INDEX

Explanations

code serialization and types

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

 они

-1.36

 作り方

-1.27

 passieren

-1.25

 mancher

-1.23

ᨈ

-1.22

bbled

-1.20

どうも

-1.17

內容簡介

-1.16

водом

-1.16

こんばんは

-1.15

POSITIVE LOGITS

</b>

1.45

 efforts

1.42

 discussion

1.40

...");

1.22

</h2>

1.19

 zicht

1.16

 level

1.16

頫

1.16

 debate

1.16

䐀

1.15

Activations Density 0.003%