INDEX
Explanations
punctuation and conjunctions
New Auto-Interp
Negative Logits
以免
0.50
despite
0.48
apesar
0.46
considering
0.45
despite
0.44
considering
0.43
lest
0.43
nonostante
0.43
fearing
0.43
avoid
0.43
POSITIVE LOGITS
而且
0.42
很简单
0.42
:"",
0.42
существуют
0.41
：
0.41
:'',
0.40
obtenue
0.39
characterised
0.39
并不是
0.39
አይደለም
0.39
Activations Density 0.152%