INDEX

Explanations

No Explanations Found

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

 filling

-0.26

(Uri

-0.26

 fill

-0.24

çļĦèĥĮåĲİ

-0.24

 suffering

-0.23

_SEC

-0.23

 integral

-0.23

/fast

-0.23

igration

-0.23

 impulse

-0.23

POSITIVE LOGITS

æĽ´å¤§

0.34

æĽ´å¼º

0.28

 Creat

0.28

izador

0.28

 Prom

0.27

æĽ´å¥½åľ°

0.27

æĽ´å¤§çļĦ

0.27

æĿ¥æī¾

0.26

 Improve

0.25

irector

0.25

Activations Density 0.000%

No Known Activations

This feature has no known activations.