INDEX

Explanations

document snippets

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

 بيها

-0.50

 quietly

-0.47

ICULO

-0.43

ביוגרפיה

-0.43

mander

-0.42

цездатний

-0.42

titura

-0.42

्रीय

-0.40

 Chrift

-0.40

 flagged

-0.40

POSITIVE LOGITS

 chance

1.34

 chances

1.27

chance

1.25

Chance

1.20

 Chances

1.15

 Chance

1.13

 CHANCE

1.13

Chances

1.12

 probability

1.05

probability

1.00

Activations Density 0.005%