INDEX

Explanations

code, orders, and news

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

 itſelf

-0.85

 varandra

-0.82

]`

-0.81

 themſelves

-0.78

 skolan

-0.75

 ſeveral

-0.75

 Theſe

-0.74

 theſe

-0.71

 leſs

-0.71

ksesta

-0.71

POSITIVE LOGITS

ad

0.60

0.57

Sy

0.55

hus

0.54

vol

0.54

ne

0.51

lou

0.51

VOL

0.50

 PropertyChanged

0.48

 ujednoznacz

0.48

Activations Density 0.221%