INDEX

Explanations

Proper nouns and locations

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

 Himo

-0.70

OGND

-0.64

afficheront

-0.61

ⓧ

-0.60

 Kram

-0.59

 Meksiku

-0.59

 насељу

-0.58

"}")

-0.58

httphttps

-0.57

 black

-0.57

POSITIVE LOGITS

 Poincaré

0.44

őzés

0.44

 self

0.44

FontAwesomeIcon

0.43

oriasis

0.43

 Joseph

0.43

HttpPost

0.42

joe

0.42

 estekak

0.42

}}^{(

0.42

Activations Density 0.002%