INDEX
Explanations
physical actions and movements of characters
New Auto-Interp
Negative Logits
hanging
-0.17
icros
-0.16
redo
-0.16
uri
-0.15
лав
-0.15
Stuff
-0.14
Pole
-0.14
hangs
-0.14
Hanging
-0.14
rades
-0.14
POSITIVE LOGITS
Lay
0.17
crawled
0.16
lying
0.16
lay
0.16
Whe
0.16
lie
0.16
heap
0.16
crawl
0.16
åľ¨åľ°
0.15
spraw
0.15
Activations Density 0.076%