INDEX

Explanations

doesn

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

ç¥¨

-0.32

åŃ³

-0.29

è¦ª

-0.28

-rad

-0.26

orna

-0.24

Knife

-0.24

æº¢ä»·

-0.24

èĸªæ°´

-0.24

 Mong

-0.24

åĪ°åľº

-0.24

POSITIVE LOGITS

è®ºè¯ģ

0.32

æĴ°åĨĻ

0.28

è®ºè¿°

0.26

loh

0.26

Argument

0.26

etic

0.26

å¯¹æ¯Ķ

0.25

ç¼®

0.25

esis

0.25

éĻĪè¿°

0.25

Activations Density 0.000%