INDEX

Explanations

configuration, command, tips

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

űt

1.27

フォルダ

1.24

ファイル

1.23

ッピー

1.17

('

1.16

File

1.15

Name

1.14

ウィ

1.11

ของผู้

1.11

ülő

1.10

POSITIVE LOGITS

 deter

1.24

 impede

1.24

 impro

1.22

 disrupting

1.20

 abhor

1.17

 depict

1.16

 brutal

1.13

 grim

1.11

 inev

1.11

 hinder

1.11

Activations Density 0.001%