INDEX

Explanations

in-depth, in-context, inclusion

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

浒

1.32

 chóng

1.24

伴

1.23

 naught

1.22

本

1.20

 purposes

1.19

 زائ

1.18

บัน

1.18

 ridiculously

1.17

กระท

1.16

POSITIVE LOGITS

やすい

1.72

activo

1.61

 silico

1.60

forcer

1.56

ositol

1.49

voi

1.48

Depth

1.47

valids

1.46

depth

1.46

sekten

1.46

Activations Density 0.195%