INDEX

Explanations

core principles and concepts

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

 nostalg

2.05

 cheeky

1.93

 flirt

1.87

 aventuras

1.82

 sassy

1.77

 hipster

1.77

 aventura

1.77

 elegante

1.76

 husky

1.75

 gangsters

1.72

POSITIVE LOGITS

Types

1.56

 classification

1.48

Definition

1.47

 determining

1.46

 characteristics

1.45

 definition

1.44

 Classification

1.44

Characteristics

1.38

 differentiation

1.37

 importance

1.36

Activations Density 0.128%