INDEX

Explanations

forum posts

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 CanadaChoose

-0.59

 correctes

-0.53

\{\\

-0.51

tım

-0.51

 (;;)

-0.50

原始内容存档于

-0.50

 kwamba

-0.48

awaiter

-0.48

 neceffary

-0.48

/*---

-0.47

POSITIVE LOGITS

WithMany

0.66

photobucket

0.66

atist

0.55

ules

0.54

 astore

0.54

án

0.50

BorderFactory

0.49

 more

0.49

cional

0.48

atsby

0.48

Activations Density 0.004%