INDEX

Explanations

hereinafter referred to

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ented

1.25

<unused332>

1.22

<unused1841>

1.22

atorship

1.22

<unused338>

1.22

<unused1072>

1.21

ꗠ

1.19

<unused951>

1.18

<unused423>

1.17

ovaniyu

1.16

POSITIVE LOGITS

 está

1.30

 hier

1.28

 peinture

1.21

Er

1.20

Nek

1.19

 für

1.19

Er

1.18

 algéb

1.16

Pag

1.16

 équ

1.16

Activations Density 0.001%