INDEX

Explanations

for example

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

to

-1.44

 både

-0.86

给

-0.85

suited

-0.84

 zarówno

-0.83

pck

-0.80

 даже

-0.78

 bagi

-0.76

 incluso

-0.76

 especially

-0.72

POSITIVE LOGITS

for

2.95

For

1.86

for

1.54

For

1.53

 misalnya

1.36

 bijvoorbeeld

1.30

 например

1.29

например

1.22

FOR

1.22

 για

1.20

Activations Density 0.074%