INDEX

Explanations

varied informational text

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

oogle

-0.27

unken

-0.27

æ³ķåºŃ

-0.25

oir

-0.25

çļĦæĶ¯æĮģ

-0.24

çļĦåİĨåı²

-0.24

zel

-0.23

pii

-0.23

strcmp

-0.23

both

-0.23

POSITIVE LOGITS

epad

0.27

'_'

0.27

'__

0.26

isse

0.25

ä¸ĢçŃīå¥ĸ

0.25

ä¸»æ¼Ķ

0.25

é¡¿

0.25

issant

0.25

 spoon

0.24

 Oral

0.23

Activations Density 0.003%