INDEX
Explanations
more information or quantity
New Auto-Interp
Negative Logits
uttosto
0.72
lighter
0.71
weaker
0.70
lighter
0.65
softer
0.62
Smaller
0.62
slower
0.61
thinner
0.61
Faster
0.60
较高
0.57
POSITIVE LOGITS
more
3.22
更多
3.08
więcej
2.95
more
2.94
mehr
2.91
більше
2.83
MORE
2.72
больше
2.70
More
2.66
更多的
2.64
Activations Density 0.219%