INDEX

Explanations

high

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Emer

-0.27

elist

-0.26

 emerg

-0.25

bel

-0.25

eldo

-0.25

åı£æ°Ķ

-0.25

Na

-0.24

è¾¹

-0.24

mond

-0.24

åı£

-0.24

POSITIVE LOGITS

arily

0.29

Ð¸ÑĩÐµÑģÐºÐ¸Ñħ

0.28

å®ĮæĪĲ

0.28

hac

0.27

å²£

0.26

 completion

0.26

 Vest

0.25

aras

0.25

Mocks

0.24

æĪĲçĨŁçļĦ

0.24

Activations Density 0.882%