INDEX

Explanations

material wants and well-being

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

faith

-1.41

塒

-1.38

龗

-1.36

ܖ

-1.35

März

-1.34

蜑

-1.34

 Chapitre

-1.30

漚

-1.29

the

-1.29

/**

-1.28

POSITIVE LOGITS

↵↵

1.56

 even

1.49

 really

1.36

 with

1.30

 Those

1.27

’

1.21

 would

1.20

see

1.19

so

1.18

 What

1.16

Activations Density 0.031%