INDEX

Explanations

pointers and memory structures

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

Of

1.24

forcer

1.20

of

1.19

iktar

1.15

ıntı

1.13

шк

1.12

ⴱ

1.11

شق

1.10



1.09

Sch

1.09

POSITIVE LOGITS

 小学

1.52

 tentunya

1.51

 yarat

1.49

 tentu

1.47

 종합

1.45

 inilah

1.42

 وكذلك

1.40

 excellence

1.37

 also

1.36

 již

1.35

Activations Density 0.573%