INDEX

Explanations

physical work and energy

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

-anchor

-0.10

ToWorld

-0.08

zeti

-0.08

encies

-0.08

lege

-0.07

å°ĸ

-0.07

atici

-0.07

udier

-0.07

ìĿ´ì¦Ī

-0.07

teg

-0.07

POSITIVE LOGITS

 sorÃ¡n

0.08

åģļåĬŁ

0.08

åĿĬ

0.07

ascal

0.07

ä½ľçĶ¨äºİ

0.07

æĺ¯ä¸Ģæ¬¡

0.07

 done

0.07

 intervening

0.07

ä¹ĭåĬŁ

0.06

ä¸īéĥ¨æĽ²

0.06

Activations Density 0.001%