INDEX

Explanations

names

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

Lee

-0.79

Lee

-0.72

Биография

-0.71

مصادر

-0.68

titutes

-0.65

itarian

-0.62

keeping

-0.59

ⓧ

-0.59

etan

-0.57

LEE

-0.56

POSITIVE LOGITS

 Stroll

0.56

Portale

0.53

 base

0.52

 disarmament

0.52

ostar

0.51

Outlined

0.50

 calib

0.50

 AssemblyCompany

0.49

arache

0.49

 }}=\

0.49

Activations Density 0.915%