INDEX

Explanations

version numbers or 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

 sudah

-1.35

ablanca

-1.34

已經

-1.34

也太

-1.32

 already

-1.31

its

-1.28

可能です

-1.25

还不是

-1.24

已经

-1.23

 telah

-1.23

POSITIVE LOGITS

 Which

1.38

 there

1.33

 esquerdo

1.32

。

1.30

 naquele

1.23

of

1.21

 These

1.19

 They

1.16

bánh

1.15

~-

1.15

Activations Density 0.682%