INDEX

Explanations

along

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

ⓧ

-0.55

__":

-0.53

 LoginComponent

-0.52

InjectAttribute

-0.48

 considérons

-0.47

apimachinery

-0.46

tamol

-0.45

UseVisualStyle

-0.45

 Observable

-0.44

adaptiveStyles

-0.44

POSITIVE LOGITS

ContentAsync

0.61

threw

0.58

makeText

0.56

OGND

0.54

filter

0.53

thrown

0.52

side

0.52

friendly

0.52

bearer

0.52

 FishBase

0.51

Activations Density 0.006%