INDEX

Explanations

acronyms and abbreviations explained

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

bery

-0.07

uzzi

-0.07

tre

-0.07

ts

-0.07

ÛĮØ±

-0.07

 PTSD

-0.07

Cov

-0.07

 uplift

-0.07

Â¾

-0.07

HDD

-0.07

POSITIVE LOGITS

 acronym

0.21

stands

0.18

 abbreviation

0.18

pron

0.15

 stands

0.15

abbreviation

0.14

standing

0.14

 pronounced

0.14

 initials

0.13

 abbrev

0.13

Activations Density 0.457%

No Known Activations