INDEX

Explanations

Liquids in systems

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

åĢĴ

-0.29

å°Ĭ

-0.29

æĤ¬

-0.28

éķ¿å¤§

-0.28

æī¿è®¤

-0.28

SUPER

-0.27

 inverted

-0.27

kap

-0.26

 slur

-0.26

æĤ¬æµ®

-0.26

POSITIVE LOGITS

 boost

0.32

éĺ²

0.29

çľŃ

0.28

(boost

0.27

 contractor

0.26

OwnerId

0.25

Ã¨les

0.24

å¸Ī

0.24

eres

0.24

boost

0.24

Activations Density 0.101%