INDEX

Explanations

ability to do something

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

 違い

-1.30

となっています

-1.23

mpagne

-1.22

urllib

-1.19

 なく

-1.17

intelligible

-1.17

AMIL

-1.15

是多少

-1.15

 како

-1.15

aphne

-1.14

POSITIVE LOGITS

las

1.41

1.40

1.34

 apalagi

1.33

of

1.32

doc

1.30

1.27

silien

1.27

they

1.26

}_{

1.23

Activations Density 0.037%