INDEX
Explanations
descriptive adverbs and specific items
New Auto-Interp
Negative Logits
moy
0.46
coord
0.43
btree
0.42
cust
0.42
mattered
0.42
ax
0.41
that
0.41
"'.$
0.41
part
0.40
abat
0.40
POSITIVE LOGITS
க
0.54
தினம்
0.52
dự
0.51
ไตล์
0.51
曛
0.50
stylu
0.49
festivities
0.48
Alltag
0.47
стиля
0.47
minimalist
0.46
Activations Density 0.025%