INDEX

Explanations

published

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

ⓧ

-0.79

 Roskov

-0.78

DockStyle

-0.77

 RouterModule

-0.68

 betweenstory

-0.68

 насељу

-0.68

 AppCompatTheme

-0.66

 Ανακτήθηκε

-0.63

 Exacts

-0.61

Искәрмәләр

-0.60

POSITIVE LOGITS

by

0.62

in

0.54

 William

0.46

 March

0.44

 February

0.44

0.43

гу

0.42

heng

0.42

Ben

0.41

ereum

0.41

Activations Density 0.005%