INDEX

Explanations

contrast that, not guilty, impossible to, few tweaks

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

 Already

-1.50

 już

-1.47

Takk

-1.38

引言

-1.38

 GRANDE

-1.38

Already

-1.37

Dobrý

-1.36

意大利

-1.36

DESCRIPCIÓN

-1.35

 despedida

-1.34

POSITIVE LOGITS

 using

1.82

and

1.76

for

1.66

 called

1.63

two

1.48

use

1.47

 used

1.44

 groene

1.43

item

1.36

or

1.35

Activations Density 0.081%