INDEX
Explanations
knowing and knowledge across languages
New Auto-Interp
Negative Logits
kBtu
0.39
grouped
0.38
Categoria
0.38
начну
0.38
jumbo
0.37
mergeddata
0.37
ákat
0.37
RM
0.36
بإ
0.36
линд
0.36
POSITIVE LOGITS
晓
0.54
connaître
0.51
知り
0.50
知
0.48
曉
0.48
conoscere
0.47
connaissance
0.47
knew
0.47
Biết
0.46
know
0.46
Activations Density 0.006%