INDEX

Explanations

measure

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

 measure

-1.28

measure

-1.21

 measures

-1.09

 measurement

-1.08

 Measure

-1.06

measures

-1.03

 measuring

-1.02

Measuring

-1.01

Measure

-1.00

measurement

-0.99

POSITIVE LOGITS

 Vikipedi

0.42

 Soup

0.40

 Lipstick

0.40

 Milk

0.38

PYX

0.38

 grandkids

0.36

zra

0.36

?,?,

0.36

 melk

0.36

 grandchildren

0.35

Activations Density 0.730%