INDEX

Explanations

object

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

é¥±

-0.26

èĥ¥

-0.26

ISK

-0.26

rase

-0.25

_Entry

-0.25

.cond

-0.24

ustum

-0.24

áº£n

-0.24

Sie

-0.24

è¤¶

-0.24

POSITIVE LOGITS

pf

0.31

-graph

0.30

variable

0.30

 defined

0.29

 oriented

0.29

èĭĶ

0.29

 variable

0.29

oriented

0.28

 written

0.28

çĶŁåĳ½

0.28

Activations Density 0.949%