INDEX

Explanations

parentheses followed by equals sign

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

anches

-0.10

icom

-0.10

(;

-0.10

eki

-0.10

.getLogger

-0.09

inder

-0.09

yles

-0.09

ArrayType

-0.09

rim

-0.09

 Carn

-0.09

POSITIVE LOGITS

 Whitney

0.08

ay

0.06

 Chop

0.06

 Tubes

0.06

 besides

0.06

.Never

0.06

OUNCE

0.06

sponsor

0.06

 defined

0.06

rium

0.06

Activations Density 0.208%