INDEX

Explanations

it followed by verb or punctuation

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

of

-1.89

 like

-1.34

 Tenden

-1.30

吝

-1.24

 craindre

-1.24

ğine

-1.23

╹

-1.18

 uzyskać

-1.15

によっては

-1.14

 even

-1.13

POSITIVE LOGITS

are

1.32

1.23

and

1.21

 Posteriormente

1.20

够了

1.18

__["

1.18

 několik

1.18

Ρ

1.17

脸上的

1.17

ñones

1.13

Activations Density 0.119%