INDEX

Explanations

code/markup

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

Sortie

-0.47

amethasone

-0.45

 '\\;'

-0.44

 FBref

-0.43

GrantedAuthority

-0.41

urion

-0.41

χρι

-0.40

othelial

-0.40

BackStack

-0.40

 userDao

-0.39

POSITIVE LOGITS

ConstraintMaker

0.75

 noDo

0.69

 Signalez

0.64

 Break

0.59

}>{

0.59

بوابة

0.59

 CanadaChoose

0.57

 Keep

0.56

Portail

0.56

[]:

0.56

Activations Density 0.014%