INDEX
Explanations
terms related to similarity and resemblance
New Auto-Interp
Negative Logits
dymyr
-0.71
يتيمه
-0.65
yoksa
-0.64
es
-0.56
넘
-0.56
kette
-0.55
DeleteMapping
-0.55
要不要
-0.53
Blackman
-0.52
imshow
-0.52
POSITIVE LOGITS
similar
3.01
Similar
2.76
similar
2.76
Similar
2.71
SIMILAR
2.52
similaire
2.16
similares
2.08
simila
1.94
similarly
1.90
similaires
1.89
Activations Density 0.074%