INDEX

Explanations

perhaps then

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

undi

-0.10

ï¸

-0.09

 Foundations

-0.09

 Para

-0.09

Para

-0.09

 unseen

-0.09

 Feinstein

-0.09

 annihil

-0.09

 articulate

-0.09

ORED

-0.08

POSITIVE LOGITS

amber

0.10

gif

0.09

esz

0.09

istes

0.09

 fascination

0.08

 searcher

0.08

ved

0.08

uche

0.08

 Javier

0.08

/of

0.08

Activations Density 0.004%