INDEX

Explanations

seeking fixes or solutions

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

-1.63

it

-1.44

-1.41

of

-1.37

={};

-1.30

никами

-1.23

ेंट

-1.21

olique

-1.20

 escenario

-1.19

 kurie

-1.19

POSITIVE LOGITS

 suggested

1.51

 weißt

1.49

懨

1.44

選手の

1.42

 jemals

1.41

 kombi

1.40

 geholfen

1.38

หรือ

1.36

 initially

1.35

 nachher

1.35

Activations Density 0.102%