INDEX

Explanations

potential, disorder, lattices

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

Köszi

-1.44

因為

-1.39

私の

-1.38

Werde

-1.37

 навіть

-1.33

 spezielle

-1.32

Дорогие

-1.30

 bestimmte

-1.29

Spenden

-1.27

 чому

-1.27

POSITIVE LOGITS

We

1.69

paksa

1.38

هره

1.36

⅙

1.33

you

1.23

ógy

1.23

∓

1.22

Our

1.21

髄

1.21

 Ours

1.20

Activations Density 0.073%