INDEX

Explanations

the word "variable" and related 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

 Spence

-0.70

 Connelly

-0.66

vun

-0.65

 Cáceres

-0.60

 endorse

-0.57

 Storey

-0.57

 Dorsey

-0.56

vic

-0.56

 Morales

-0.56

 Connolly

-0.54

POSITIVE LOGITS

 variable

2.48

 Variable

2.42

variable

2.36

 VARIABLE

2.31

Variable

2.27

 variables

2.22

 Vari

2.20

 Variables

2.20

Vari

2.09

 vari

2.02

Activations Density 0.449%