INDEX

Explanations

influenced by

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

 fungerer

-1.28

 начинает

-1.24

 sikkert

-1.23

呖

-1.20

 hemsida

-1.20

 vermag

-1.19

ナブル

-1.18

 battaglia

-1.16

 help

-1.16

 входит

-1.14

POSITIVE LOGITS

by

4.34

of

2.17

 bởi

1.93

 oleh

1.81

 توسط

1.80

 tarafından

1.56

 with

1.49

by

1.48

 بواسطة

1.41

 любого

1.37

Activations Density 0.083%