INDEX

Explanations

Identifying markers

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 signature

-1.08

signature

-1.03

 signatures

-0.95

signatures

-0.94

 Signature

-0.93

 Signatures

-0.91

 SIGNATURE

-0.90

Signature

-0.85

WebMethod

-0.84

UrlResolution

-0.84

POSITIVE LOGITS

Външни

0.48

DockStyle

0.41

 quoc

0.41

 developer

0.39

Siv

0.38

ña

0.38

dff

0.37

 Conseils

0.36

sei

0.36

fler

0.36

Activations Density 0.004%