INDEX

Explanations

formal writing

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

ashes

-0.71

ashe

-0.71

 mourut

-0.68



-0.65

 مشين

-0.64

 spalle

-0.64

 refusé

-0.63

licherweise

-0.62

 themſelves

-0.60

 avoient

-0.60

POSITIVE LOGITS

SourceChecksum

0.53

Injectable

0.49

 numberOfRows

0.45

 slip

0.44

 Wikimedijinoj

0.43

 blur

0.43

 rocks

0.42

 shooter

0.42

MessageOf

0.41

StoreMessageInfo

0.41

Activations Density 0.307%