INDEX

Explanations

Documentation, explanations

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 REMEMBER

-0.93

 remember

-0.93

remember

-0.83

 remembers

-0.82

دانشنامهٔ

-0.81

 Roskov

-0.80

isContained

-0.79

:✨

-0.79

 Remember

-0.75

Remember

-0.74

POSITIVE LOGITS

 modern

0.41

ToScroll

0.40

inOrder

0.37

</>

0.35

 accurate

0.34

нью

0.34

civ

0.33

 medial

0.33

currentColor

0.32

Civ

0.32

Activations Density 0.000%