INDEX

Explanations

broader + context/field/role

New Auto-Interp

Configuration

Prompts (Dashboard)

238,145 prompts, 512 tokens each

Dataset (Dashboard)

lmsys + oasst1

Embeds

IFrame

Link

Negative Logits

 sólo

0.40

Entire

0.39

 keseluruhan

0.37

 менее

0.36

 簡単

0.36

 CGRectMake

0.36

 thêm

0.35

arding

0.35

 일반

0.35

 более

0.35

POSITIVE LOGITS

 unimportant

0.38

ད

0.37

Ժ

0.36

 forbidden

0.36

 contradictions

0.36

Ange

0.36

 looming

0.36

 shitty

0.36

睡觉

0.36

谲

0.36

Activations Density 0.008%

No Known Activations