INDEX

Explanations

deleting

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

 deleted

-1.55

 delete

-1.45

 deleting

-1.42

 deletes

-1.33

 Deleted

-1.31

deleted

-1.30

delete

-1.24

Deleting

-1.21

 deletion

-1.16

Deleted

-1.11

POSITIVE LOGITS

BorderFactory

0.61

apnews

0.58

brtc

0.58

<?

0.57

plings

0.54

ContentValues

0.54

 Мексичка

0.54

//

0.53

posedge

0.53

ceased

0.52

Activations Density 1.643%