INDEX

Explanations

code, numbers, and specific identifiers

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

久し

-1.63

総評

-1.41

 hırka

-1.37

archiw

-1.34

 anúncio

-1.28

accompag

-1.26

 czasem

-1.26

Хороший

-1.25

 caiu

-1.25

 beijo

-1.24

POSITIVE LOGITS

1.57

 again

1.34

</i>

1.29

1.28

</em>

1.22

”

1.21

its

1.21

1.20

1.17

is

1.16

Activations Density 0.001%