INDEX

Explanations

code block following `using`

indentation (leading whitespace) in code, particularly deeper nesting levels.

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

-1.57

to

-1.48

യും

-1.42

</b>

-1.40

 with

-1.38

紲

-1.38

 varier

-1.34

ğum

-1.34

=\""

-1.34

[])

-1.30

POSITIVE LOGITS

 вслед

1.52

 estando

1.48

 llamaba

1.38

 tomando

1.37

 riten

1.34

 attacco

1.34

 determinados

1.33

各样的

1.31

 there

1.30

 conditions

1.30

Activations Density 0.149%