INDEX

Explanations

Requests for user action

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

agli

-0.29

è¯·æ³¨æĦı

-0.28

 note

-0.27

note

-0.26

azard

-0.26

Ã³m

-0.26

\Modules

-0.26

Cou

-0.25

ansi

-0.25

 sÃ»r

-0.25

POSITIVE LOGITS

@class

0.31

è½¬è½½è¯·

0.28

adius

0.27

blr

0.26

 Calories

0.26

 tagged

0.25

 contributes

0.25

pollo

0.25

oir

0.25

æ±Ģ

0.24

Activations Density 0.009%