INDEX

Explanations

logic and mathematical proofs

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

rungsseite

-0.69

<bos>

-0.60

expandindo

-0.59

хьтан

-0.57

 supérieurs

-0.55

 >=",

-0.53

niająca

-0.52

 Signalez

-0.50

المشاركات

-0.49

GEBURTS

-0.48

POSITIVE LOGITS

esgue

0.66

çek

0.63

 pylint

0.57

kegaard

0.57

mability

0.56

iculous

0.56

 pixabay

0.55

 Weyl

0.55

culada

0.55

[::-

0.54

Activations Density 0.095%