INDEX
Explanations
granite or gran followed by specific words
New Auto-Interp
Negative Logits
unlabeled
0.40
un
0.40
meng
0.39
ces
0.38
alupe
0.37
unu
0.37
aba
0.37
大
0.37
inflammatory
0.37
experimental
0.36
POSITIVE LOGITS
Gran
0.55
GRAN
0.53
Gran
0.52
granular
0.48
granular
0.46
gran
0.46
permisos
0.44
gran
0.42
粒
0.41
granules
0.40
Activations Density 0.011%