INDEX

Explanations

diagnostic manual (DSM)

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

is

1.79

в

1.57

at

1.48

1.42

jects

1.40

1.33

as

1.33

़

1.32

Is

1.30

Intermediate

1.29

POSITIVE LOGITS

 décès

2.39

 réellement

2.33

 différences

2.31

 IUnary

2.26

 elytris

2.25

𒊶

2.24

 getSchema

2.22

<unused1399>

2.21

<unused1968>

2.21

𐱃

2.21

Activations Density 0.089%