INDEX

Explanations

common patterns and specific terms

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

ḿ

-1.60

 innumer

-1.58

蛱

-1.56

厛

-1.56

毟

-1.49

 mesmos

-1.48

皷

-1.46

strange

-1.45



-1.45

 interesses

-1.45

POSITIVE LOGITS

 others

1.98

 other

1.91

 high

1.86

on

1.84

 first

1.77

It

1.76

 with

1.65

 Other

1.63

 three

1.53

 large

1.52

Activations Density 0.051%