INDEX

Explanations

here to help or listen

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 높아

1.33

גיע

1.28

präsident

1.24

 приводит

1.24

 শাক

1.23

 hastal

1.19

 चाहे

1.15

 ডাকা

1.15

 zieht

1.15

 উচিত

1.15

POSITIVE LOGITS

spirited

1.60

intro

1.59

 opportune

1.55

 chorizo

1.55

 incisions

1.53

<unused298>

1.51

 itineraries

1.51

 concise

1.51

<unused2032>

1.50

urb

1.50

Activations Density 0.146%