INDEX

Explanations

words ending in -er / mer / ter / ler

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

とき

-1.42

ときに

-1.35

ときは

-0.93

electrode

-0.89

timbangan

-0.88

sesion

-0.85

 decorativos

-0.84

 coloridos

-0.84

-0.83

slu

-0.80

POSITIVE LOGITS

rrrr

1.45

rrrrr

1.38

ific

1.34

ized

1.28

rrr

1.24

rrrrrrrr

1.17

tainment

1.16

rrrrrr

1.16

ised

1.16

rific

1.16

Activations Density 0.432%