INDEX

Explanations

berg Bergeron Days vs everyone

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

 Hindus

-0.11

woods

-0.11

uder

-0.11

lick

-0.11

obe

-0.10

 Paige

-0.10

riot

-0.10

 Rosie

-0.10

 fossils

-0.10

/licenses

-0.10

POSITIVE LOGITS

aget

0.08

HM

0.08

 Dere

0.08

 ..↵

0.07

Dot

0.07

hurst

0.07

 freel

0.07

akra

0.07

 risky

0.07

 dise

0.07

Activations Density 0.001%