INDEX

Explanations

vit

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

ashion

-0.28

 Passage

-0.28

actories

-0.28

 padd

-0.27

anchise

-0.26

æīĵåĩº

-0.26

ç»ıè¿ĩå¤ļå¹´

-0.26

adian

-0.26

[obj

-0.26

digest

-0.26

POSITIVE LOGITS

oster

0.32

uper

0.31

å¡Ķ

0.28

OST

0.27

ORED

0.27

CESS

0.26

uro

0.26

lingen

0.25

oller

0.25

ray

0.25

Activations Density 0.008%