INDEX

Explanations

pondering, considering, or pondicherry

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

HING

-0.12

tes

-0.12

 Giov

-0.12

ters

-0.11

viar

-0.11

imb

-0.11

ten

-0.11

toa

-0.11

oop

-0.11

 Yours

-0.11

POSITIVE LOGITS

ously

0.15

osa

0.13

 ponder

0.10

erer

0.10

ous

0.10

osity

0.09

why

0.09

ÛĮØ±

0.09

ance

0.09

 deeply

0.08

Activations Density 0.003%