INDEX

Explanations

1/f noise and fluctuations

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

 southward

-0.88

 northward

-0.85

INTO

-0.72

颔

-0.72

Earl

-0.70

 justiça

-0.70

 maximale

-0.70

 passaram

-0.69

 liquidated

-0.69

 distributive

-0.69

POSITIVE LOGITS

 noise

1.58

 pink

1.56

 flicker

1.30

 white

1.28

noise

1.23

 Noise

1.23

Pink

1.22

 fluctuations

1.21

pink

1.20

 Pink

1.20

Activations Density 0.075%