INDEX

Explanations

figure

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

 mourut

-0.53

ientí

-0.52

$^\

-0.49

RefreshLayout

-0.48

 Azevedo

-0.47

 malheureux

-0.46

Eure

-0.46

 superiori

-0.44

 pouvoit

-0.43

 Cientí

-0.43

POSITIVE LOGITS

Identyfik

0.70

expandindo

0.61

llten

0.59

ligg

0.57

 initComponents

0.56

llum

0.56

engager

0.56

scry

0.56

walk

0.54

descend

0.52

Activations Density 0.011%