INDEX

Explanations

code and math

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

 avoient

-0.78

Tikang

-0.72

 manqué

-0.69

 BrowserModule

-0.68

 entanto

-0.68

 étoient

-0.66

 émotion

-0.64

ήσει

-0.64

 Majefty

-0.63

:;

-0.62

POSITIVE LOGITS

<bos>

0.79

))^{

0.44

sizeCache

0.41

the

0.40

 Lightboxes

0.36

expandindo

0.36

HostException

0.36

ordano

0.35

enterOuterAlt

0.35

ьаж

0.35

Activations Density 0.002%