INDEX

Explanations

gun- prefix for names and variations

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

anton

-0.11

ively

-0.11

 Muss

-0.09

exter

-0.09

adel

-0.09

ift

-0.09

 trous

-0.09

/fr

-0.09

 geom

-0.09

ORS

-0.09

POSITIVE LOGITS

pow

0.22

ners

0.14

ned

0.14

ning

0.13

linger

0.13

ther

0.13

shots

0.13

ner

0.13

metal

0.12

zburg

0.12

Activations Density 0.049%