INDEX

Explanations

forensics, studies, data

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

expandindo

-0.71

InvalidProtocol

-0.56

 ostavi

-0.53

 auguri

-0.50

 foret

-0.48

Eksterne

-0.47

 quente

-0.47

 voraus

-0.46

 hotter

-0.46

jetunion

-0.46

POSITIVE LOGITS

ser

0.70

 ball

0.65

 GEOLOGY

0.60

 onCreateView

0.59

od

0.59

 bals

0.54

 TOXIC

0.54

 BALL

0.52

 Anthropology

0.52

 anthropology

0.52

Activations Density 0.004%

No Known Activations