INDEX

Explanations

code component and class names

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

ondres

-1.44

並且

-1.38

lysses

-1.34

to

-1.29

𝗷

-1.27

 arasındaki

-1.25

торами

-1.23

Jis

-1.23

Ꮞ

-1.21

 therefore

-1.19

POSITIVE LOGITS

If

1.34

 даже

1.26

۲

1.24

Make

1.23

 иногда

1.20

Život

1.20

 sogar

1.19

Here

1.19

section

1.17

1.16

Activations Density 0.531%