INDEX

Explanations

ranked search results

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

ature

-0.10

ultimate

-0.09

ago

-0.09

oo

-0.09

ese

-0.09

ForResult

-0.09

ish

-0.09

oho

-0.08

oen

-0.08

ning

-0.08

POSITIVE LOGITS

ë¬¼

0.12

ion

0.11

çĻĸ

0.11

ë¬¼ìĿĦ

0.11

 ÑĩÐµÐ³Ð¾

0.11

ä¸°ç¡ķ

0.11

ä¸įä½³

0.10

swith

0.10

ifle

0.10

ponsible

0.10

Activations Density 0.069%