INDEX

Explanations

improvement or improved concepts

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

 different

-1.44

离开了

-1.41

我们

-1.29

-1.28

 brancas

-1.27

Похо

-1.27

 suicidio

-1.23

 pomiędzy

-1.20

哪个

-1.19

Different

-1.18

POSITIVE LOGITS

or

1.79

 with

1.45

in

1.45

 your

1.29

the

1.25

ᴮ

1.25

 bertahan

1.23

Zalety

1.22

ordinator

1.18

RouterModule

1.16

Activations Density 0.128%