INDEX

Explanations

with

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

-0.59

in

-0.57

the

-0.56

 them

-0.51

-0.49

 member

-0.49

-0.48

 cientí

-0.47

up

-0.47

 domestiques

-0.47

POSITIVE LOGITS

ThroughAttribute

0.71

 Audiodateien

0.71

таратура

0.71

 MainAxisSize

0.70

Билгалдахарш

0.68

CompilerServices

0.68

BibitemShut

0.66

 ferons

0.66

 surla

0.66

Revenir

0.65

Activations Density 0.076%