INDEX

Explanations

loosen

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

 betweenstory

-1.30

RenderAtEndOf

-1.29

aarrggbb

-1.15

RetentionPolicy

-1.11

Personendaten

-1.07

 weaken

-1.06

 disambiguazione

-1.06

WithIOException

-1.03

 weakens

-1.02

 AssemblyCompany

-0.98

POSITIVE LOGITS

up

0.97

in

0.65

to

0.60

the

0.60

ley

0.59

 full

0.59

up

0.57

0.56

Activations Density 1.585%