INDEX

Explanations

titles starting with the

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 mété

-1.53

 rembourse

-1.51

 siè

-1.47

 maksi

-1.38

 poë

-1.38

졍

-1.35

 dépenses

-1.34

 DOING

-1.32

 conformité

-1.29

满脸

-1.27

POSITIVE LOGITS

 only

1.95

not

1.91

one

1.87

 making

1.78

 lost

1.76

 truly

1.73

 great

1.73

 time

1.73

 first

1.71

all

1.69

Activations Density 0.063%