INDEX
Explanations
numerical references and citations in texts
New Auto-Interp
Negative Logits
third
-0.37
Third
-0.35
3
-0.33
three
-0.33
Three
-0.32
第ä¸ī
-0.32
ä¸ī
-0.32
Four
-0.32
03
-0.32
ä¸ī
-0.32
POSITIVE LOGITS
6
0.31
7
0.27
sixth
0.22
Sixth
0.20
ï¼ĸ
0.20
5
0.19
Û¶
0.19
006
0.19
06
0.18
8
0.18
Activations Density 0.111%