INDEX

Explanations

average/mean

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

 averages

-0.30

 dispersed

-0.28

 averaging

-0.27

[assembly

-0.26

åĪĨæķ£

-0.24

åħ¨æł¡

-0.24

 Carn

-0.24

 away

-0.24

 stripper

-0.24

æķ£å¸ĥ

-0.24

POSITIVE LOGITS

Case

0.30

æ¡Īä¾ĭ

0.29

æ¡Ī

0.29

case

0.29

 case

0.28

peed

0.28

ÐļÑĥÑĢÑģ

0.28

_case

0.27

WithValue

0.26

á»§a

0.26

Activations Density 0.108%