹´
0.28
ä¸Ģ次
0.28
Activations Density 0.372%