INDEX

Explanations

a or one of

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

adol

-0.11

 yellow

-0.10

 seemingly

-0.10

 seventeen

-0.10

ardin

-0.10

 eleven

-0.10

 worse

-0.09

 mysterious

-0.09

 eighty

-0.09

actually

-0.09

POSITIVE LOGITS

 preferably

0.38

 either

0.31

 whichever

0.25

 ideally

0.25

 whatever

0.24

 EITHER

0.24

either

0.22

 Either

0.22

Either

0.22

prefer

0.22

Activations Density 1.027%