INDEX

Explanations

episode

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

 authorities

-0.61

 defaultstate

-0.60

authorities

-0.55

 Authorities

-0.52

 AUTHORITIES

-0.52

lüssel

-0.51

Authorities

-0.50

 disambiguazione

-0.49

begre

-0.48

évaluateur

-0.47

POSITIVE LOGITS

новниш

0.74

asilkan

0.63

Datuak

0.62

NgModule

0.57

__':

0.57

 Walkover

0.57

 kasarigan

0.56

fjspx

0.56

PyExc

0.55

Diwedd

0.54

Activations Density 0.090%