INDEX

Explanations

Scientific summaries and links

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

 يتيمه

-0.61

Personendaten

-0.56

 autorytatywna

-0.56

 Kome

-0.56

PreferredItem

-0.55

 незавершена

-0.55

 Signalez

-0.55

:✨

-0.55

хьтан

-0.54

BagConstraints

-0.50

POSITIVE LOGITS

DW

0.52

ături

0.50

 fallu

0.49

 Parigi

0.49

 ring

0.49

MMdd

0.48

 sonno

0.47

 andato

0.47

DW

0.47

ضور

0.47

Activations Density 0.001%