INDEX

Explanations

noise

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

anine

-0.29

Tick

-0.29

itted

-0.28

reff

-0.28

éĩįä¸Ń

-0.27

 tick

-0.27

.deb

-0.27

vac

-0.26

avity

-0.25

etz

-0.24

POSITIVE LOGITS

 floor

0.35

åľ°æĿ¿

0.34

 floors

0.33

-cancel

0.28

floor

0.28

æľ¨åľ°æĿ¿

0.28

roperties

0.28

-dom

0.28

 shaping

0.27

èĩ

0.27

Activations Density 0.894%