INDEX

Explanations

phrases starting with specific words

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

ksesta

-1.27

らしいです

-1.22

voer

-1.22

Вариант

-1.20

Глава

-1.17

Часть

-1.16

早就

-1.16

 один

-1.16

と思われます

-1.16

agerie

-1.15

POSITIVE LOGITS

to

1.55

 prohibido

1.44

 zuges

1.42

₂,

1.41

:[],

1.40

 obligado

1.39

 extremadamente

1.38

 newValue

1.37

 lack

1.36

 even

1.33

Activations Density 0.003%

No Known Activations