INDEX

Explanations

online or digital services and content

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

 loose

-0.08

 Zion

-0.08

 loft

-0.08

 loaf

-0.08

RF

-0.08

 elev

-0.08

 Scouts

-0.08

DEM

-0.07

 Imper

-0.07

 incomplete

-0.07

POSITIVE LOGITS

inden

0.11

 Grat

0.11

 Nich

0.10

ayne

0.10

anja

0.10

TL

0.09

 aure

0.09

 Nicolas

0.09

aris

0.09

pch

0.09

Activations Density 0.002%

No Known Activations