INDEX

Explanations

technical units or programming constructs

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



-2.55

 otras

-2.39

竊

-2.20

 москва

-2.16

🫃

-2.11

icznej

-2.08

 аккура

-2.06

 solicitud

-2.02

 ผม

-1.92

姈

-1.91

POSITIVE LOGITS

3.11

end

2.77

“

2.72

2.50

棪

2.48

section

2.47

her

2.44

from

2.38

}=\

2.38



2.31

Activations Density 0.001%