INDEX
Explanations
hypothetical examples of desires
New Auto-Interp
Negative Logits
ataupun
1.11
insbesondere
1.06
terutama
1.01
soprattutto
1.01
surtout
0.99
특히
0.98
nonché
0.98
maupun
0.97
comunque
0.97
bespoke
0.95
POSITIVE LOGITS
Suppose
1.02
might
0.95
Example
0.94
Suppose
0.94
Example
0.93
say
0.91
would
0.90
might
0.89
Might
0.86
Baseball
0.85
Activations Density 0.449%