INDEX

Explanations

develop

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

ValueStyle

-0.85

 snippetHide

-0.73

 صوتيه

-0.73

+#+#

-0.71

clusal

-0.69

StoryboardSegue

-0.69

RenderAtEndOf

-0.69

CloseOperation

-0.69

EndContext

-0.66

afficheront

-0.66

POSITIVE LOGITS

 bank

0.42

0.39

oo

0.39

hi

0.39

sk

0.39

 banks

0.38

 chocolate

0.38

hesus

0.36

ғ

0.36

bank

0.36

Activations Density 0.010%