INDEX

Explanations

action and consequence

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

 дуже

-1.78

 öde

-1.48

 ihan

-1.48

 більше

-1.45

if

-1.45

 bunu

-1.44

our

-1.42

 мені

-1.41

ظهار

-1.39

 bize

-1.34

POSITIVE LOGITS

 will

1.84

🪛

1.77

 será

1.71

 expected

1.68

 vertrou

1.57

нє

1.56

 likely

1.55

 indruk

1.54

 wezen

1.54

ñadir

1.54

Activations Density 0.596%