INDEX

Explanations

internal components or logic

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

Gar

1.72

לע

1.65

↵↵

1.65

↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵

1.56

شو

1.56

تا

1.55

↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵

1.54

Charged

1.53

Fruit

1.53

Czas

1.51

POSITIVE LOGITS

<unused2140>

1.33

 undersea

1.28

 triplet

1.27

த்த

1.26

 terrestrial

1.25

ール

1.24

ashvili

1.24

ॅमिली

1.24

 geometric

1.24

 accumulator

1.24

Activations Density 0.000%