INDEX

Explanations

so followed by adjectives/adverbs

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

 bestimmten

-1.34

up

-1.29

かなり

-1.24

 tudi

-1.21

 verheiratet

-1.20

冷漠

-1.15

 nawet

-1.11

くん

-1.11

 miała

-1.10

}$,

-1.08

POSITIVE LOGITS

we

1.51

 that

1.30

ようになりました

1.26

 praised

1.23

 anbef

1.18

kuri

1.17

となりました

1.15

 contactez

1.13

 tiap

1.13

 kost

1.12

Activations Density 0.019%