INDEX

Explanations

ever

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ever

-1.01

 kasarigan

-0.76

 ever

-0.73

theless

-0.72

 varandra

-0.71

sidemargin

-0.70

EVER

-0.69

 skolan

-0.69

 Administrativna

-0.69

Ever

-0.67

POSITIVE LOGITS

 example

0.59

 statement

0.59

 attempt

0.54

 measurement

0.54

 effort

0.52

 total

0.52

 test

0.52

 joint

0.51

few

0.50

Activations Density 0.071%