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

 grandchildren

-0.28

abant

-0.28

bay

-0.25

selectedIndex

-0.24

ofi

-0.24

ocket

-0.24

æ´Ľéĺ³

-0.24

è¿İ

-0.23

ofs

-0.23

&M

-0.23

POSITIVE LOGITS

èľļ

0.27

0.26

car

0.26

erte

0.25

alist

0.24

Dr

0.24

 Balanced

0.24

 news

0.23

 Fold

0.23

oÅĽÄĩ

0.23

Activations Density 0.057%

No Known Activations

This feature has no known activations.