INDEX

Explanations

code snippets or keywords

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

 also

-1.90

 those

-1.83

 知道

-1.66

or

-1.66

 these

-1.63

Also

-1.59

on

-1.54

 также

-1.41

-1.38

也是

-1.35

POSITIVE LOGITS



1.68

饽

1.44

睁开

1.42

'){

1.41

 hade

1.40

goes

1.39

 reple

1.34

means

1.33

'>";

1.30

 hones

1.28

Activations Density 0.048%