INDEX

Explanations

programming keywords followed by syntax

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

to

-2.28

 which

-1.88

for

-1.80

and

-1.55

 some

-1.52

 with

-1.48

 while

-1.45

imbang

-1.41

 this

-1.26

 including

-1.26

POSITIVE LOGITS

镦

1.45

.(

1.40

 возник

1.37

1.35

 beschränkt

1.34

是因為

1.33

把它

1.30

擽

1.30

 tornou

1.30

쨩

1.27

Activations Density 0.038%