INDEX

Explanations

code and structured text syntax

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

のですか

-1.33

 where

-1.23

 will

-1.23

ようです

-1.20

はずです

-1.16

 suceso

-1.16

ようになりました

-1.14

 Also

-1.13

as

-1.13

of

-1.12

POSITIVE LOGITS

 kemer

1.34

titud

1.33

<bos>

1.30

 oignon

1.27

 leher

1.25

暁

1.25

 DOMINGO

1.23

卨

1.23

 réinitial

1.21

PÉ

1.20

Activations Density 0.001%