INDEX

Explanations

references to code-related functions, syntax, or programming constructs

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

 unspecified

-0.61

継

-0.53

 Expl

-0.51

 fancy

-0.50

[{\

-0.50

 merkwür

-0.50

 zast

-0.48

odd

-0.47

спекти

-0.47

odd

-0.47

POSITIVE LOGITS

unique

2.70

 unique

2.58

Unique

2.53

 Unique

2.52

 UNIQUE

2.28

UNIQUE

2.03

 uniques

1.99

 unieke

1.93

 uniquely

1.91

 uniqueness

1.90

Activations Density 0.032%