INDEX

Explanations

council election

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

 televised

-0.14

 masturb

-0.12

 masturbating

-0.11

 weddings

-0.11

SEX

-0.10

 Rentals

-0.10

SEX

-0.10

 divorce

-0.10

 interviews

-0.10

 debugging

-0.10

POSITIVE LOGITS

 vague

0.09

 innoc

0.09

 chromosomes

0.08

 innocence

0.08

nu

0.08

Clip

0.08

 nuanced

0.08

eec

0.08

 dusty

0.08

 intuition

0.08

Activations Density 1.061%