INDEX

Explanations

good followed by context

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

 such

-1.69

 solch

-1.34

the

-1.31

that

-1.25

各式

-1.18

Such

-1.17

大量的

-1.16

 Forty

-1.15

 midst

-1.13

 parcour

-1.12

POSITIVE LOGITS

」（

1.51

 cepillo

1.50

ależ

1.48

ToMove

1.41

 Obtén

1.39

chaffen

1.33

Crap

1.32

ırım

1.30

֟

1.29

 twarzy

1.28

Activations Density 0.039%